- Email: [email protected]
An Energy Flow Management Algorithm for a Photovoltaic Solar Home
Available online at www.sciencedirect.com
ScienceDirect Energy Procedia 111 (2017) 934 – 943
8th International Conference on Sustainability in Energy and Buildings, SEB-16, 11-13 September 2016, Turin, ITALY
An energy flow management algorithm for a photovoltaic solar home Fathia Chekired*, Zoubeyr Smara, Achour Mahrane, Madjid Chikh, Smail Berkane Unité de Développement des Equipements Solaires /UDES Centre de Développement des Energies Renouvelables /CDER, 42415.W.Tipaza,Algérie
Abstract In this paper, an energy flow management algorithm for a grid-connected-photovoltaic system with battery storage devoted to supply a home is presented. This algorithm favours the fulfilment of the home energy demand by the energy produced by the photovoltaic generator or stored in the batteries than that got from the grid. This algorithm manages the flow of energy in the house through the combination of four switches. It was applied to the case of a home installed in a coastal region of Bou-Ismaïl (Algeria).The efficiency of the algorithm was tested for two weeks, a favourable week in summer and an unfavourable week in winter regarding the weather conditions. The simulation of the on grid PV system was done using real data of the irradiation and the temperature acquired by the meteorological station of the Bou-Ismaïl site and a home load profile for each season. The results obtained revealed that the energy demand satisfaction for the house is high in the favourable summer week and is only 33% in the winter unfavourable week. To meet a high rate of self-consumption a Home Energy Management is required. © 2017 Published by by Elsevier Ltd.Ltd. ThisPeer-review is an open access article under the CC BY-NC-ND license 2017The TheAuthors. Authors. Published Elsevier under responsibility of [KES International.]. (http://creativecommons.org/licenses/by-nc-nd/4.0/). K d h t lt i (PV) l h t ti fil Peer-review under responsibility of KES International. Keywords: photovoltaic (PV); solar home,; energy management; consumption profiles;
1.Introduction In an energy environment where the energy demand is continuously growing, the fossil resources are declining and the global warming is dramatically increasing, many countries have opted for the adoption of measures to reduce energy consumption and for an energy transition using renewable energy sources. This implies a new architecture of the energy supply system, which moves from a centralized to a decentralized
* Corresponding author. Tel: +213-244-102-00; fax: +213-244-101-33.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of KES International. doi:10.1016/j.egypro.2017.03.256
935
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
generation. In this context, photovoltaic (PV) generation appears as the most promising alternative thanks to its maturity, its environmentally friendly characteristics, low maintenance and the fact that the sun is everywhere. The residential and building sector is in many countries the first energy consumer, about 40% of the global energy consumption [1], therefore efficient utilization and control of energy consumption at household level is crucial [2]. The challenge resides in the matching of the intermittent energy production with the dynamic power demand [3]. This needs to design an energy management strategy in order to optimize the use of PV source and storage and to match the local production with local consumption. The available energy produced by these sources and the home energy demand while insuring user comfort. In the literature, several works are particularly focused on optimizing the energy. This optimization is based on the study and analysis of the houses consumption profiles. Consequently, studies have been directed to "smart homes" [4-7]. In this paper the case of a home powered by on grid photovoltaic system with storage is investigated. An energy flow management algorithm was developed. It manages the energy flow available in the home through a combination of switches to meet the energy demand optimally using the energy produced by the photovoltaic generator and/or the energy stored in the batteries and/or the energy drawn from the grid. The purpose is to minimize this last and ultimately reach the self-consumption mode. Two particular weeks have been chosen, one in summer and one in winter, in order to tests the algorithm. The paper is organized as follows. In Section 2, the chosen photovoltaic system which feeds the home to satisfy the energy demand profile is presented. The models used to simulate the PV system are given in section 3. The proposed energy flow management algorithm and the operating mode of the system are presented in section 4, while in Section 5, the obtained simulation results, using the MATLAB-SIMULINK are given and interpreted. 2. PV System description and energy demand of the home 2.1. Description of the proposed PV system In order to study how to meet the energy needs of a family by using mainly the energy produced by a photovoltaic installation, the sizing of the PV system has been done by using PVSST 1.0 locally developed software [8], which used real irradiation and temperature data of the site, an autonomy of one day and the home load profile. The optimal PV system configuration obtained is an array of 3.2kWp and a battery bank of 12kWh. To feed the loads of the house, 4kW DC/AC converter has been chosen. In order to prevent the batteries from a total discharge, the lowest level of the State Of Charge (SOC) was set to 25%. This SOC corresponds to a capacity of 3kWh. The SOCmax was set to 85% which corresponds to 10kWh. As shown on figure 1, the PV system operation depends on the combination status of four switches (K1, K2, K3, K4) which is related to the energy demand, the energy available from the PV generator, the SOC of the battery bank and the energy taken off from the grid. In order to meet optimally the energy needs of the home, an energy flow management algorithm has been developed. Energy management K3 K4 DC PV Array
Ipv
Iload DC AC
AC DC K1
K2 Ib Battery
Fig. 1. Grid-connected PV system with storage for the solar home.
