- Email: [email protected]
Post-design system control for integrated space heating systems in residential buildings in cold regions
Building and Environment 94 (2015) 313e324
Contents lists available at ScienceDirect
Building and Environment journal homepage: www.elsevier.com/locate/buildenv
Post-design system control for integrated space heating systems in residential buildings in cold regions Xinming Li, Mustafa Gül*, Mohamed Al-Hussein Department of Civil and Environmental Engineering, University of Alberta, 9105 116th Street NW, Edmonton, Alberta, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 March 2015 Received in revised form 2 August 2015 Accepted 16 August 2015 Available online 19 August 2015
Ground source heat pump is an effective and environmentally friendly system for space heating and cooling in residential buildings. For cold regions, where heating demand is greater than cooling demand, multi-source heat pumps integrating renewable energy sources are an alternative that ensures occupant comfort. This paper presents a research methodology for investigating the performance of multi-source heat pump systems in residential buildings in cold regions. One year of monitoring data is collected from a four-storey residential building located in Fort McMurray, Alberta, Canada. This data is analyzed to evaluate the performance of the multi-source heat pump system. Enhanced heating system control is also devised which shows potential benefits for system efficiency improvement and non-renewable energy source savings. A system control set point forecasting model is also proposed for the purpose of reducing operation costs. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Renewable energy Ground source heat pump Coefficient of performance Forecasting model Energy savings Post-design system control
1. Introduction Greenhouse gas emissions in Canada have increased rapidly in recent decades, having grown at a rate of 9 mega tons per year from 591 mega tons in 1990 to the peak point of 749 mega tons in 2007 [1]. There has been significant improvement in reducing CO2 emissions since 2007 and researchers continue investigating renewable energy sources to replace non-renewable energy sources for heating, transportation, and industry. This paper focuses on the use of renewable energy sources for residential heating purposes. The use of renewable resources brings environmental benefits through the reduction of greenhouse gas emissions. Due to the benefits of energy savings and low CO2 emissions, ground source heat pump (GSHP) system is shown to be effective in both commercial and residential buildings for indoor heating, cooling and domestic hot water supply. Akpan et al. [2] showed that the geothermal heat pump is an effective thermal system for cold regions, both economically and environmentally (reducing greenhouse gases by 28% compared with the conventional system), compared with photovoltaics or other thermal collectors. The
* Corresponding author. E-mail address: [email protected] (M. Gül). http://dx.doi.org/10.1016/j.buildenv.2015.08.015 0360-1323/© 2015 Elsevier Ltd. All rights reserved.
ground-water heat pump system is also found to offer savings in electrical energy compared with both air-conditioner for cooling and coal-fired boiler for heating [3]. Coefficient of performance (COP) is an effective metric to evaluate the system performance of heat pump systems. The COP for heating is defined as the ratio of energy production to the power consumption of corresponding equipment, whereas the COP for cooling is defined as the ratio of thermal energy removed to the power consumption of corresponding equipment. Benli and Durmus [4] monitored a ground-source heat pump for a year and half in Turkey, having the COP of the GSHP in the range from 2.3 to 3.8 and the system COP in the range from 2 to 3.5. As other studies have shown, GSHP operation hours influence the COP of the heat pump as well. The COP of heat pumps decreases after a long period of operation in cold areas [5]. Involving more compressors may also reduce the COP for the heat pumps and for the entire system. The system performance can be analyzed in daily, monthly, and seasonal increments. A representative factor is Seasonal Performance Factor (SPF), which evaluates the system performance by heating season or cooling season [6,7]. The GSHP has a higher COP/ SPF when the demands for heating and cooling are balanced than when the demands are unbalanced, i.e., the GSHP is sensitive to the weather and potential risks exist when the GSHP is the only dependable system in regions with unbalanced heating and cooling demands [8]. Li et al. [9] explained and compared the geothermal
314
X. Li et al. / Building and Environment 94 (2015) 313e324
field energy distribution of operating the GSHP with both balanced and imbalanced thermal demands, highlighting the significance of using an integrated system. Other researchers have focused on achieving a steady temperature in an underground geothermal field utilized as one of the energy sources. Von Cube and Steimle [10] discussed the advantages of combining a geothermal heating system with other heat sources. As they pointed out, pre-heating the thermal source for the heat pump can lead to an increase of system COP. Li et al. [11] substituted the use of boilers by using a combination of groundwater flow and GSHP for both heating and cooling in twelve greenhouses in a cold region, Akabira, Japan. Their study suggested that adding groundwater flow in the field could reduce the fluctuation of field temperature caused by imbalanced heating and cooling demands, which could further achieve higher system efficiency. The system they described contributed to CO2 emission reductions of 20% and 22% compared with air-source heat pump and kerosene system, respectively. Combining a solar energy system with GSHP is another alternative, which has drawn interest among researchers. The idea of using solar energy to assist GSHP was proposed as a desirable system by Metz in 1982. He compared the results from both experimental and computing (through TRNSYS simulation software) approaches. However, the results were found to vary based on different soils, climates, and experimental designs [12]. The depth of the ground exchanger also influences the results. Wang et al. [13] simulated a solar-ground coupled heat pump with better COP for a long-term cooling and heating period due to geothermal field heat recovery and storage during the transitional seasons and summer. They showed that the system performs better with the vertical geothermal exchanger at a depth of more than 50 m than other depth of vertical exchanger. It was also shown that the complexity and properties of soil affect the heat transfer, where water content in the soil facilitates better thermal conduction for the use of the GSHP. Bi et al. [14] measured the performance of a solar source heat pump, a GSHP, as well as of a solar-assisted ground source heat pump (SGSHP) based on an experimental system under observation for five months. They evaluated the feasibility of using SGSHP and provided valuable experimental results. Bakirci et al. [15] reviewed research results published prior to 2011 and noted that the focus in previous research had been on either solar-assisted heat pump or GSHP individually, with only limited research focusing on systems that integrate both sources. More recently the solar-assisted GSHP system is of increasing interest among researchers. Research results have proven that the utilization SGSHP is advantageous both economically and environmentally compared with conventional energy sources [16,17]. Numerous simulations have been conducted by researchers to analyze SGSHP for the purpose either of revising the existing system or proposing new system designs prior to practical implementation [18e20]. Si et al. [12] designed two solar-GSHP systems for space heating, cooling, and hot water heating and introduced two models. In the first model, the solar thermal collectors not only reheat the circulating fluid from the geothermal field to the heat pump during winter, but also deliver the surplus energy in order to recover the field in summer, while the second model only stores solar energy to the water tank in daytime and recharges the field at night. The results of a simulation carried out as part of their study showed that the first model had better capacity for restoring field temperature than the second model. The system discussed in this paper resembles the first model from the Si et al. study. The challenge of using the GSHP for only heating but not for cooling in a cold region is even bigger. Kjellson et al. [13] simulated different combinations of solar collectors and geothermal exchangers to reduce the electricity demand in the system. They identified two key benefits: (1) that the solar energy system produces heat in summer
and recharges the ground in winter; and (2) that, with the support of the solar energy, the thermal interference between boreholes can be reduced so that the ground temperature will be reduced. Other researchers have also analyzed SGSHP systems built on experimental data based on a system model in a lab or small building [21e26]. Experimental analysis has also validated the notion of using solar as a thermal collector during the daytime and storing energy to the ground for the recovery support of the geothermal field [21]. These studies have revealed that the efficiency of ground heat exchangers can be enhanced from a COP of 2.37e2.72. Ozgener and Hepbasli [22,23] monitored a solar-assisted vertical GSHP heating system for a greenhouse in Turkey and analyzed the performance of this system in terms of energetic and exergetic aspects. The average COP of GSHP achieved 2.84 while the average exergy efficiency was 68.11%, and the COP of entire system was found to be 5e20% lower than the COP of the heat pump itself. Dikici and Akbulut [24] experimented with a solar-assisted heat pump system in a test room and obtained a higher COP of 3.08 during the season with heating demand. Another SGSHP was tested in Turkey with a heat pump COP in the range of 3.0e3.4 [15]. Yu et al. [25] combined solar energy with GSHP with the aim of achieving equilibrium in the geothermal field energy for a villa in the cold region of Beijing, China. Their results showed that the SGSHP not only satisfies heating and cooling demand but also keeps the geothermal field temperature balanced for long-term usage and facilitates a higher COP of the system. After a long-term investment and acceptance of integrating GSHP with other systems, some researchers have also sought the development of improved control strategies for the system being operated. Yavuzturk and Spitler [26] proposed several control strategies for system operation by simulating the combination of using a supplemental heat rejecter with a cooling-dominated GSHP for a small office building. Other researchers have conducted experiments on the system and enhanced the system control based on the experimental results. Mokhtar et al. [27] utilized a type of artificial neural network system to propose an intelligent building management system control in order to minimize the energy waste, optimize load control, and increase the system efficiency. Wang et al. [28] adjusted the system cooling control strategy based on experiment data collected from a specified short period of time. They recommended a system control strategy reference for the hybrid GSHP system in practice. Thus, existing experimental data provides useful information for future reference. Without the enhanced system control algorithm, the considerable benefits of using renewable source energy as the heat source may not be maximally achieved. The limitation of these research results is that there is always variation from one system to the other. The system control strategy devised for one system may not be suitable for other systems. However, in this paper, we propose a system control improvement strategy based on live sensorbased system monitoring and live system performance analysis. The research results not only can serve as a reference for future study and implementation, but the methodology can also be implemented in other systems for the purpose of system control. Furthermore, this research is focused on integrated GSHP system control, thereby addressing a gap in the existing body of research. 2. Research objective and methodology The main objective of the research presented in this paper is to propose enhanced system control and improve the performance of an integrated heating system in a residential building under occupancy based on comprehensive, live, sensor-based experimental monitoring. The framework can be applicable to other buildings as well, with modifications in the control algorithm and set points.
X. Li et al. / Building and Environment 94 (2015) 313e324
The integrated heating system monitored in this study combines the energy from solar thermal collectors and a drain water heat recovery (DWHR) system to support the GSHP system, as the multisource heat pump (MSHP), in order to improve the COP of the system. The purpose of this design is twofold: (1) in winter, to collect the energy from solar thermal collectors, DWHR and geothermal field in order to assist the heat pumps for heating; and (2) in summer, to store the surplus solar energy and energy recycled from the DWHR in summer in order to recover heat loss in the geothermal field. A research framework of system analysis (as shown in Fig. 1) is proposed to improve system control based on live monitoring data. From the live sensor-based monitoring, the system performance can be recorded and analyzed, and will be further used for the comparison between renewable and non-renewable energy sources, developing a forecasting model and proposing enhanced system control. The framework input requires the data collections of temperature, flow rate, pressure, pump status, and electricity usage of heating equipment, as well as local outdoor temperature data and the unit prices of natural gas and electricity. Temperature, flow rate, and pressure in the circulating fluid in system pipes are meaured using temperature sensor, flow rate meter, and pressure sensor, respectively. OneOff status sensors are installed on the main circulating pumps and CT power meters to capture the electricity consumption from circulating pumps and heat pumps. Based on the data collection and analysis, the energy production and efficiency of each system can be calculated using Equations (1) and (2).
Q ¼ C r Rv t TSupply TReturn COP of MSHPðheatingÞ ¼
QHP
source
277:78 þ EMSHP EMSHP
where Q: Thermal Energy, kilojoules (kJ)
(1)
(2)
315
C: Specific Heat, kJ/kg C
r: Density of fluid, kg/m3
Rv: Volume flow rate of fluid, m3/hr t: Serving time or operation time, hr TSupply: Temperature of supply pipe, C TReturn: Temperature of return pipe, C QHP_source: Energy production from source side of heat pumps, GJ EMSHP: Multi-source heat pump electricity consumption, kWh The efficiency of each system can be identified using this method. And, if the system works with low efficiency, it requires further modification by either changing the system control algorithm or revising the original system design. Usually, revising the system design is challenging and has high risk of wasting the original capital cost or incurring additional cost. Thus, modification of the existing system control algorithm is more cost-effective, and can be achieved by raising the corresponding system corrective measures based on the system performance analysis. After comparing the performance of the existing system to the estimated performance of the system with proposed corrective measures, an enhanced system control algorithm can be achieved. The expected outcomes of this research can be associated with both environmental and economic benefits. With regard to the environmental aspect, researchers are able to learn about CO2 emission reduction resulting from renewable sources superseding non-renewable sources in the monitored heating system. Economically, because the total operational costs, including operation fees and energy costs, varies due to the fluctuations in energy source unit prices, a forecasting model aimed at minimizing operational costs is also established. The outcome of this forecasting model also contributes to the overall system control algorithm. There are also constraints petaining to the research results, especially the location and weather conditions of the monitored building, system design, and occupant behavior. These constraints affect the analysis of the findings. Consequently, the experimental design provides the researchers
Fig. 1. Research methodology for system control analysis based on live integrated heating system monitoring.
316
X. Li et al. / Building and Environment 94 (2015) 313e324
to access and analyze data in order to ascertain the system performance parameters and to analyze the potential improvement of the existing heating system design. As output, an approach to enhanced heating system control is proposed and shown to be more environmentally-friendly than the existing system control, also resulting in cost savings. The same experimental and analysis methodology can also be applied to other projects for system performance analysis. 3. Methodology implementation The methodology is implemented in a four-storey, 1295 m2 (13,937 ft2) wood-framed residential building in Fort McMurray, Alberta, Canada. Fort McMurray is located in a cold-climate region that experiences an average annual temperature of 1.3 C and average heating degree days (HDD) of 6094 [29,30]. As depicted in, the building mainly relies on the MSHP system for space heating of 70 apartments units (combination of both one-bedroom and twobedroom apartments) through domestic heating radiators. The geothermal system is only used for heating, and 10 solar panels as well as 11 DWHR systems are integrated in order to support the geothermal system for underground field energy recovery. Electricity is utilized to operate all these renewable energy source systems. Natural gas-fired boilers, as the secondary system, are the auxiliary heat supply in winter. The system schematic drawing, including sensor locations, is given in Fig. 2. The integrated heating system is controlled by a Building Management System (BMS). The solar energy component of the system operates when the temperature in the geothermal field return pipe is lower than the temperature at the solar panel. Freeze protection is designed to be activated in winter when the outdoor temperature falls below 2 C. In this case the freeze protection system runs for 20 min at a time. The DWHR pump runs based on the temperature of the drainage pipe (set point is 2 C). DWHR pump refers to the circulating pump on each drainage stack, while the injection pump refers to the upstream circulating pump serving all 11 stacks. The heating mode is “On” when the outdoor temperature is below 15 C. Heat pumps remains “On” until the outdoor temperature falls to 20 C, at which point the boilers turn “On”. The heat pump and boiler operations are controlled by
specific temperature set points of the hot water tank. The existing system control presents a number of opportunities for potential improvement. This research is focused on the system production performance and system efficiency analysis. Based on the collected data, the authors are able to propose an improved and efficient control system based on observation and analysis of the existing system. The system controls of the solar energy system, DWHR system, heat pumps, and boilers, highlighted in Fig. 2 are separately analyzed in greater detail, and strategies for improved system control are proposed later in this paper. To analyze the system operation performance under occupancy, Li et al. [9] proposed an experimental design to monitor the system based on three months of system performance data. The research presented in this paper is built on the study described in Li et al., which presented the implementation of the proposed methodology using one year of data (March 15, 2013 to March 14, 2014) collected from 32 sensors. The data analyzed is drawn from 12 temperature sensors, 10 OneOff status sensors, six CT power meters, and four flow rates on the circulating pipes. The temperature sensorsdThermocouples (TX)dare installed on the supply and return pipes of each system and the flow rates (FlowX) represent Ultrasonic flow meter measurements taken on March 15, 2013. In addition, six electricity sensors (CTX-X) are installed on the field circulating pumps and heat pumps, which are the main power consuming devices, in order to monitor their power usage. The status of whether the circulating pumps and heat pumps (STX-X) are “On” or “Off” is also recorded. These sensors are collecting data at intervals ranging from 10 s to 1 min. Table 1 presents the detailed energy production and efficiency calculation for each system using the data from corresponding sensors.
3.1. Annual energy production, system efficiency and operation cost from renewable sources (solar, DWHR, geothermal field) Based on the data collected from March 15, 2013 to March 14, 2014, annual heating system performance can be evaluated monthly. Hourly intervals of data are selected for the calculations in this paper. The fluctuations in monthly heating performance are presented graphically for each system.
Fig. 2. Integrated heating system schematic with sensor locations [9].
X. Li et al. / Building and Environment 94 (2015) 313e324
317
Table 1 Circulating fluid and energy production calculations (legends refer to Fig. 2) [9]. System
Fluid
Energy production (GJ)
Solar energy system
13.6% Methanol
DWHR
13.6% Methanol
MSHP
15% Glycol
Boilers
15% Glycol
QSolar ¼ 4.00 kJ/kg C 971.68 Kg/M3 Flow3 t (T5 T4) 106 COPSolar ¼ QSolar/(ST1 ST1_pump_power) QDWHR ¼ 4.00 kJ/kg C 971.68 Kg/M3 Flow6 t (T12 T13) 106 COPDWHR ¼ QDWHR/ST10 ST10_pump_power) QMSHP ¼ 3.90 kJ/kg C 1016.98 Kg/M3 Flow5 t (T3 T1) 106 COPMSHP ¼ QMSHP/CT1e4 QBoiler1 ¼ 3.90 kJ/kg C 1016.98 Kg/M3 Flow4_1 t (T14 T1) 106 QBoiler2 ¼ 3.90 kJ/kg C 1016.98 Kg/M3 Flow4_2 t (T15 T1) 106 COPBoiler1 ¼ QBoiler1/(ST6 ST6_pump_power) COPBoiler2 ¼ QBoiler2/(ST7 ST7_pump_power)
Note: pump power draw rates are obtained from the pump specifications.
3.1.1. Solar energy system Ten 2.87 m2 solar panels are installed at 45-degree obliquity on the south roof of the residential building. 13.6% methanol is used as the energy transmission fluid in the system. The solar energy system production is calculated for each month, as presented in Fig. 3. The solar energy system is found to produce 52 GJ, but to consume 6 GJ (1664 kWh) of electricity at a cost of $145 (all dollar amounts in this paper are in Canadian Dollars) for the time period under investigation. Energy waste occurs in winter time during the annual performance as estimated in the calculation. This energy waste is due to the system freeze protection, since the solar heat transfer station (including the primary and secondary circulating pumps) is required to run every 20 min when the outdoor temperature is below 2 C. As a result, 10%e20% or more of energy production from the solar energy system is wasted through solar system upper loop, which is exposed to outdoor conditions in winter nights, and the loop is found to waste energy instead of generating energy in December, 2013 and January, 2014. The power usage from the secondary circulating pump, which runs continuously, is also included in the COP calculation. The monthly average COP of the solar energy system is also found to decrease from 14 to below 0 from summer to winter. Because of the more abundant solar energy in summer, and the energy waste in winter as discussed previously, the COP of the solar energy system in summer is higher than the COP in winter. The average monthly production cost of using conventional electricity to drive the circulating pumps is found to be $3.6/GJ, with $2.1/GJ in summer and $8.9/GJ in winter based on the monthly electric price varying from $0.07/kWh to $0.11/kWh in the course of the year of data collection [31]. It should be noted here that the energy delivery fee or other fees are not included in energy unit prices discussed in this paper. In order to improve the COP of the solar energy system, the authors proposed increasing the concentration of the anti-freeze fluid (Tyfocor L) to 50%, where the freeze protection is not
required until the outdoor temperature drops to 40 C. In this case, the need for circulating freeze protection can be reduced or eliminated. Based on the technical information of Tyfocor L, the specific heat capacity has linear relation to the fluid temperature while the density changes non-linearly versus the fluid temperature [32]. The changes of specific heat capacity and density along with the temperature of the fluid have been calculated. On the basis of the recalculation, the system obtains higher COP in winter but lower energy production in summer. The electricity consumption can be reduced by 13.2%, corresponding to an operational cost decrease from $145 to $127 for the monitored year; however, the energy production will decrease from 52 GJ to 48 GJ, because, the higher the concentration of the anti-freeze fluid, the lower specific heat it will have, thus the lower energy it will gain. Thus, based on this analysis and in consideration of the high costs of materials and labor in Fort McMurray, it is not worth replacing the anti-freeze fluid as it has a long payback period (approximately 30e50 years). However, in the future heating system design, designers should emphasize selecting the optimal concentration of antifreeze fluid. 3.1.2. DWHR system Eleven DWHR pipe stacks are installed to recycle the wasted energy in hot water from the 70 apartment units. One of these stacks is monitored for the purpose of this study, and it is assumed that the production of each stack is equivalent. The monthly energy production and monthly average COP of the DWHR are presented in Fig. 4. For 11 stacks in total, the DWHR system can produce 340 GJ by consuming 51 GJ of electricity (14 MWh) with monthly production varying from 15.6 GJ to 35.2 GJ, as illustrated in Fig. 4. The energy production depends on both the efficiency of the DWHR system and also the hot water consumption by occupants. The monthly average COP of the DWHR ranges from 3.8 to 8.7, with a production cost of $3.6/GJ. Evidently, the DWHR system works with
Fig. 3. Solar energy system production and monthly average COP.
318
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 4. DWHR system production and monthly average COP.
higher efficiency in winter than in summer due to the fact that occupants consume more hot water in winter than in summer. However, the DWHR units run regardless of the volume of hot water coming to the drainage pipe. Since the injection pump is always “On”, electricity consumption is wasted if the recycled energy does not exceed the electricity usage of the injection pump and circulating pump. In order to improve the COP of the DWHR system, the control point for running the circulating pumps should be adjusted for each drain water stack. Based on the temperature of the drainage pipe, diverse set points are proposed for different seasons. According to the analysis of energy production versus the electricity usage of the circulating pumps, set points are 6 C from March to May, 10 C from June to October and 4 C from November to February. It is anticipated that this adjustment will prevent unnecessary power consumption and improves system efficiency. After recalculating the expected system performance with the proposed control set points, ideal electricity savings of 10% can be achieved, reducing the annual operation cost from $1202 to $1088 with 12.7 GJ in corresponding energy reductions. The DWHR system production cost can be reduced from $3.6/GJ to $3.2/GJ and the COP can be improved by proportions ranging up to 39.9%, as plotted in Fig. 5. Therefore, recommendations are devised to adjust the set points of the drain pipes, to vary from 4 to 10 C, according to the given month/season. However, it should be noted that this recommendation is based on the data collection from one DWHR stack. The more stacks that can be monitored the more accurately the set points can be estimated. Another limitation of the present study has to do with the fact that the set points are fixed numbers. Ideally, the set points vary based on occupant water usage demand and geothermal heating system working conditions. The next step of this project is to generate a method by which to obtain the set points automatically based on timely temperature data.
3.1.3. Geothermal field Two field circulating pumps drive the geothermal field to generate heat in the winter and increase the temperature on the source side of four water source heat pumps. 80 underground geothermal heat exchangers (vertical U-tube) are buried at a depth of 91 m underground. The energy production from the field differs seasonally, as presented in Fig. 6. It provides heat in the winter when heat pumps are in operation and recovers the heat in the summer and occasionally in the winter when heat pumps are not in operation. In total, the geothermal field is found to have provided 236 GJ of energy in winter and to have recovered from the solar energy system and DWHR system 182 GJ of energy over the course of the year of data collection. Overall, it produces 54 GJ more energy than the amount recovered; thus, corrective measures should be taken to recover more energy to the field by either using less of the geothermal field as a heating source or providing more energy sources to support the geothermal field recovery. Of the 182 GJ of recovered energy in the geothermal field, the solar energy system accounts for 16% and the remaining 84% is from the DWHR system (A further comparative analysis between these two energy sources is presented in the next section.). The total operational cost of the geothermal field is $2,812, corresponding to 118 GJ (32,783 kWh) of electricity consumption. These are valuable data for future similar heating system design.