Load
936
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
2.2. Solar home energy demand The case treated in this paper is a house of 57m2 area situated in the coastal region of Bou-Ismaîl (Latitude: 36 ° 38 ‘33’’North and Longitude: 36 ° 38 ‘33’’North) occupied by a family of three members. The energy demand of a home depends on many factors such as the number of appliances, the electrical power used by each appliance and the amount of use of appliances determined by the behavior of the occupants in the home. 2.2.1. Home appliances The home appliances considered in the study are listed in the Table 1 with their power ratings and estimated operating time per day. Table 1. Typical home appliances power ratings and estimated operating time per day [7, 9]. Appliance Cold
Appliance Type Refrigerator Hi-Fi Iron Vacuum PC TV 1 TV Receiver box Hob Washing machine Oil bath radiator air conditioner Lighting
Consumer Electronics
Cooking Wet Electric space heating Air conditioning Lighting
Mean cycle power 120 100 1000 2000 100 124 40 2000 1500 1800 1500 30
Quantity 1 1 1 1 1 1 1 1 1 1 1 5
Number of operating hours 12 h/day 2h/day (1/2h)/week (1/2h) /week 2h/day 3h/day 3h/day 3h/day 1h/ 3 days Just in winter :3 to 4h/day Just in summer : 3 to 4h/day All the 5 lighting ;12h/day
2.2.2. Home energy demand To design the most efficient PV power system for supplying the dwelling, the solar energy resources of a site and the energy consumption of the household should be known. Unfortunately, the electricity demand is rarely available, so the daily load profile must be generated. In the studied case, the home load profile has been determined by considering the daily load profile of each appliance, their power ratings and their estimated operating time per day, then their summation leads to the daily consumption home profile. To have a realistic load profile, the electrical consumption of a given week for winter and summer has been considered as shown on figure 2.
Load consumption (Kwh)
3.5
Summer profile Winter profile
3
2.5
2
1.5
1
0.5
0
0
20
40
60
80
100
120
Time in hours Fig. 2. One week home load profile for winter and summer.