3.2. Operation cost-effectiveness analysis between renewable energy sources The renewable energy sources used in this project are solar energy, DWHR, and geothermal energy. Renewable sources are environmentally friendly since they incur comparably less greenhouse gas emissions in producing energy. However, electricity consumption is still necessary for system operation, thus the operational costs of each source must be compared. Based on the analysis of system production costs from Section
Fig. 5. Enhanced COP of DWHR system.
X. Li et al. / Building and Environment 94 (2015) 313e324
319
Fig. 6. Geo-field energy production and heat recovery.
3.1, the solar energy system is more cost-effective than the DWHR system in summer while the opposite is true in winter. After implementing the proposed post-design control system, the production cost of the DWHR is expected to be lower than that of the solar energy system. However, this only includes the operational cost. If the initial cost is also taken into consideration, the DWHR is more cost-effective based on life cycle analysis since the initial cost of the solar energy system is much higher than the DWHR system. On the other hand, there are limitations to the DWHR system, as its production rate is sensitive to occupant water usage and the system control algorithm. The production of DWHR is constrained by the water usage of occupants, whereas the solar energy system is constrained by the surface area of solar panels as well as weather conditions. For the renewable source energy systems, beside the capital costs, the only input of system operation is electricity consumption. During the year of data collection, $4159 is determined to have been expended to power the systems, corresponding to 48 MWh of electricity consumption. Of this, 3.5% is for the solar energy system, 29.1% is for the DWHR, while 67.4% is for powering the geothermal field circulating pumps. Consequently, as shown in Fig. 7, 446 GJ of energy is produced: 12.1% by the solar energy system, 76.2% by the DWHR system, and 11.7% by the geothermal field. Total energy output comprises purchased electricity as well as 271 GJ of thermal energy produced by the renewable sources. 15.7 tons of CO2 emissions have been avoided owing to the current renewable source system design, compared with using a conventional natural gas heating system. This estimation is based on the natural gas usage standards and statistics specified by Natural Resources Canada [33] and the CO2 emissions coefficients given by the U.S. Energy Information Administration [34]. This amount of CO2 is approximately equivalent to the CO2 emissions produced by 60,704 km of automobile travel.
3.3. Energy production comparison among resources (MSHPs, boilers) The energy produced by all the renewable sources is transferred to heat pumps, which increases the temperature on the source side of the heat pumps. The energy produced by the MSHPs and boilers is conveyed to two hot water tanks for heat preservation. The hot water tanks are connected with the final heating supply and return loop in order to deliver the heated water to the heating radiator in the apartments. 3.3.1. Multi-source heat pump production The MSHPs contain the water-to-water heat pumps, with the solar energy, DWHR system energy, and geothermal energy as sources. Sensors are located on the load side of the heat pumps, collecting the temperature and flow rate data for heat pump energy production, i.e., the energy transmitted from the MSHPs to the hot water tanks. Assuming the flow rate ranges from 148 to 170 standard liters per minute (SLPM), the energy production from heat pumps can be estimated as specified in Fig. 8. It varies monthly depending on the heating demand and the control system. The MSHPs are found to have produced the most energy in April and November within the monitored year, when the daily average outdoor temperature is higher than 20 C and the heating supply mainly relies on the heat pumps. In the remaining months of the cold season, natural gas-burning boilers afford sufficient energy to satisfy the heating demands of the occupants. In total, the system is found to have produced 690 GJ of energy while consuming 350 GJ (97 MWh) of electricity at a cost of $7089. The MSHP system operates with a seasonal-average COP (commonly known as Seasonal Performance Factor (SPF) [35]) for heating of 2.24. The monthly average COP of this system fluctuates, as illustrated by the curve in Fig. 8. Evidently, they perform with higher monthly
Fig. 7. Electricity input and corresponding energy output from each renewable source system.
320
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 8. MSHP energy production and monthly average COP.
average COP (approximately 2e2.5) in winter, while they perform with lower monthly average COP (below 2) in summer. Two factors explain this result of low efficiency in summer. First, because the system is only used for heating, not for cooling, the heating demand is low in summer. However, the heat pumps are still in heating mode, although sometimes only one heat pump is operated for heating. This results in a waste of electricity, which further decreases the COP of the heat pumps. Second, typically in summer, the geothermal field is recovering and cannot provide sufficient heat for summer nights when the outdoor temperature falls below 20 C. Thus, the COP of MSHPs can be improved by turning heat pumps into idle mode when there is less heating demand in summer in order to reduce power usage and increase COP. Moreover, their COPs can also be improved by adding more energy on the source side of the system or decreasing the heating demand on the load side. 3.3.2. Natural gas-burning boilers During the year of monitoring, the boilers play a vital role for heating supply in winter, which accounts for 58% of the annual heating load, while the remaining 42% of the heating load is provided by the MSHPs. The unit price of natural gas differs monthly from $1.7/GJ to $9.0/GJ based on Direct Energy historic data for customers in the relevant region of Alberta, Canada [36]. The natural gas and electricity unit price comparison is given in Table 2 (energy delivery fee is not included). Fig. 9 presents the input and output during the monitored period, showing MSHPs to have consumed 350 GJ (97 MWh) of electricity, costing $7,089, and produced 690 GJ of energy, while the boilers have consumed 3037 m3 of natural gas, costing $4,650, and produced 966 GJ of energy. Thus, the production cost of MSHPs is higher than the production cost of boilers. Due to the fact that the unit cost of natural gas is lower than electricity in Fort McMurray during the monitoring period, a production rate comparison between the boilers and MSHPs is also analyzed, as illustrated in Fig. 10. Overall, the production cost of the natural gas burned by the boilers is less than that of the MSHPs. The exception is March 2014, due to the rapid increase in the unit price of natural gas to $9/GJ in March, 2014. $9/GJ is observed to be an appropriate threshold, such that
natural gas prices above this warrant changing from boilers to MSHPs as a cost-effective solution. That is to say, if the unit price of natural gas is higher than $9/GJ and electricity remains approximately $0.1/kWh, the MSHP is worth using economically. However, the threshold varies monthly because of the fluctuation of unit price of these two sources. Overall, it is found that MSHPs can achieve a total production cost of $11.5/GJ to $13.2/GJ with the current electricity price vacillating between $0.07/kWh to $0.11/ kWh, whereas natural gas would reach a total production cost of $4.4/GJ with a boiler efficiency of 85%, based on the natural gas price range from $1.78/GJ to $9.02/GJ for Alberta delivery. It is therefore more cost-effective to use the boiler as the only heater for the entire building until the natural gas price increases beyond the threshold of approximately $10.0/GJ. The total operation cost from boilers and MSHPs is found to be approximately $12,000 during the year of monitoring. Due to the cost-effectiveness of the boiler, if the boiler is the only heating supply, the total operation cost would be estimated to be approximately $8,000, resulting in $4000 in savings (see Fig. 11). However, the use of solar and DWHR to assist the GSHP system makes it much more environmentally-friendly, as 15.7 tons of CO2 emissions are avoided owing to the renewable source system design. More economical system design can also be achieved if the COP of the heat pumps increases. Thus, based on the system monitoring design and data analysis methodology, building owners can encounter environmental benefits and/or heating system operation cost savings. The research results also lead to a better system control for future system operation. A post-design system control that will be further detailed and summarized in the next section has been proposed based on the current analysis. 3.4. Post-design system control Based on the analysis conducted with the monitoring data, it is determined that the performance of the heating system can be improved and the energy used by the heating system can be reduced. In the interest of energy saving and cost efficiency, a postdesign system control is proposed as represented in Fig. 12. As discussed in Section 3.1.1, due to the high investment cost of
Table 2 Unit price of natural gas and electricity [31,36]. Source
Time 2013
Natural gas Electricity
($/GJ) ($/kWh) ($/GJ)
2014
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
2.93 0.07 19.44
3.76 0.08 22.22
4.19 0.07 19.44
4.59 0.07 19.44
1.85 0.11 30.56
1.78 0.11 30.56
2.32 0.11 30.56
2.43 0.08 22.22
3.72 0.08 22.22
3.30 0.08 22.22
4.09 0.09 25.00
4.57 0.07 19.44
9.02 0.07 19.44
Although the cost of electricity is not common to be presented in $/GJ, it is given in this table for comparison purpose.
X. Li et al. / Building and Environment 94 (2015) 313e324
321
Fig. 9. Energy production comparison between MSHPs and boilers.
Fig. 10. Production rate comparison between MSHPs and boilers.
Fig. 11. Operation cost of the MSHP and boiler.
changing the solar energy system design, no adjustment is proposed for the current stage of the solar energy system (Fig. 12a). And according to the analysis in Section 3.1.2, changing the control system set points of the drain pipes (current range is from 4 C to 10 C) for different months and seasons is recommended to improve the performance of the DWHR system (Fig. 12b). In terms of the MSHPs, more problems are identified and corresponding suggestions are devised. To obtain a higher COP of heat pumps, three solutions are listed below. The post-design system control is summarized in Fig. 12c, with proposed changes denoted by the dark-shaded objects in the figure.
higher temperature. In this case, the geothermal field will recover more energy within a year, making the energy recovery roughly equal to the energy production. Further analysis regarding this suggestion will be discussed in Section 3.5. 2) Reduce the leaving load temperature (the outlet temperature on the load side of the heat pump). The energy demand on the load side of heat pumps also affects their working performance. The COP of heat pumps can be improved by decreasing the set point of the hot water tank, particularly in summer when less heating load is needed.
1) Adjust the set point temperature for switching from MSHPs to boilers.
3) Group heat pumps into pairs and set boiler priority.
Since the overall energy production of the geothermal field is higher than energy recovery and the production cost of natural gas is lower than that of electricity for the monitored period, the MSHP can be utilized with less frequency. This can be approached by increasing the switching point of heat pumps and boilers to a
The heat pump consumes a small amount of power when it is in idle mode. In most cases, only one or two heat pumps are required to operate in spring and autumn. The remaining two heat pumps waste power when they are in idle mode for long periods. To avoid this unnecessary electricity consumption, an effective method can
322
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 12. Post-design system control.
be to group the four heat pumps into two pairs, which, meanwhile, also offers benefits in terms of reduced machine maintenance needs.
3.5. Forecasting model by optimizing switching point of heating systems A forecasting model is under development to obtain the optimal switching point between heat pumps and boilers, which can be updated based on current energy source unit price information. In this model, the inputs are annual heating degree days/hours, heating load supply, and electricity and natural gas unit prices. Heating degree days/heating degree hours (HDD/HDH) represent the heating demand for a building. This can be calculated as the sum of the differences between the daily/hourly average
temperatures and the base temperature as illustrated in Equation (3) [37]. This base temperature represents the indoor temperature controlled by occupants. Assuming the base temperature for this residential building averages 18 C (recommended by Government of Canada) in the monitored year, the HDD versus daily heating supply by the heating system is plotted in Fig. 13, with a high correlation coefficient of 88.7%. The monitored year is observed to have been unusually cold, particularly in December, 2013, and to have included 54 days when the outdoor temperature fell below 20 C. The total heating load is 1634 GJ, while the HDD is 6,438, which is higher than the Fort McMurray average HDD (6094) for the past 25 years [38]. The outdoor temperature data from the monitored sensor is verified using Canadian Government climate statistics, with a correlation coefficient of 99.5% between the data from system monitoring and the government climate statistics [38].
Fig. 13. HDD and daily heating supply.
X. Li et al. / Building and Environment 94 (2015) 313e324
Thus the outdoor temperatures collected are deemed to be reliable and meaningful for the geographical location of the monitored residential building.
HDD ¼
i¼N X ðTb Ti Þ
(3)
i¼1
where. Tb: Base temperature ( C) Ti: Daily average temperature ( C) N: Number of days
323
However, the unit price of energy source varies monthly. The next step in developing the forecasting model involves incorporating fluctuations in the energy unit price over time. The resident can opt for a yearly fixed energy price contract with an energy company. In addition, the base temperature may differ among apartment units. Thus, a more extensive model can be established based on occupant information, apartment vacancy status, and forecast weather conditions. The research team is seeking to ultimately achieve an intelligent forecasting model capable of adjusting the set points based on energy unit price fluctuations over time. 4. Limitations of the experimental data
HDH is chosen as the input for forecasting since the correlation coefficient between HDH and hourly heating supply is 0.947. A linear and a non-linear relationship are assumed in order to obtain the formulary relationship between hourly heating energy supply and outdoor temperature. Setting average daily outdoor temperature as the variable, the heating energy supply can be calculated, generated formulas as given in Equation (4) and Equation (5). The R square values of these deduced equations are 0.8514 and 0.9548, respectively (Fig. 14). The operation cost can also be ascertained, counting energy unit price in the calculation. Whether or not the outdoor temperature exceeds the system switching point determines which energy source to use.
Qsupply ¼ 0:2738 Tout þ 4:7259ðTout < 17 CÞ
(4)
2 0:2433 Tout þ 2:9162ðTout < 20 CÞ Qsupply ¼ 0:0063 Tout
(5) where Qsupply: Daily heating energy supply (GJ) Tout: Average daily outdoor temperature ( C) The forecasting model is able to recalculate the total system operation cost under various system switching temperatures, as illustrated in Fig. 15. Scenarios with set points of 19 C, 18 C, 17 C, 16 C, 15 C are implemented in this study. After recalculating the operation cost under these scenarios, 16 C is selected as the optimal set point, as it incurs the least annual operation cost under both linear and non-linear estimations. Economically, with the new set point, the proposed designed system offers savings in the range of 10.9%e12.1% in operational costs by using linear and non-linear regression, respectively. This outcome is analogous to the conclusion obtained by Nguyen et al. that the comparably low rate of natural gas results in GSHP systems being less economical to install for heatingdominant buildings in North America [8]. The model is still under construction as of the time of submission of this paper.
The sensor-based monitoring can be improved if it is designed before the construction of the building. The sensors installed for the present experiment are surface sensors connected to the pipes. If permitted by building management, inserted sensors would be a better option yielding more accurate data, which would further increase the reliability of the results. However, due to limited access to the heating system in the case study and the budget of this project, the current experimental design is the best design achievable under the given circumstances. 5. Conclusion and future study This project targets energy saving and renewable resource applications, particularly by means of a GSHP. A GSHP supported with energy from solar and DWHR sources is proposed to satisfy energy demand in a residential facility through an innovative and energy efficient approach. This multi-source space heating system for residential building is monitored for one year according to sensorbased experimental design. Based on data collection, energy production calculation and COP analysis for each part of the system is carried out. Results indicate that the usage of the geothermal heating system only for heating, not for cooling, is a challenge. A better system control is recommended with comprehensive live monitored data collection and analysis. The authors have also analyzed the system performance on the aspects of environment and economy. The results show that the use of an integrated heating system leads to the environmental benefit of an annual reduction of CO2 emissions by 15 tons compared with a natural gas system. In terms of economics, an enhanced postdesign system control is also investigated to increase the existing system efficiency, as well as to minimize the operational cost of the heating system. The enhanced post-design system control recommendations can be achieved by programming system control algorithms. A heating system operation cost forecasting model is also established for estimating the switching temperature criteria of using heat pumps and boilers, which should ultimately result in system operation savings in the range of 10.9%e12.1%, as expected. In this model, the outdoor temperature and the unit price of
Fig. 14. Fitting data with linear and non-linear regression.
324
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 15. Estimated total operation cost savings for systems.
electricity and natural gas are the inputs. This forecasting model will be further developed, including integration of the fluctuation of energy unit price and full automation and adjustment of setting switching point. Acknowledgments This research work has been developed within the scope of a project undertaken by the University of Alberta, entitled, “Integrated Construction and Energy Management System for Affordable Housing in Fort McMurray”. The authors would like to thank the Natural Sciences and Engineering Research Council of Canada (NSERC), CRDPJ 411101-10, as well as project sponsors, Wood Buffalo Housing and Development Corporation, Integrated Management and Realty Ltd., Cormode & Dickson Construction Ltd., Hydraft Development Services Inc., TLJ Engineering Consultants, and BCT Structures, for their support and assistance. References [1] Government of Canada, National greenhouse gas emissions, Environ. Can. (2014). https://www.ec.gc.ca/indicateurs-indicators/default.asp?lang¼en&n¼ FBF8455E-1 (accessed August, 2014). [2] I.E. Akpan, M. Sasaki, N. Endoh, Optimization of domestic-size renewable energy system designs suitable for cold climate regions, Jpn. Soc. Mech. Eng. (JSME) Int. J. Ser. B 49 (4) (2006) 1241e1252. [3] N. Zhu, P. Hu, W. Wang, J. Yu, F. Lei, Performance analysis of ground watersource heat pump system with improved control strategies for building retrofit, Renew. Energy 80 (2015) 324e330. [4] H. Benli, A. Durmus¸, Evaluation of ground-source heat pump combined latent heat storage system performance in greenhouse heating, Energy Build. 41 (2) (2009) 220e228. [5] Y. Liu, Y.X. Wang, Y.Z. Zhang, Experimental study on a ground source heat pump system in cold area, Appl. Mech. Mater. 71e78 (2011) 2566e2571. Montero, J.F. Urchueguía, Analysis of the energy n, A. [6] C. Montagud, J.M. Corbera performance of a ground source heat pump system after five years of operation, Energy Build. 43 (12) (2011) 3618e3626. [7] C.J. Wood, H. Liu, S.B. Riffat, An investigation of the heat pump performance and ground temperature of a piled foundation heat exchanger system for a residential building, Energy 35 (12) (2010) 4932e4940. [8] H.V. Nguyen, Y.L.E. Law, M. Alavy, P.R. Walsh, W.H. Leong, S.B. Dworkin, An analysis of the factors affecting hybrid ground-source heat pump installation potential in North America, Appl. Energy 125 (2014) 28e38. [9] X. Li, M. Gül, T. Sharmin, I. Nikolaidis, M. Al-Hussein, A framework to monitor the integrated multi-source space heating systems to improve the design of the control system, Energy Build. 72 (2014) 398e410. [10] H.L. Von Cube, F. Steimle, Heat Pump Technology, Butter Worths, London, 1981. [11] H. Li, K. Nagano, Y. Lai, K. Shibata, H. Fujii, Evaluating the performance of a large borehole ground source heat pump for greenhouses in northern Japan, Energy 63 (2013) 387e399. [12] P.D. Metz, The use of ground-coupled tanks in solar-assisted heat-pump, J. Sol. Energy Eng. 104 (4) (1982) 366e372. [13] F. Wang, Z. Liu, Z. Li, Y. Liu, Z. Wang, M. Zheng, Performance analysis of seasonal soil heat storage air conditioning system in solar ground coupled heat pump, in: Proceedings, International Conference on Electronic & Mechanical Engineering and Information Technology, IEEE, Harbin City, China, Aug. 12e14, 2011, pp. 1599e1602. [14] Y. Bi, T. Guo, L. Zhang, L. Chen, Solar and ground source heat-pump system, Appl. Energy 78 (2004) 231e245. [15] K. Bakirci, O. Ozyurt, K. Comakli, O. Comakli, Energy analysis of a solar-ground source heat pump system with vertical closed-loop for heating applications,
Energy 36 (5) (2011) 3224e3232. [16] J.B. Zhao, Comprehensive utilization of solar assisted heat pump and ground source heat pump in severe cold region, Appl. Mech. Mater. 448e453 (2013) 2790e2793. [17] C. Xi, L. Lin, Y. Hongxing, Long term operation of a solar assisted ground coupled heat pump system for space heating and domestic hot water, Energy Build. 43 (8) (2011) 1835e1844. [18] F. Busato, R. Lazzarin, M. Noro, Ground or solar source for space heating: Which is better? Energetic assessment based on a case history, Energy Build. 102 (2015) 347e356. [19] Q. Si, M. Okumiya, X. Zhang, Performance evaluation and optimization of a novel solar-ground source heat pump system, Energy Build. 70 (2014) 237e245. €m, B. Perers, Optimization of systems with the com[20] E. Kjellsson, G. Hellstro bination of ground-source heat pump and solar collectors in dwellings, Energy 35 (6) (2010) 2667e2673. [21] W. Yang, L. Sun, Y. Chen, Experimental investigations of the performance of a solar-ground source heat pump system operated in heating modes, Energy Build. 89 (2015) 97e111. [22] O. Ozgener, A. Hepbasli, Performance analysis of a solar-assisted groundsource heat pump system for greenhouse heating: An experimental study, Build. Environ. 40 (2005) 1040e1050. [23] O. Ozgener, A. Hepbasli, A parametrical study on the energetic and exergetic assessment of a solar-assisted vertical ground-source heat pump system used for heating a greenhouse, Build. Environ. 42 (2005) 11e24. [24] A. Dikici, A. Akbulut, Performance characteristics and energy-exergy analysis of solar-assisted heat pump system, Build. Environ. 43 (11) (2008) 1961e1972. [25] T. Yu, P. Liu, G.M. Chu, Y.F. Li, Seasonal underground storage of SGSHP system design based on equilibrium geothermal temperature, Adv. Mater. Res. 512e515 (2012) 864e868. [26] C. Yavuzturk, J.D. Spitler, Comparative study of operating and control strategies for hybrid ground-source heat pump systems using a short time step simulation model, ASHRAE 106 (2000) 192e209. [27] M. Mokhtar, M. Stables, X. Liu, J. Howe, Intelligent multi-agent system for building heat distribution control with combined gas boilers and ground source heat pump, Energy Build. 62 (2013) 615e626. [28] W. Jinggang, G. Xiaoxia, Study of Operating Control Strategies for Hybrid Ground Source Heat Pump System with Supplemental Cooling Tower, 2009, pp. 511e514. [29] Environment Canada, Temperature over the last 25 years (annual data) for Fort McMurray.http://fortmcmurray.weatherstats.ca/charts/temperature25years.html; (accessed August, 2014). [30] Environment Canada, Heating degree days over the last 25 years (annual data) for Fort McMurray. http://fortmcmurray.weatherstats.ca/charts/hdd-25years. html; (accessed August, 2014). [31] Direct Energy Regulated Services. Your Electricity Rates, 2013. http://www2. directenergy.com/direct-energy-regulations-educational/about-your-rates/ your-electricity-rates.aspx (accessed August, 2014). [32] Tyfocor Chemie GmbH. Tyfocor L. Technical Information, 2009. http://www. resol.de/Produktdokumente/TYFOCOR-L.daten.pdf (accessed July, 2015). [33] Natural Resources Canada, Natural Gas: a Primer, 2014. http://www.nrcan.gc. ca/energy/natural-gas/5641#produced (accessed August, 2014). [34] U.S. Energy Information Administration, Carbon Dioxide Emissions Coefficient, 2013. http://www.eia.gov/environment/emissions/co2_vol_mass.cfm (accessed August, 2014). [35] ICAX Interseasonal Heat Transfer. Seasonal Performance Factor, 2007. http:// www.icax.co.uk/Seasonal_Performance_Factor.html (accessed July, 2015). [36] Direct Energy Regulated Services, Your Natural Gas Rate, 2013. http://www2. directenergy.com/direct-energy-regulations-educational/about-your-rates/ your-natural-gas-rates.aspx (accessed August, 2014). [37] D. Thevenard, Methods for Estimating Heating and Cooling Degree-days to Any Base Temperature, ASHRAE, 2011. LV-11e021:884-891. [38] Government of Canada, Climate e Daily Data Report for March, 2013, 2013. http://climate.weather.gc.ca/climateData/dailydata_e.html? timeframe¼2&Prov¼AB&StationID¼49490&dlyRange¼2011-10-20%7C201408-18&cmdB1¼Go&Year¼2013&Month¼3&cmdB1¼Go# (accessed August, 2014).