140
160
180
937
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943 Table 2. Energy demand seasonal figures. Season
Weekly min value in kWh
Weekly max value in kWh
Daily average in kWh
Winter
12.30
18.30
13.92
Summer
13.87
16.27
14.64
Table 2 summarizes the main figures of the seasonal load profile. For the summer, the daily energy demand varies between 13.87kWh and 16.27kWh. For the winter, the energy demand varies between 12.03kWh and 18.30kWh. For these two seasons, the high energy demand is explained by the use of energy intensive equipment for heating and air conditioning. 3. PV system modeling In this section, the modeling of the PV generator, the batteries and the power converter will be presented. For PV system modeling, a one diode model for the PV generator has been used [10]. For the batteries, the Copetti’s model has been chosen as it is well suited for lead acid batteries allowing the simulation of the whole operating process charge–discharge-overcharge while considering the temperature change of the battery [11, 12]. For the DC/DC converter, a boost converter controlled by a MPPT Fuzzy logic Controller has been used [13, 14] and the DC/AC converter has been modeled by its yield curve. 4.Energy flow management strategy proposed for the PV solar home To improve the performance of the studied PV system devoted for a solar home, an energy management algorithm must be implemented in order to: • Produce a maximum power from the photovoltaic generator, • Protect the batteries against overcharge and deep discharge, • Satisfy the energy needs of the user by avoiding energy shortage, • Power the grid when there is an excess of energy. 4.1.Operating modes of the PV system To meet the energy management priority set for the house, the electricity generated by the PV system must first feed into the electrical appliances, if there is any excess energy this will be stored in the battery bank, the rest of energy will be if any fed into the grid. The energy flow available in the house was led under the control of the algorithm to the planed target according to one of the following operating modes. • Mode1: The PV arrays generate sufficient energy to feed the load and charge the battery. • Mode 2: The battery is charged, and the PV arrays generate sufficient energy to the load, so the PV injects the energy produced on the Grid. • Mode 3: In this mode, the energy available in PV array is not sufficient to supply the load, the battery bank supplements the energy required by the load. • Mode 4: The PV energy production is insufficient and the battery bank is completely discharged, the load is fed by the grid. Table 3. Switches operating mode. Switch Mode 1
K1 On
K2
K3 Off
Off
K4
Status
Off
PV feeding loads, Batteries in charge The excess energy produced by the PV is injected into the grid. Batteries supply the energy required by the loads The loads are fed by the grid.
Mode 2
On
Off
On
Off
Mode 3
Off
On
Off
Off
Mode 4
Off
Off
Off
On
938
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
4.2.Energy flow management algorithm The figure 3 depicts the detailed flowchart of the flow energy management algorithm. The main idea is that the photovoltaic production should be utilized as much as possible to reach the self-consumption. The priority is given to the supply of the loads, then the charge of the batteries and at last, the excess energy is fed into the grid. The algorithm operation is based mainly on the PV production, the SOC of the battery and the Ibat which indicates if the battery is in charge or discharge and if it is able to feed the loads or not. Depending on the state of theses parameters one of the modes indicated in the previous paragraph is used. Start Start Input data : Consumption Profile, Irradiation, Temperature,…) PV power produced, Power consumption, The battery charge current
No No No
PPRO=PLOA
PPRO
PPRO>PLOAD
Yes
Yes
Ibat=0
Ibat=Ibat-charge
No
SOC>SOCMAX
Yes
Ibat=Ibat-discharge No
Yes
PV feeding loads Mode1
SOC>SOCMin
The load is fed by the grid Mode4
PV feeding loads Batteries in charge Mode 1
Batteries supply the energy required by the loads Mode3
Batteries disconnection from the PV Array PV array injects the excess energy produced into the Grid. Mode2
Fig. 3. Energy management strategy.
5.Simulation results and discussion In order to test the efficiency of the flow energy management algorithm the satisfaction of the energy demand of the solar home by the photovoltaic system, has been evaluated for two chosen cases. The first one is ‘a favorable week’ for the PV production relatively to weather conditions in summer and the second one is ‘an unfavorable week’ in the winter. For each case, depending on the PV produced and the load energy profile a combination of the switches, directs the energy flow to the given target. The simulations were done using the irradiation and the temperature data acquired by the meteorological station of the site of Bou-Ismaïl. The PV electrical production was calculated using the PV system model presented briefly in paragraph 3. The algorithm was performed using the appropriate load profile of each season. The figures 4 and 5 show the states of the K1, K2, K3 and K4 switches and the variation of the photovoltaic production, the energy demand (load profile), the energy battery storage and the grid energy for the summer and winter week respectively.
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
• Case 1: favorable week in summer
Fig. 4. (a) K1, K2, K3 and K4 states and the variation of PV production, load consumption, battery storage and grid energy versus time for the ‘favorable week’ in summer; (b) the weekly PV production and load consumption for the ‘favourable week’ in summer (21/08/2015-27/08/2015).