Contents lists available at ScienceDirect
Building and Environment journal homepage: www.elsevier.com/locate/buildenv
Post-design system control for integrated space heating systems in residential buildings in cold regions Xinming Li, Mustafa Gül*, Mohamed Al-Hussein Department of Civil and Environmental Engineering, University of Alberta, 9105 116th Street NW, Edmonton, Alberta, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 March 2015 Received in revised form 2 August 2015 Accepted 16 August 2015 Available online 19 August 2015
Ground source heat pump is an effective and environmentally friendly system for space heating and cooling in residential buildings. For cold regions, where heating demand is greater than cooling demand, multi-source heat pumps integrating renewable energy sources are an alternative that ensures occupant comfort. This paper presents a research methodology for investigating the performance of multi-source heat pump systems in residential buildings in cold regions. One year of monitoring data is collected from a four-storey residential building located in Fort McMurray, Alberta, Canada. This data is analyzed to evaluate the performance of the multi-source heat pump system. Enhanced heating system control is also devised which shows potential benefits for system efficiency improvement and non-renewable energy source savings. A system control set point forecasting model is also proposed for the purpose of reducing operation costs. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Renewable energy Ground source heat pump Coefficient of performance Forecasting model Energy savings Post-design system control
1. Introduction Greenhouse gas emissions in Canada have increased rapidly in recent decades, having grown at a rate of 9 mega tons per year from 591 mega tons in 1990 to the peak point of 749 mega tons in 2007 [1]. There has been significant improvement in reducing CO2 emissions since 2007 and researchers continue investigating renewable energy sources to replace non-renewable energy sources for heating, transportation, and industry. This paper focuses on the use of renewable energy sources for residential heating purposes. The use of renewable resources brings environmental benefits through the reduction of greenhouse gas emissions. Due to the benefits of energy savings and low CO2 emissions, ground source heat pump (GSHP) system is shown to be effective in both commercial and residential buildings for indoor heating, cooling and domestic hot water supply. Akpan et al. [2] showed that the geothermal heat pump is an effective thermal system for cold regions, both economically and environmentally (reducing greenhouse gases by 28% compared with the conventional system), compared with photovoltaics or other thermal collectors. The
* Corresponding author. E-mail address: [email protected] (M. Gül). http://dx.doi.org/10.1016/j.buildenv.2015.08.015 0360-1323/© 2015 Elsevier Ltd. All rights reserved.
ground-water heat pump system is also found to offer savings in electrical energy compared with both air-conditioner for cooling and coal-fired boiler for heating [3]. Coefficient of performance (COP) is an effective metric to evaluate the system performance of heat pump systems. The COP for heating is defined as the ratio of energy production to the power consumption of corresponding equipment, whereas the COP for cooling is defined as the ratio of thermal energy removed to the power consumption of corresponding equipment. Benli and Durmus [4] monitored a ground-source heat pump for a year and half in Turkey, having the COP of the GSHP in the range from 2.3 to 3.8 and the system COP in the range from 2 to 3.5. As other studies have shown, GSHP operation hours influence the COP of the heat pump as well. The COP of heat pumps decreases after a long period of operation in cold areas [5]. Involving more compressors may also reduce the COP for the heat pumps and for the entire system. The system performance can be analyzed in daily, monthly, and seasonal increments. A representative factor is Seasonal Performance Factor (SPF), which evaluates the system performance by heating season or cooling season [6,7]. The GSHP has a higher COP/ SPF when the demands for heating and cooling are balanced than when the demands are unbalanced, i.e., the GSHP is sensitive to the weather and potential risks exist when the GSHP is the only dependable system in regions with unbalanced heating and cooling demands [8]. Li et al. [9] explained and compared the geothermal
314
X. Li et al. / Building and Environment 94 (2015) 313e324
field energy distribution of operating the GSHP with both balanced and imbalanced thermal demands, highlighting the significance of using an integrated system. Other researchers have focused on achieving a steady temperature in an underground geothermal field utilized as one of the energy sources. Von Cube and Steimle [10] discussed the advantages of combining a geothermal heating system with other heat sources. As they pointed out, pre-heating the thermal source for the heat pump can lead to an increase of system COP. Li et al. [11] substituted the use of boilers by using a combination of groundwater flow and GSHP for both heating and cooling in twelve greenhouses in a cold region, Akabira, Japan. Their study suggested that adding groundwater flow in the field could reduce the fluctuation of field temperature caused by imbalanced heating and cooling demands, which could further achieve higher system efficiency. The system they described contributed to CO2 emission reductions of 20% and 22% compared with air-source heat pump and kerosene system, respectively. Combining a solar energy system with GSHP is another alternative, which has drawn interest among researchers. The idea of using solar energy to assist GSHP was proposed as a desirable system by Metz in 1982. He compared the results from both experimental and computing (through TRNSYS simulation software) approaches. However, the results were found to vary based on different soils, climates, and experimental designs [12]. The depth of the ground exchanger also influences the results. Wang et al. [13] simulated a solar-ground coupled heat pump with better COP for a long-term cooling and heating period due to geothermal field heat recovery and storage during the transitional seasons and summer. They showed that the system performs better with the vertical geothermal exchanger at a depth of more than 50 m than other depth of vertical exchanger. It was also shown that the complexity and properties of soil affect the heat transfer, where water content in the soil facilitates better thermal conduction for the use of the GSHP. Bi et al. [14] measured the performance of a solar source heat pump, a GSHP, as well as of a solar-assisted ground source heat pump (SGSHP) based on an experimental system under observation for five months. They evaluated the feasibility of using SGSHP and provided valuable experimental results. Bakirci et al. [15] reviewed research results published prior to 2011 and noted that the focus in previous research had been on either solar-assisted heat pump or GSHP individually, with only limited research focusing on systems that integrate both sources. More recently the solar-assisted GSHP system is of increasing interest among researchers. Research results have proven that the utilization SGSHP is advantageous both economically and environmentally compared with conventional energy sources [16,17]. Numerous simulations have been conducted by researchers to analyze SGSHP for the purpose either of revising the existing system or proposing new system designs prior to practical implementation [18e20]. Si et al. [12] designed two solar-GSHP systems for space heating, cooling, and hot water heating and introduced two models. In the first model, the solar thermal collectors not only reheat the circulating fluid from the geothermal field to the heat pump during winter, but also deliver the surplus energy in order to recover the field in summer, while the second model only stores solar energy to the water tank in daytime and recharges the field at night. The results of a simulation carried out as part of their study showed that the first model had better capacity for restoring field temperature than the second model. The system discussed in this paper resembles the first model from the Si et al. study. The challenge of using the GSHP for only heating but not for cooling in a cold region is even bigger. Kjellson et al. [13] simulated different combinations of solar collectors and geothermal exchangers to reduce the electricity demand in the system. They identified two key benefits: (1) that the solar energy system produces heat in summer
and recharges the ground in winter; and (2) that, with the support of the solar energy, the thermal interference between boreholes can be reduced so that the ground temperature will be reduced. Other researchers have also analyzed SGSHP systems built on experimental data based on a system model in a lab or small building [21e26]. Experimental analysis has also validated the notion of using solar as a thermal collector during the daytime and storing energy to the ground for the recovery support of the geothermal field [21]. These studies have revealed that the efficiency of ground heat exchangers can be enhanced from a COP of 2.37e2.72. Ozgener and Hepbasli [22,23] monitored a solar-assisted vertical GSHP heating system for a greenhouse in Turkey and analyzed the performance of this system in terms of energetic and exergetic aspects. The average COP of GSHP achieved 2.84 while the average exergy efficiency was 68.11%, and the COP of entire system was found to be 5e20% lower than the COP of the heat pump itself. Dikici and Akbulut [24] experimented with a solar-assisted heat pump system in a test room and obtained a higher COP of 3.08 during the season with heating demand. Another SGSHP was tested in Turkey with a heat pump COP in the range of 3.0e3.4 [15]. Yu et al. [25] combined solar energy with GSHP with the aim of achieving equilibrium in the geothermal field energy for a villa in the cold region of Beijing, China. Their results showed that the SGSHP not only satisfies heating and cooling demand but also keeps the geothermal field temperature balanced for long-term usage and facilitates a higher COP of the system. After a long-term investment and acceptance of integrating GSHP with other systems, some researchers have also sought the development of improved control strategies for the system being operated. Yavuzturk and Spitler [26] proposed several control strategies for system operation by simulating the combination of using a supplemental heat rejecter with a cooling-dominated GSHP for a small office building. Other researchers have conducted experiments on the system and enhanced the system control based on the experimental results. Mokhtar et al. [27] utilized a type of artificial neural network system to propose an intelligent building management system control in order to minimize the energy waste, optimize load control, and increase the system efficiency. Wang et al. [28] adjusted the system cooling control strategy based on experiment data collected from a specified short period of time. They recommended a system control strategy reference for the hybrid GSHP system in practice. Thus, existing experimental data provides useful information for future reference. Without the enhanced system control algorithm, the considerable benefits of using renewable source energy as the heat source may not be maximally achieved. The limitation of these research results is that there is always variation from one system to the other. The system control strategy devised for one system may not be suitable for other systems. However, in this paper, we propose a system control improvement strategy based on live sensorbased system monitoring and live system performance analysis. The research results not only can serve as a reference for future study and implementation, but the methodology can also be implemented in other systems for the purpose of system control. Furthermore, this research is focused on integrated GSHP system control, thereby addressing a gap in the existing body of research. 2. Research objective and methodology The main objective of the research presented in this paper is to propose enhanced system control and improve the performance of an integrated heating system in a residential building under occupancy based on comprehensive, live, sensor-based experimental monitoring. The framework can be applicable to other buildings as well, with modifications in the control algorithm and set points.
X. Li et al. / Building and Environment 94 (2015) 313e324
The integrated heating system monitored in this study combines the energy from solar thermal collectors and a drain water heat recovery (DWHR) system to support the GSHP system, as the multisource heat pump (MSHP), in order to improve the COP of the system. The purpose of this design is twofold: (1) in winter, to collect the energy from solar thermal collectors, DWHR and geothermal field in order to assist the heat pumps for heating; and (2) in summer, to store the surplus solar energy and energy recycled from the DWHR in summer in order to recover heat loss in the geothermal field. A research framework of system analysis (as shown in Fig. 1) is proposed to improve system control based on live monitoring data. From the live sensor-based monitoring, the system performance can be recorded and analyzed, and will be further used for the comparison between renewable and non-renewable energy sources, developing a forecasting model and proposing enhanced system control. The framework input requires the data collections of temperature, flow rate, pressure, pump status, and electricity usage of heating equipment, as well as local outdoor temperature data and the unit prices of natural gas and electricity. Temperature, flow rate, and pressure in the circulating fluid in system pipes are meaured using temperature sensor, flow rate meter, and pressure sensor, respectively. OneOff status sensors are installed on the main circulating pumps and CT power meters to capture the electricity consumption from circulating pumps and heat pumps. Based on the data collection and analysis, the energy production and efficiency of each system can be calculated using Equations (1) and (2).
Q ¼ C r Rv t TSupply TReturn COP of MSHPðheatingÞ ¼
QHP
source
277:78 þ EMSHP EMSHP
where Q: Thermal Energy, kilojoules (kJ)
(1)
(2)
315
C: Specific Heat, kJ/kg C
r: Density of fluid, kg/m3
Rv: Volume flow rate of fluid, m3/hr t: Serving time or operation time, hr TSupply: Temperature of supply pipe, C TReturn: Temperature of return pipe, C QHP_source: Energy production from source side of heat pumps, GJ EMSHP: Multi-source heat pump electricity consumption, kWh The efficiency of each system can be identified using this method. And, if the system works with low efficiency, it requires further modification by either changing the system control algorithm or revising the original system design. Usually, revising the system design is challenging and has high risk of wasting the original capital cost or incurring additional cost. Thus, modification of the existing system control algorithm is more cost-effective, and can be achieved by raising the corresponding system corrective measures based on the system performance analysis. After comparing the performance of the existing system to the estimated performance of the system with proposed corrective measures, an enhanced system control algorithm can be achieved. The expected outcomes of this research can be associated with both environmental and economic benefits. With regard to the environmental aspect, researchers are able to learn about CO2 emission reduction resulting from renewable sources superseding non-renewable sources in the monitored heating system. Economically, because the total operational costs, including operation fees and energy costs, varies due to the fluctuations in energy source unit prices, a forecasting model aimed at minimizing operational costs is also established. The outcome of this forecasting model also contributes to the overall system control algorithm. There are also constraints petaining to the research results, especially the location and weather conditions of the monitored building, system design, and occupant behavior. These constraints affect the analysis of the findings. Consequently, the experimental design provides the researchers
Fig. 1. Research methodology for system control analysis based on live integrated heating system monitoring.
316
X. Li et al. / Building and Environment 94 (2015) 313e324
to access and analyze data in order to ascertain the system performance parameters and to analyze the potential improvement of the existing heating system design. As output, an approach to enhanced heating system control is proposed and shown to be more environmentally-friendly than the existing system control, also resulting in cost savings. The same experimental and analysis methodology can also be applied to other projects for system performance analysis. 3. Methodology implementation The methodology is implemented in a four-storey, 1295 m2 (13,937 ft2) wood-framed residential building in Fort McMurray, Alberta, Canada. Fort McMurray is located in a cold-climate region that experiences an average annual temperature of 1.3 C and average heating degree days (HDD) of 6094 [29,30]. As depicted in, the building mainly relies on the MSHP system for space heating of 70 apartments units (combination of both one-bedroom and twobedroom apartments) through domestic heating radiators. The geothermal system is only used for heating, and 10 solar panels as well as 11 DWHR systems are integrated in order to support the geothermal system for underground field energy recovery. Electricity is utilized to operate all these renewable energy source systems. Natural gas-fired boilers, as the secondary system, are the auxiliary heat supply in winter. The system schematic drawing, including sensor locations, is given in Fig. 2. The integrated heating system is controlled by a Building Management System (BMS). The solar energy component of the system operates when the temperature in the geothermal field return pipe is lower than the temperature at the solar panel. Freeze protection is designed to be activated in winter when the outdoor temperature falls below 2 C. In this case the freeze protection system runs for 20 min at a time. The DWHR pump runs based on the temperature of the drainage pipe (set point is 2 C). DWHR pump refers to the circulating pump on each drainage stack, while the injection pump refers to the upstream circulating pump serving all 11 stacks. The heating mode is “On” when the outdoor temperature is below 15 C. Heat pumps remains “On” until the outdoor temperature falls to 20 C, at which point the boilers turn “On”. The heat pump and boiler operations are controlled by
specific temperature set points of the hot water tank. The existing system control presents a number of opportunities for potential improvement. This research is focused on the system production performance and system efficiency analysis. Based on the collected data, the authors are able to propose an improved and efficient control system based on observation and analysis of the existing system. The system controls of the solar energy system, DWHR system, heat pumps, and boilers, highlighted in Fig. 2 are separately analyzed in greater detail, and strategies for improved system control are proposed later in this paper. To analyze the system operation performance under occupancy, Li et al. [9] proposed an experimental design to monitor the system based on three months of system performance data. The research presented in this paper is built on the study described in Li et al., which presented the implementation of the proposed methodology using one year of data (March 15, 2013 to March 14, 2014) collected from 32 sensors. The data analyzed is drawn from 12 temperature sensors, 10 OneOff status sensors, six CT power meters, and four flow rates on the circulating pipes. The temperature sensorsdThermocouples (TX)dare installed on the supply and return pipes of each system and the flow rates (FlowX) represent Ultrasonic flow meter measurements taken on March 15, 2013. In addition, six electricity sensors (CTX-X) are installed on the field circulating pumps and heat pumps, which are the main power consuming devices, in order to monitor their power usage. The status of whether the circulating pumps and heat pumps (STX-X) are “On” or “Off” is also recorded. These sensors are collecting data at intervals ranging from 10 s to 1 min. Table 1 presents the detailed energy production and efficiency calculation for each system using the data from corresponding sensors.
3.1. Annual energy production, system efficiency and operation cost from renewable sources (solar, DWHR, geothermal field) Based on the data collected from March 15, 2013 to March 14, 2014, annual heating system performance can be evaluated monthly. Hourly intervals of data are selected for the calculations in this paper. The fluctuations in monthly heating performance are presented graphically for each system.
Fig. 2. Integrated heating system schematic with sensor locations [9].
X. Li et al. / Building and Environment 94 (2015) 313e324
317
Table 1 Circulating fluid and energy production calculations (legends refer to Fig. 2) [9]. System
Fluid
Energy production (GJ)
Solar energy system
13.6% Methanol
DWHR
13.6% Methanol
MSHP
15% Glycol
Boilers
15% Glycol
QSolar ¼ 4.00 kJ/kg C 971.68 Kg/M3 Flow3 t (T5 T4) 106 COPSolar ¼ QSolar/(ST1 ST1_pump_power) QDWHR ¼ 4.00 kJ/kg C 971.68 Kg/M3 Flow6 t (T12 T13) 106 COPDWHR ¼ QDWHR/ST10 ST10_pump_power) QMSHP ¼ 3.90 kJ/kg C 1016.98 Kg/M3 Flow5 t (T3 T1) 106 COPMSHP ¼ QMSHP/CT1e4 QBoiler1 ¼ 3.90 kJ/kg C 1016.98 Kg/M3 Flow4_1 t (T14 T1) 106 QBoiler2 ¼ 3.90 kJ/kg C 1016.98 Kg/M3 Flow4_2 t (T15 T1) 106 COPBoiler1 ¼ QBoiler1/(ST6 ST6_pump_power) COPBoiler2 ¼ QBoiler2/(ST7 ST7_pump_power)
Note: pump power draw rates are obtained from the pump specifications.
3.1.1. Solar energy system Ten 2.87 m2 solar panels are installed at 45-degree obliquity on the south roof of the residential building. 13.6% methanol is used as the energy transmission fluid in the system. The solar energy system production is calculated for each month, as presented in Fig. 3. The solar energy system is found to produce 52 GJ, but to consume 6 GJ (1664 kWh) of electricity at a cost of $145 (all dollar amounts in this paper are in Canadian Dollars) for the time period under investigation. Energy waste occurs in winter time during the annual performance as estimated in the calculation. This energy waste is due to the system freeze protection, since the solar heat transfer station (including the primary and secondary circulating pumps) is required to run every 20 min when the outdoor temperature is below 2 C. As a result, 10%e20% or more of energy production from the solar energy system is wasted through solar system upper loop, which is exposed to outdoor conditions in winter nights, and the loop is found to waste energy instead of generating energy in December, 2013 and January, 2014. The power usage from the secondary circulating pump, which runs continuously, is also included in the COP calculation. The monthly average COP of the solar energy system is also found to decrease from 14 to below 0 from summer to winter. Because of the more abundant solar energy in summer, and the energy waste in winter as discussed previously, the COP of the solar energy system in summer is higher than the COP in winter. The average monthly production cost of using conventional electricity to drive the circulating pumps is found to be $3.6/GJ, with $2.1/GJ in summer and $8.9/GJ in winter based on the monthly electric price varying from $0.07/kWh to $0.11/kWh in the course of the year of data collection [31]. It should be noted here that the energy delivery fee or other fees are not included in energy unit prices discussed in this paper. In order to improve the COP of the solar energy system, the authors proposed increasing the concentration of the anti-freeze fluid (Tyfocor L) to 50%, where the freeze protection is not
required until the outdoor temperature drops to 40 C. In this case, the need for circulating freeze protection can be reduced or eliminated. Based on the technical information of Tyfocor L, the specific heat capacity has linear relation to the fluid temperature while the density changes non-linearly versus the fluid temperature [32]. The changes of specific heat capacity and density along with the temperature of the fluid have been calculated. On the basis of the recalculation, the system obtains higher COP in winter but lower energy production in summer. The electricity consumption can be reduced by 13.2%, corresponding to an operational cost decrease from $145 to $127 for the monitored year; however, the energy production will decrease from 52 GJ to 48 GJ, because, the higher the concentration of the anti-freeze fluid, the lower specific heat it will have, thus the lower energy it will gain. Thus, based on this analysis and in consideration of the high costs of materials and labor in Fort McMurray, it is not worth replacing the anti-freeze fluid as it has a long payback period (approximately 30e50 years). However, in the future heating system design, designers should emphasize selecting the optimal concentration of antifreeze fluid. 3.1.2. DWHR system Eleven DWHR pipe stacks are installed to recycle the wasted energy in hot water from the 70 apartment units. One of these stacks is monitored for the purpose of this study, and it is assumed that the production of each stack is equivalent. The monthly energy production and monthly average COP of the DWHR are presented in Fig. 4. For 11 stacks in total, the DWHR system can produce 340 GJ by consuming 51 GJ of electricity (14 MWh) with monthly production varying from 15.6 GJ to 35.2 GJ, as illustrated in Fig. 4. The energy production depends on both the efficiency of the DWHR system and also the hot water consumption by occupants. The monthly average COP of the DWHR ranges from 3.8 to 8.7, with a production cost of $3.6/GJ. Evidently, the DWHR system works with
Fig. 3. Solar energy system production and monthly average COP.
318
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 4. DWHR system production and monthly average COP.
higher efficiency in winter than in summer due to the fact that occupants consume more hot water in winter than in summer. However, the DWHR units run regardless of the volume of hot water coming to the drainage pipe. Since the injection pump is always “On”, electricity consumption is wasted if the recycled energy does not exceed the electricity usage of the injection pump and circulating pump. In order to improve the COP of the DWHR system, the control point for running the circulating pumps should be adjusted for each drain water stack. Based on the temperature of the drainage pipe, diverse set points are proposed for different seasons. According to the analysis of energy production versus the electricity usage of the circulating pumps, set points are 6 C from March to May, 10 C from June to October and 4 C from November to February. It is anticipated that this adjustment will prevent unnecessary power consumption and improves system efficiency. After recalculating the expected system performance with the proposed control set points, ideal electricity savings of 10% can be achieved, reducing the annual operation cost from $1202 to $1088 with 12.7 GJ in corresponding energy reductions. The DWHR system production cost can be reduced from $3.6/GJ to $3.2/GJ and the COP can be improved by proportions ranging up to 39.9%, as plotted in Fig. 5. Therefore, recommendations are devised to adjust the set points of the drain pipes, to vary from 4 to 10 C, according to the given month/season. However, it should be noted that this recommendation is based on the data collection from one DWHR stack. The more stacks that can be monitored the more accurately the set points can be estimated. Another limitation of the present study has to do with the fact that the set points are fixed numbers. Ideally, the set points vary based on occupant water usage demand and geothermal heating system working conditions. The next step of this project is to generate a method by which to obtain the set points automatically based on timely temperature data.
3.1.3. Geothermal field Two field circulating pumps drive the geothermal field to generate heat in the winter and increase the temperature on the source side of four water source heat pumps. 80 underground geothermal heat exchangers (vertical U-tube) are buried at a depth of 91 m underground. The energy production from the field differs seasonally, as presented in Fig. 6. It provides heat in the winter when heat pumps are in operation and recovers the heat in the summer and occasionally in the winter when heat pumps are not in operation. In total, the geothermal field is found to have provided 236 GJ of energy in winter and to have recovered from the solar energy system and DWHR system 182 GJ of energy over the course of the year of data collection. Overall, it produces 54 GJ more energy than the amount recovered; thus, corrective measures should be taken to recover more energy to the field by either using less of the geothermal field as a heating source or providing more energy sources to support the geothermal field recovery. Of the 182 GJ of recovered energy in the geothermal field, the solar energy system accounts for 16% and the remaining 84% is from the DWHR system (A further comparative analysis between these two energy sources is presented in the next section.). The total operational cost of the geothermal field is $2,812, corresponding to 118 GJ (32,783 kWh) of electricity consumption. These are valuable data for future similar heating system design.