939
940
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
• Case2: unfavorable week in winter
Fig. 5. (a) K1,K2, K3 and K4 state and the variation of PV production, load consumption, battery storage and grid energy versus time for the ‘unfavorable week’ in winter; (b) the weekly PV production and load consumption for the ‘unfavorable week’ in winter (31/01/2015-06/02/2015).
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
From the results presented in figure 4 and 5, the excess energy produced in the favorable week in summer is 8.5%, this percentage corresponds to an energy of 9kWh which is injected to the grid. On the other hand, in winter, 67% of the energy demand of the unfavorable week is satisfied by the grid (65kWh) this is due to the low photovoltaic energy production during this season. To better explain the energy flow in the studied PV system, two particular cases will be discussed. The first one corresponds to a favorable day in summer (25/08/2015), and the second corresponds to an unfavorable day in winner (03/02/2015). • Favorable day in summer
Fig. 6. (a) K1, K2, K3 and K4 state; (b) the variation of PV production, load consumption, battery storage and grid energy versus time for a ‘favorable day’ in summer (25/08/2015).
941
942
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
• Unfavorable day in winter
Fig. 7. (a) K1, K2, K3 and K4 states ;(b) the variation of PV production, load consumption, battery storage and grid energy versus time for ‘unfavorable day’ in winter (03/02/2015).
As it can be seen on figure 6, between midnight and 5am there is no PV production and the energy demand is very low, the battery is at its low level (25%). Between 5am and 5pm, the PV array produces 2kWh. This electric
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
production is so high that it can at the same time satisfy the energy demand of the loads, the charging of the batteries and supply the excess energy to the grid. After 5pm, as the PV production is low, the batteries supply a part of the demand. Between 7pm and 9pm all the energy needed is taken from the batteries. After 9pm the batteries continues to discharge in the loads but the demand is so high that the rest of the energy needed was withdrawn from the grid. Figure 7 indicates that, for the chosen day it can be noticed that the irradiation is at low level all the day and this conducts to a low PV production. In this case, the energy demand is exclusively satisfied by the grid. The battery bank is all day long near its lowest level (25%). The highest energy demand is situated between 5pm and 9pm.
6.Conclusion Thanks to the decreasing of the cost of PV installations, the grid parity comes a reality. Nowadays, it is not profitable as before to sale the produced photovoltaic electricity. It is more interesting for the PV owners to use it for their own needs and try to reach their self-consumption. In this article, as a first step an energy flow management algorithm which directs the energy flow according to priorities set was presented. It appears that even if the energy flow algorithm is necessary it is not sufficient. To go further and to use the PV electricity produced as much as possible, a Home Energy Management (HEM) is needed. This last will allow the balancing between the demand and the generation, by controlling the deferrable loads, reduce the energy consumption for example by using natural light as long as possible, exchanging excess energy produced between neighbours instead of drawing off energy from the grid. Acknowledgements We would like to thank Dr A. Diaf, UDES, for his valuable advices on the writing of this article. References [1] OECD/IEA, 2013. Transition to sustainable buildings. Strategies and opportunities to 2050. [2] Asare-Bediako B, Ramirez Elizondo L.M, Ribeiro P.F, Kling W.L. Consideration of Electricity and Heat Load Profiles for Intelligent Energy Management Systems. UPEC 2011. 46th International Universities' Power Engineering Conference. Soest. Germany; 5-8th September 2011. [3] Riffonneau Y, Bacha S, Barruel F, and Ploix S. Optimal Power Flow Management for Grid Connected PV Systems with Batteries. IEEE Trans. Sustain. Energy VOL. 2, NO. 3, JULY 2011. [4] Gudi N, Wang L, Devabhaktuni V. A demand side management based simulation platform incorporating heuristic optimization for management of household appliances. Int J Elec Power 2012;43(1):185–193. [5] Di Giorgio A, Pimpinella L. An event driven smart home controller enabling consumer economic saving and automated demand side management. Appl Energ 2012; 96:92–103. [6] Chen X, Wei T, Hu S. Uncertainty-aware household appliance scheduling considering dynamic electricity pricing in smart home. IEEE