3.2. Operation cost-effectiveness analysis between renewable energy sources The renewable energy sources used in this project are solar energy, DWHR, and geothermal energy. Renewable sources are environmentally friendly since they incur comparably less greenhouse gas emissions in producing energy. However, electricity consumption is still necessary for system operation, thus the operational costs of each source must be compared. Based on the analysis of system production costs from Section
Fig. 5. Enhanced COP of DWHR system.
X. Li et al. / Building and Environment 94 (2015) 313e324
319
Fig. 6. Geo-field energy production and heat recovery.
3.1, the solar energy system is more cost-effective than the DWHR system in summer while the opposite is true in winter. After implementing the proposed post-design control system, the production cost of the DWHR is expected to be lower than that of the solar energy system. However, this only includes the operational cost. If the initial cost is also taken into consideration, the DWHR is more cost-effective based on life cycle analysis since the initial cost of the solar energy system is much higher than the DWHR system. On the other hand, there are limitations to the DWHR system, as its production rate is sensitive to occupant water usage and the system control algorithm. The production of DWHR is constrained by the water usage of occupants, whereas the solar energy system is constrained by the surface area of solar panels as well as weather conditions. For the renewable source energy systems, beside the capital costs, the only input of system operation is electricity consumption. During the year of data collection, $4159 is determined to have been expended to power the systems, corresponding to 48 MWh of electricity consumption. Of this, 3.5% is for the solar energy system, 29.1% is for the DWHR, while 67.4% is for powering the geothermal field circulating pumps. Consequently, as shown in Fig. 7, 446 GJ of energy is produced: 12.1% by the solar energy system, 76.2% by the DWHR system, and 11.7% by the geothermal field. Total energy output comprises purchased electricity as well as 271 GJ of thermal energy produced by the renewable sources. 15.7 tons of CO2 emissions have been avoided owing to the current renewable source system design, compared with using a conventional natural gas heating system. This estimation is based on the natural gas usage standards and statistics specified by Natural Resources Canada [33] and the CO2 emissions coefficients given by the U.S. Energy Information Administration [34]. This amount of CO2 is approximately equivalent to the CO2 emissions produced by 60,704 km of automobile travel.
3.3. Energy production comparison among resources (MSHPs, boilers) The energy produced by all the renewable sources is transferred to heat pumps, which increases the temperature on the source side of the heat pumps. The energy produced by the MSHPs and boilers is conveyed to two hot water tanks for heat preservation. The hot water tanks are connected with the final heating supply and return loop in order to deliver the heated water to the heating radiator in the apartments. 3.3.1. Multi-source heat pump production The MSHPs contain the water-to-water heat pumps, with the solar energy, DWHR system energy, and geothermal energy as sources. Sensors are located on the load side of the heat pumps, collecting the temperature and flow rate data for heat pump energy production, i.e., the energy transmitted from the MSHPs to the hot water tanks. Assuming the flow rate ranges from 148 to 170 standard liters per minute (SLPM), the energy production from heat pumps can be estimated as specified in Fig. 8. It varies monthly depending on the heating demand and the control system. The MSHPs are found to have produced the most energy in April and November within the monitored year, when the daily average outdoor temperature is higher than 20 C and the heating supply mainly relies on the heat pumps. In the remaining months of the cold season, natural gas-burning boilers afford sufficient energy to satisfy the heating demands of the occupants. In total, the system is found to have produced 690 GJ of energy while consuming 350 GJ (97 MWh) of electricity at a cost of $7089. The MSHP system operates with a seasonal-average COP (commonly known as Seasonal Performance Factor (SPF) [35]) for heating of 2.24. The monthly average COP of this system fluctuates, as illustrated by the curve in Fig. 8. Evidently, they perform with higher monthly
Fig. 7. Electricity input and corresponding energy output from each renewable source system.
320
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 8. MSHP energy production and monthly average COP.
average COP (approximately 2e2.5) in winter, while they perform with lower monthly average COP (below 2) in summer. Two factors explain this result of low efficiency in summer. First, because the system is only used for heating, not for cooling, the heating demand is low in summer. However, the heat pumps are still in heating mode, although sometimes only one heat pump is operated for heating. This results in a waste of electricity, which further decreases the COP of the heat pumps. Second, typically in summer, the geothermal field is recovering and cannot provide sufficient heat for summer nights when the outdoor temperature falls below 20 C. Thus, the COP of MSHPs can be improved by turning heat pumps into idle mode when there is less heating demand in summer in order to reduce power usage and increase COP. Moreover, their COPs can also be improved by adding more energy on the source side of the system or decreasing the heating demand on the load side. 3.3.2. Natural gas-burning boilers During the year of monitoring, the boilers play a vital role for heating supply in winter, which accounts for 58% of the annual heating load, while the remaining 42% of the heating load is provided by the MSHPs. The unit price of natural gas differs monthly from $1.7/GJ to $9.0/GJ based on Direct Energy historic data for customers in the relevant region of Alberta, Canada [36]. The natural gas and electricity unit price comparison is given in Table 2 (energy delivery fee is not included). Fig. 9 presents the input and output during the monitored period, showing MSHPs to have consumed 350 GJ (97 MWh) of electricity, costing $7,089, and produced 690 GJ of energy, while the boilers have consumed 3037 m3 of natural gas, costing $4,650, and produced 966 GJ of energy. Thus, the production cost of MSHPs is higher than the production cost of boilers. Due to the fact that the unit cost of natural gas is lower than electricity in Fort McMurray during the monitoring period, a production rate comparison between the boilers and MSHPs is also analyzed, as illustrated in Fig. 10. Overall, the production cost of the natural gas burned by the boilers is less than that of the MSHPs. The exception is March 2014, due to the rapid increase in the unit price of natural gas to $9/GJ in March, 2014. $9/GJ is observed to be an appropriate threshold, such that
natural gas prices above this warrant changing from boilers to MSHPs as a cost-effective solution. That is to say, if the unit price of natural gas is higher than $9/GJ and electricity remains approximately $0.1/kWh, the MSHP is worth using economically. However, the threshold varies monthly because of the fluctuation of unit price of these two sources. Overall, it is found that MSHPs can achieve a total production cost of $11.5/GJ to $13.2/GJ with the current electricity price vacillating between $0.07/kWh to $0.11/ kWh, whereas natural gas would reach a total production cost of $4.4/GJ with a boiler efficiency of 85%, based on the natural gas price range from $1.78/GJ to $9.02/GJ for Alberta delivery. It is therefore more cost-effective to use the boiler as the only heater for the entire building until the natural gas price increases beyond the threshold of approximately $10.0/GJ. The total operation cost from boilers and MSHPs is found to be approximately $12,000 during the year of monitoring. Due to the cost-effectiveness of the boiler, if the boiler is the only heating supply, the total operation cost would be estimated to be approximately $8,000, resulting in $4000 in savings (see Fig. 11). However, the use of solar and DWHR to assist the GSHP system makes it much more environmentally-friendly, as 15.7 tons of CO2 emissions are avoided owing to the renewable source system design. More economical system design can also be achieved if the COP of the heat pumps increases. Thus, based on the system monitoring design and data analysis methodology, building owners can encounter environmental benefits and/or heating system operation cost savings. The research results also lead to a better system control for future system operation. A post-design system control that will be further detailed and summarized in the next section has been proposed based on the current analysis. 3.4. Post-design system control Based on the analysis conducted with the monitoring data, it is determined that the performance of the heating system can be improved and the energy used by the heating system can be reduced. In the interest of energy saving and cost efficiency, a postdesign system control is proposed as represented in Fig. 12. As discussed in Section 3.1.1, due to the high investment cost of
Table 2 Unit price of natural gas and electricity [31,36]. Source
Time 2013
Natural gas Electricity
($/GJ) ($/kWh) ($/GJ)
2014
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
2.93 0.07 19.44
3.76 0.08 22.22
4.19 0.07 19.44
4.59 0.07 19.44
1.85 0.11 30.56
1.78 0.11 30.56
2.32 0.11 30.56
2.43 0.08 22.22
3.72 0.08 22.22
3.30 0.08 22.22
4.09 0.09 25.00
4.57 0.07 19.44
9.02 0.07 19.44
Although the cost of electricity is not common to be presented in $/GJ, it is given in this table for comparison purpose.
X. Li et al. / Building and Environment 94 (2015) 313e324
321
Fig. 9. Energy production comparison between MSHPs and boilers.
Fig. 10. Production rate comparison between MSHPs and boilers.
Fig. 11. Operation cost of the MSHP and boiler.
changing the solar energy system design, no adjustment is proposed for the current stage of the solar energy system (Fig. 12a). And according to the analysis in Section 3.1.2, changing the control system set points of the drain pipes (current range is from 4 C to 10 C) for different months and seasons is recommended to improve the performance of the DWHR system (Fig. 12b). In terms of the MSHPs, more problems are identified and corresponding suggestions are devised. To obtain a higher COP of heat pumps, three solutions are listed below. The post-design system control is summarized in Fig. 12c, with proposed changes denoted by the dark-shaded objects in the figure.
higher temperature. In this case, the geothermal field will recover more energy within a year, making the energy recovery roughly equal to the energy production. Further analysis regarding this suggestion will be discussed in Section 3.5. 2) Reduce the leaving load temperature (the outlet temperature on the load side of the heat pump). The energy demand on the load side of heat pumps also affects their working performance. The COP of heat pumps can be improved by decreasing the set point of the hot water tank, particularly in summer when less heating load is needed.
1) Adjust the set point temperature for switching from MSHPs to boilers.
3) Group heat pumps into pairs and set boiler priority.
Since the overall energy production of the geothermal field is higher than energy recovery and the production cost of natural gas is lower than that of electricity for the monitored period, the MSHP can be utilized with less frequency. This can be approached by increasing the switching point of heat pumps and boilers to a
The heat pump consumes a small amount of power when it is in idle mode. In most cases, only one or two heat pumps are required to operate in spring and autumn. The remaining two heat pumps waste power when they are in idle mode for long periods. To avoid this unnecessary electricity consumption, an effective method can
322
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 12. Post-design system control.
be to group the four heat pumps into two pairs, which, meanwhile, also offers benefits in terms of reduced machine maintenance needs.
3.5. Forecasting model by optimizing switching point of heating systems A forecasting model is under development to obtain the optimal switching point between heat pumps and boilers, which can be updated based on current energy source unit price information. In this model, the inputs are annual heating degree days/hours, heating load supply, and electricity and natural gas unit prices. Heating degree days/heating degree hours (HDD/HDH) represent the heating demand for a building. This can be calculated as the sum of the differences between the daily/hourly average
temperatures and the base temperature as illustrated in Equation (3) [37]. This base temperature represents the indoor temperature controlled by occupants. Assuming the base temperature for this residential building averages 18 C (recommended by Government of Canada) in the monitored year, the HDD versus daily heating supply by the heating system is plotted in Fig. 13, with a high correlation coefficient of 88.7%. The monitored year is observed to have been unusually cold, particularly in December, 2013, and to have included 54 days when the outdoor temperature fell below 20 C. The total heating load is 1634 GJ, while the HDD is 6,438, which is higher than the Fort McMurray average HDD (6094) for the past 25 years [38]. The outdoor temperature data from the monitored sensor is verified using Canadian Government climate statistics, with a correlation coefficient of 99.5% between the data from system monitoring and the government climate statistics [38].
Fig. 13. HDD and daily heating supply.
X. Li et al. / Building and Environment 94 (2015) 313e324
Thus the outdoor temperatures collected are deemed to be reliable and meaningful for the geographical location of the monitored residential building.
HDD ¼
i¼N X ðTb Ti Þ
(3)
i¼1
where. Tb: Base temperature ( C) Ti: Daily average temperature ( C) N: Number of days
323
However, the unit price of energy source varies monthly. The next step in developing the forecasting model involves incorporating fluctuations in the energy unit price over time. The resident can opt for a yearly fixed energy price contract with an energy company. In addition, the base temperature may differ among apartment units. Thus, a more extensive model can be established based on occupant information, apartment vacancy status, and forecast weather conditions. The research team is seeking to ultimately achieve an intelligent forecasting model capable of adjusting the set points based on energy unit price fluctuations over time. 4. Limitations of the experimental data
HDH is chosen as the input for forecasting since the correlation coefficient between HDH and hourly heating supply is 0.947. A linear and a non-linear relationship are assumed in order to obtain the formulary relationship between hourly heating energy supply and outdoor temperature. Setting average daily outdoor temperature as the variable, the heating energy supply can be calculated, generated formulas as given in Equation (4) and Equation (5). The R square values of these deduced equations are 0.8514 and 0.9548, respectively (Fig. 14). The operation cost can also be ascertained, counting energy unit price in the calculation. Whether or not the outdoor temperature exceeds the system switching point determines which energy source to use.
Qsupply ¼ 0:2738 Tout þ 4:7259ðTout < 17 CÞ
(4)
2 0:2433 Tout þ 2:9162ðTout < 20 CÞ Qsupply ¼ 0:0063 Tout
(5) where Qsupply: Daily heating energy supply (GJ) Tout: Average daily outdoor temperature ( C) The forecasting model is able to recalculate the total system operation cost under various system switching temperatures, as illustrated in Fig. 15. Scenarios with set points of 19 C, 18 C, 17 C, 16 C, 15 C are implemented in this study. After recalculating the operation cost under these scenarios, 16 C is selected as the optimal set point, as it incurs the least annual operation cost under both linear and non-linear estimations. Economically, with the new set point, the proposed designed system offers savings in the range of 10.9%e12.1% in operational costs by using linear and non-linear regression, respectively. This outcome is analogous to the conclusion obtained by Nguyen et al. that the comparably low rate of natural gas results in GSHP systems being less economical to install for heatingdominant buildings in North America [8]. The model is still under construction as of the time of submission of this paper.
The sensor-based monitoring can be improved if it is designed before the construction of the building. The sensors installed for the present experiment are surface sensors connected to the pipes. If permitted by building management, inserted sensors would be a better option yielding more accurate data, which would further increase the reliability of the results. However, due to limited access to the heating system in the case study and the budget of this project, the current experimental design is the best design achievable under the given circumstances. 5. Conclusion and future study This project targets energy saving and renewable resource applications, particularly by means of a GSHP. A GSHP supported with energy from solar and DWHR sources is proposed to satisfy energy demand in a residential facility through an innovative and energy efficient approach. This multi-source space heating system for residential building is monitored for one year according to sensorbased experimental design. Based on data collection, energy production calculation and COP analysis for each part of the system is carried out. Results indicate that the usage of the geothermal heating system only for heating, not for cooling, is a challenge. A better system control is recommended with comprehensive live monitored data collection and analysis. The authors have also analyzed the system performance on the aspects of environment and economy. The results show that the use of an integrated heating system leads to the environmental benefit of an annual reduction of CO2 emissions by 15 tons compared with a natural gas system. In terms of economics, an enhanced postdesign system control is also investigated to increase the existing system efficiency, as well as to minimize the operational cost of the heating system. The enhanced post-design system control recommendations can be achieved by programming system control algorithms. A heating system operation cost forecasting model is also established for estimating the switching temperature criteria of using heat pumps and boilers, which should ultimately result in system operation savings in the range of 10.9%e12.1%, as expected. In this model, the outdoor temperature and the unit price of
Fig. 14. Fitting data with linear and non-linear regression.
324
X. Li et al. / Building and Environment 94 (2015) 313e324
Fig. 15. Estimated total operation cost savings for systems.
electricity and natural gas are the inputs. This forecasting model will be further developed, including integration of the fluctuation of energy unit price and full automation and adjustment of setting switching point. Acknowledgments This research work has been developed within the scope of a project undertaken by the University of Alberta, entitled, “Integrated Construction and Energy Management System for Affordable Housing in Fort McMurray”. The authors would like to thank the Natural Sciences and Engineering Research Council of Canada (NSERC), CRDPJ 411101-10, as well as project sponsors, Wood Buffalo Housing and Development Corporation, Integrated Management and Realty Ltd., Cormode & Dickson Construction Ltd., Hydraft Development Services Inc., TLJ Engineering Consultants, and BCT Structures, for their support and assistance. References [1] Government of Canada, National greenhouse gas emissions, Environ. Can. (2014). https://www.ec.gc.ca/indicateurs-indicators/default.asp?lang¼en&n¼ FBF8455E-1 (accessed August, 2014). [2] I.E. Akpan, M. Sasaki, N. Endoh, Optimization of domestic-size renewable energy system designs suitable for cold climate regions, Jpn. Soc. Mech. Eng. (JSME) Int. J. Ser. B 49 (4) (2006) 1241e1252. [3] N. Zhu, P. Hu, W. Wang, J. Yu, F. Lei, Performance analysis of ground watersource heat pump system with improved control strategies for building retrofit, Renew. Energy 80 (2015) 324e330. [4] H. Benli, A. Durmus¸, Evaluation of ground-source heat pump combined latent heat storage system performance in greenhouse heating, Energy Build. 41 (2) (2009) 220e228. [5] Y. Liu, Y.X. Wang, Y.Z. Zhang, Experimental study on a ground source heat pump system in cold area, Appl. Mech. Mater. 71e78 (2011) 2566e2571. Montero, J.F. Urchueguía, Analysis of the energy n, A. [6] C. Montagud, J.M. Corbera performance of a ground source heat pump system after five years of operation, Energy Build. 43 (12) (2011) 3618e3626. [7] C.J. Wood, H. Liu, S.B. Riffat, An investigation of the heat pump performance and ground temperature of a piled foundation heat exchanger system for a residential building, Energy 35 (12) (2010) 4932e4940. [8] H.V. Nguyen, Y.L.E. Law, M. Alavy, P.R. Walsh, W.H. Leong, S.B. Dworkin, An analysis of the factors affecting hybrid ground-source heat pump installation potential in North America, Appl. Energy 125 (2014) 28e38. [9] X. Li, M. Gül, T. Sharmin, I. Nikolaidis, M. Al-Hussein, A framework to monitor the integrated multi-source space heating systems to improve the design of the control system, Energy Build. 72 (2014) 398e410. [10] H.L. Von Cube, F. Steimle, Heat Pump Technology, Butter Worths, London, 1981. [11] H. Li, K. Nagano, Y. Lai, K. Shibata, H. Fujii, Evaluating the performance of a large borehole ground source heat pump for greenhouses in northern Japan, Energy 63 (2013) 387e399. [12] P.D. Metz, The use of ground-coupled tanks in solar-assisted heat-pump, J. Sol. Energy Eng. 104 (4) (1982) 366e372. [13] F. Wang, Z. Liu, Z. Li, Y. Liu, Z. Wang, M. Zheng, Performance analysis of seasonal soil heat storage air conditioning system in solar ground coupled heat pump, in: Proceedings, International Conference on Electronic & Mechanical Engineering and Information Technology, IEEE, Harbin City, China, Aug. 12e14, 2011, pp. 1599e1602. [14] Y. Bi, T. Guo, L. Zhang, L. Chen, Solar and ground source heat-pump system, Appl. Energy 78 (2004) 231e245. [15] K. Bakirci, O. Ozyurt, K. Comakli, O. Comakli, Energy analysis of a solar-ground source heat pump system with vertical closed-loop for heating applications,
Energy 36 (5) (2011) 3224e3232. [16] J.B. Zhao, Comprehensive utilization of solar assisted heat pump and ground source heat pump in severe cold region, Appl. Mech. Mater. 448e453 (2013) 2790e2793. [17] C. Xi, L. Lin, Y. Hongxing, Long term operation of a solar assisted ground coupled heat pump system for space heating and domestic hot water, Energy Build. 43 (8) (2011) 1835e1844. [18] F. Busato, R. Lazzarin, M. Noro, Ground or solar source for space heating: Which is better? Energetic assessment based on a case history, Energy Build. 102 (2015) 347e356. [19] Q. Si, M. Okumiya, X. Zhang, Performance evaluation and optimization of a novel solar-ground source heat pump system, Energy Build. 70 (2014) 237e245. €m, B. Perers, Optimization of systems with the com[20] E. Kjellsson, G. Hellstro bination of ground-source heat pump and solar collectors in dwellings, Energy 35 (6) (2010) 2667e2673. [21] W. Yang, L. Sun, Y. Chen, Experimental investigations of the performance of a solar-ground source heat pump system operated in heating modes, Energy Build. 89 (2015) 97e111. [22] O. Ozgener, A. Hepbasli, Performance analysis of a solar-assisted groundsource heat pump system for greenhouse heating: An experimental study, Build. Environ. 40 (2005) 1040e1050. [23] O. Ozgener, A. Hepbasli, A parametrical study on the energetic and exergetic assessment of a solar-assisted vertical ground-source heat pump system used for heating a greenhouse, Build. Environ. 42 (2005) 11e24. [24] A. Dikici, A. Akbulut, Performance characteristics and energy-exergy analysis of solar-assisted heat pump system, Build. Environ. 43 (11) (2008) 1961e1972. [25] T. Yu, P. Liu, G.M. Chu, Y.F. Li, Seasonal underground storage of SGSHP system design based on equilibrium geothermal temperature, Adv. Mater. Res. 512e515 (2012) 864e868. [26] C. Yavuzturk, J.D. Spitler, Comparative study of operating and control strategies for hybrid ground-source heat pump systems using a short time step simulation model, ASHRAE 106 (2000) 192e209. [27] M. Mokhtar, M. Stables, X. Liu, J. Howe, Intelligent multi-agent system for building heat distribution control with combined gas boilers and ground source heat pump, Energy Build. 62 (2013) 615e626. [28] W. Jinggang, G. Xiaoxia, Study of Operating Control Strategies for Hybrid Ground Source Heat Pump System with Supplemental Cooling Tower, 2009, pp. 511e514. [29] Environment Canada, Temperature over the last 25 years (annual data) for Fort McMurray.http://fortmcmurray.weatherstats.ca/charts/temperature25years.html; (accessed August, 2014). [30] Environment Canada, Heating degree days over the last 25 years (annual data) for Fort McMurray. http://fortmcmurray.weatherstats.ca/charts/hdd-25years. html; (accessed August, 2014). [31] Direct Energy Regulated Services. Your Electricity Rates, 2013. http://www2. directenergy.com/direct-energy-regulations-educational/about-your-rates/ your-electricity-rates.aspx (accessed August, 2014). [32] Tyfocor Chemie GmbH. Tyfocor L. Technical Information, 2009. http://www. resol.de/Produktdokumente/TYFOCOR-L.daten.pdf (accessed July, 2015). [33] Natural Resources Canada, Natural Gas: a Primer, 2014. http://www.nrcan.gc. ca/energy/natural-gas/5641#produced (accessed August, 2014). [34] U.S. Energy Information Administration, Carbon Dioxide Emissions Coefficient, 2013. http://www.eia.gov/environment/emissions/co2_vol_mass.cfm (accessed August, 2014). [35] ICAX Interseasonal Heat Transfer. Seasonal Performance Factor, 2007. http:// www.icax.co.uk/Seasonal_Performance_Factor.html (accessed July, 2015). [36] Direct Energy Regulated Services, Your Natural Gas Rate, 2013. http://www2. directenergy.com/direct-energy-regulations-educational/about-your-rates/ your-natural-gas-rates.aspx (accessed August, 2014). [37] D. Thevenard, Methods for Estimating Heating and Cooling Degree-days to Any Base Temperature, ASHRAE, 2011. LV-11e021:884-891. [38] Government of Canada, Climate e Daily Data Report for March, 2013, 2013. http://climate.weather.gc.ca/climateData/dailydata_e.html? timeframe¼2&Prov¼AB&StationID¼49490&dlyRange¼2011-10-20%7C201408-18&cmdB1¼Go&Year¼2013&Month¼3&cmdB1¼Go# (accessed August, 2014).