Trans. Smart Grid 2013; 2:932–941. [7] Tascikaraoglu A, Boynuegri A.R, Uzunoglu M. A demand side management strategy based on forecasting of residential renewable sources: A smart home system in Turkey. Energ Buildings 2014; 80:309–320. [8] Chikh M, Mahrane A, Bouachri F. PVSST 1.0 sizing and simulation tool for PV systems. Energy Procedia 2011; 6:75–84. [9] Al-Alawi A, Islam S.M, Demand side management for remote area power supply systems incorporating solar irradiance model, Renew energ 2004; 29:2027–2036. [10] Sera D, Teodorescu R, Rodriguez P. PV panel model based on datasheet values. Proc. IEEE Int.Symp. Ind. Electron. http://dx.doi.org/10.1109/ISIE.2007.4374981 2007; 2392–2396. [11] Copetti J. B, Chenlo F. Internal resistance characterization of lead acid battery for PV rates. Proceedings of the 11th European PV Solar Energy. Conference, Montreux, 12-16 October 1992; 1116-1119. [12] Copetti J. B, Chenlo F. A general battery model for PV system simulation. Prog Photovoltaic Res Appl 1993; 283- 292. [13] Chekired F, Mellit A, Kalogirou S.A, Larbes C. Intelligent maximum power point trackers for photovoltaic applications using FPGA chip: A comparative study. Elsevier, Sol Energy 2014; 101:83–99. [14] Chekired F, Larbes C, Rekioua D, Haddad F. Implementation of a MPPT fuzzy controller for photovoltaic systems on FPGA circuit. Elsevier, Energy Procedia 2011; 6: 541–549.
943
ScienceDirect Energy Procedia 111 (2017) 934 – 943
8th International Conference on Sustainability in Energy and Buildings, SEB-16, 11-13 September 2016, Turin, ITALY
An energy flow management algorithm for a photovoltaic solar home Fathia Chekired*, Zoubeyr Smara, Achour Mahrane, Madjid Chikh, Smail Berkane Unité de Développement des Equipements Solaires /UDES Centre de Développement des Energies Renouvelables /CDER, 42415.W.Tipaza,Algérie
Abstract In this paper, an energy flow management algorithm for a grid-connected-photovoltaic system with battery storage devoted to supply a home is presented. This algorithm favours the fulfilment of the home energy demand by the energy produced by the photovoltaic generator or stored in the batteries than that got from the grid. This algorithm manages the flow of energy in the house through the combination of four switches. It was applied to the case of a home installed in a coastal region of Bou-Ismaïl (Algeria).The efficiency of the algorithm was tested for two weeks, a favourable week in summer and an unfavourable week in winter regarding the weather conditions. The simulation of the on grid PV system was done using real data of the irradiation and the temperature acquired by the meteorological station of the Bou-Ismaïl site and a home load profile for each season. The results obtained revealed that the energy demand satisfaction for the house is high in the favourable summer week and is only 33% in the winter unfavourable week. To meet a high rate of self-consumption a Home Energy Management is required. © 2017 Published by by Elsevier Ltd.Ltd. ThisPeer-review is an open access article under the CC BY-NC-ND license 2017The TheAuthors. Authors. Published Elsevier under responsibility of [KES International.]. (http://creativecommons.org/licenses/by-nc-nd/4.0/). K d h t lt i (PV) l h t ti fil Peer-review under responsibility of KES International. Keywords: photovoltaic (PV); solar home,; energy management; consumption profiles;
1.Introduction In an energy environment where the energy demand is continuously growing, the fossil resources are declining and the global warming is dramatically increasing, many countries have opted for the adoption of measures to reduce energy consumption and for an energy transition using renewable energy sources. This implies a new architecture of the energy supply system, which moves from a centralized to a decentralized
* Corresponding author. Tel: +213-244-102-00; fax: +213-244-101-33.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of KES International. doi:10.1016/j.egypro.2017.03.256
935
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
generation. In this context, photovoltaic (PV) generation appears as the most promising alternative thanks to its maturity, its environmentally friendly characteristics, low maintenance and the fact that the sun is everywhere. The residential and building sector is in many countries the first energy consumer, about 40% of the global energy consumption [1], therefore efficient utilization and control of energy consumption at household level is crucial [2]. The challenge resides in the matching of the intermittent energy production with the dynamic power demand [3]. This needs to design an energy management strategy in order to optimize the use of PV source and storage and to match the local production with local consumption. The available energy produced by these sources and the home energy demand while insuring user comfort. In the literature, several works are particularly focused on optimizing the energy. This optimization is based on the study and analysis of the houses consumption profiles. Consequently, studies have been directed to "smart homes" [4-7]. In this paper the case of a home powered by on grid photovoltaic system with storage is investigated. An energy flow management algorithm was developed. It manages the energy flow available in the home through a combination of switches to meet the energy demand optimally using the energy produced by the photovoltaic generator and/or the energy stored in the batteries and/or the energy drawn from the grid. The purpose is to minimize this last and ultimately reach the self-consumption mode. Two particular weeks have been chosen, one in summer and one in winter, in order to tests the algorithm. The paper is organized as follows. In Section 2, the chosen photovoltaic system which feeds the home to satisfy the energy demand profile is presented. The models used to simulate the PV system are given in section 3. The proposed energy flow management algorithm and the operating mode of the system are presented in section 4, while in Section 5, the obtained simulation results, using the MATLAB-SIMULINK are given and interpreted. 2. PV System description and energy demand of the home 2.1. Description of the proposed PV system In order to study how to meet the energy needs of a family by using mainly the energy produced by a photovoltaic installation, the sizing of the PV system has been done by using PVSST 1.0 locally developed software [8], which used real irradiation and temperature data of the site, an autonomy of one day and the home load profile. The optimal PV system configuration obtained is an array of 3.2kWp and a battery bank of 12kWh. To feed the loads of the house, 4kW DC/AC converter has been chosen. In order to prevent the batteries from a total discharge, the lowest level of the State Of Charge (SOC) was set to 25%. This SOC corresponds to a capacity of 3kWh. The SOCmax was set to 85% which corresponds to 10kWh. As shown on figure 1, the PV system operation depends on the combination status of four switches (K1, K2, K3, K4) which is related to the energy demand, the energy available from the PV generator, the SOC of the battery bank and the energy taken off from the grid. In order to meet optimally the energy needs of the home, an energy flow management algorithm has been developed. Energy management K3 K4 DC PV Array
Ipv
Iload DC AC
AC DC K1
K2 Ib Battery
Fig. 1. Grid-connected PV system with storage for the solar home.
Load
936
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
2.2. Solar home energy demand The case treated in this paper is a house of 57m2 area situated in the coastal region of Bou-Ismaîl (Latitude: 36 ° 38 ‘33’’North and Longitude: 36 ° 38 ‘33’’North) occupied by a family of three members. The energy demand of a home depends on many factors such as the number of appliances, the electrical power used by each appliance and the amount of use of appliances determined by the behavior of the occupants in the home. 2.2.1. Home appliances The home appliances considered in the study are listed in the Table 1 with their power ratings and estimated operating time per day. Table 1. Typical home appliances power ratings and estimated operating time per day [7, 9]. Appliance Cold
Appliance Type Refrigerator Hi-Fi Iron Vacuum PC TV 1 TV Receiver box Hob Washing machine Oil bath radiator air conditioner Lighting
Consumer Electronics
Cooking Wet Electric space heating Air conditioning Lighting
Mean cycle power 120 100 1000 2000 100 124 40 2000 1500 1800 1500 30
Quantity 1 1 1 1 1 1 1 1 1 1 1 5
Number of operating hours 12 h/day 2h/day (1/2h)/week (1/2h) /week 2h/day 3h/day 3h/day 3h/day 1h/ 3 days Just in winter :3 to 4h/day Just in summer : 3 to 4h/day All the 5 lighting ;12h/day
2.2.2. Home energy demand To design the most efficient PV power system for supplying the dwelling, the solar energy resources of a site and the energy consumption of the household should be known. Unfortunately, the electricity demand is rarely available, so the daily load profile must be generated. In the studied case, the home load profile has been determined by considering the daily load profile of each appliance, their power ratings and their estimated operating time per day, then their summation leads to the daily consumption home profile. To have a realistic load profile, the electrical consumption of a given week for winter and summer has been considered as shown on figure 2.
Load consumption (Kwh)
3.5
Summer profile Winter profile
3
2.5
2
1.5
1
0.5
0
0
20
40
60
80
100
120
Time in hours Fig. 2. One week home load profile for winter and summer.
140
160
180
937
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943 Table 2. Energy demand seasonal figures. Season
Weekly min value in kWh
Weekly max value in kWh
Daily average in kWh
Winter
12.30
18.30
13.92
Summer
13.87
16.27
14.64
Table 2 summarizes the main figures of the seasonal load profile. For the summer, the daily energy demand varies between 13.87kWh and 16.27kWh. For the winter, the energy demand varies between 12.03kWh and 18.30kWh. For these two seasons, the high energy demand is explained by the use of energy intensive equipment for heating and air conditioning. 3. PV system modeling In this section, the modeling of the PV generator, the batteries and the power converter will be presented. For PV system modeling, a one diode model for the PV generator has been used [10]. For the batteries, the Copetti’s model has been chosen as it is well suited for lead acid batteries allowing the simulation of the whole operating process charge–discharge-overcharge while considering the temperature change of the battery [11, 12]. For the DC/DC converter, a boost converter controlled by a MPPT Fuzzy logic Controller has been used [13, 14] and the DC/AC converter has been modeled by its yield curve. 4.Energy flow management strategy proposed for the PV solar home To improve the performance of the studied PV system devoted for a solar home, an energy management algorithm must be implemented in order to: • Produce a maximum power from the photovoltaic generator, • Protect the batteries against overcharge and deep discharge, • Satisfy the energy needs of the user by avoiding energy shortage, • Power the grid when there is an excess of energy. 4.1.Operating modes of the PV system To meet the energy management priority set for the house, the electricity generated by the PV system must first feed into the electrical appliances, if there is any excess energy this will be stored in the battery bank, the rest of energy will be if any fed into the grid. The energy flow available in the house was led under the control of the algorithm to the planed target according to one of the following operating modes. • Mode1: The PV arrays generate sufficient energy to feed the load and charge the battery. • Mode 2: The battery is charged, and the PV arrays generate sufficient energy to the load, so the PV injects the energy produced on the Grid. • Mode 3: In this mode, the energy available in PV array is not sufficient to supply the load, the battery bank supplements the energy required by the load. • Mode 4: The PV energy production is insufficient and the battery bank is completely discharged, the load is fed by the grid. Table 3. Switches operating mode. Switch Mode 1
K1 On
K2
K3 Off
Off
K4
Status
Off
PV feeding loads, Batteries in charge The excess energy produced by the PV is injected into the grid. Batteries supply the energy required by the loads The loads are fed by the grid.
Mode 2
On
Off
On
Off
Mode 3
Off
On
Off
Off
Mode 4
Off
Off
Off
On
938
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
4.2.Energy flow management algorithm The figure 3 depicts the detailed flowchart of the flow energy management algorithm. The main idea is that the photovoltaic production should be utilized as much as possible to reach the self-consumption. The priority is given to the supply of the loads, then the charge of the batteries and at last, the excess energy is fed into the grid. The algorithm operation is based mainly on the PV production, the SOC of the battery and the Ibat which indicates if the battery is in charge or discharge and if it is able to feed the loads or not. Depending on the state of theses parameters one of the modes indicated in the previous paragraph is used. Start Start Input data : Consumption Profile, Irradiation, Temperature,…) PV power produced, Power consumption, The battery charge current
No No No
PPRO=PLOA
PPRO
PPRO>PLOAD
Yes
Yes
Ibat=0
Ibat=Ibat-charge
No
SOC>SOCMAX
Yes
Ibat=Ibat-discharge No
Yes
PV feeding loads Mode1
SOC>SOCMin
The load is fed by the grid Mode4
PV feeding loads Batteries in charge Mode 1
Batteries supply the energy required by the loads Mode3
Batteries disconnection from the PV Array PV array injects the excess energy produced into the Grid. Mode2
Fig. 3. Energy management strategy.
5.Simulation results and discussion In order to test the efficiency of the flow energy management algorithm the satisfaction of the energy demand of the solar home by the photovoltaic system, has been evaluated for two chosen cases. The first one is ‘a favorable week’ for the PV production relatively to weather conditions in summer and the second one is ‘an unfavorable week’ in the winter. For each case, depending on the PV produced and the load energy profile a combination of the switches, directs the energy flow to the given target. The simulations were done using the irradiation and the temperature data acquired by the meteorological station of the site of Bou-Ismaïl. The PV electrical production was calculated using the PV system model presented briefly in paragraph 3. The algorithm was performed using the appropriate load profile of each season. The figures 4 and 5 show the states of the K1, K2, K3 and K4 switches and the variation of the photovoltaic production, the energy demand (load profile), the energy battery storage and the grid energy for the summer and winter week respectively.
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
• Case 1: favorable week in summer
Fig. 4. (a) K1, K2, K3 and K4 states and the variation of PV production, load consumption, battery storage and grid energy versus time for the ‘favorable week’ in summer; (b) the weekly PV production and load consumption for the ‘favourable week’ in summer (21/08/2015-27/08/2015).
939
940
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
• Case2: unfavorable week in winter
Fig. 5. (a) K1,K2, K3 and K4 state and the variation of PV production, load consumption, battery storage and grid energy versus time for the ‘unfavorable week’ in winter; (b) the weekly PV production and load consumption for the ‘unfavorable week’ in winter (31/01/2015-06/02/2015).
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
From the results presented in figure 4 and 5, the excess energy produced in the favorable week in summer is 8.5%, this percentage corresponds to an energy of 9kWh which is injected to the grid. On the other hand, in winter, 67% of the energy demand of the unfavorable week is satisfied by the grid (65kWh) this is due to the low photovoltaic energy production during this season. To better explain the energy flow in the studied PV system, two particular cases will be discussed. The first one corresponds to a favorable day in summer (25/08/2015), and the second corresponds to an unfavorable day in winner (03/02/2015). • Favorable day in summer
Fig. 6. (a) K1, K2, K3 and K4 state; (b) the variation of PV production, load consumption, battery storage and grid energy versus time for a ‘favorable day’ in summer (25/08/2015).
941
942
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
• Unfavorable day in winter
Fig. 7. (a) K1, K2, K3 and K4 states ;(b) the variation of PV production, load consumption, battery storage and grid energy versus time for ‘unfavorable day’ in winter (03/02/2015).
As it can be seen on figure 6, between midnight and 5am there is no PV production and the energy demand is very low, the battery is at its low level (25%). Between 5am and 5pm, the PV array produces 2kWh. This electric
Fathia Chekired et al. / Energy Procedia 111 (2017) 934 – 943
production is so high that it can at the same time satisfy the energy demand of the loads, the charging of the batteries and supply the excess energy to the grid. After 5pm, as the PV production is low, the batteries supply a part of the demand. Between 7pm and 9pm all the energy needed is taken from the batteries. After 9pm the batteries continues to discharge in the loads but the demand is so high that the rest of the energy needed was withdrawn from the grid. Figure 7 indicates that, for the chosen day it can be noticed that the irradiation is at low level all the day and this conducts to a low PV production. In this case, the energy demand is exclusively satisfied by the grid. The battery bank is all day long near its lowest level (25%). The highest energy demand is situated between 5pm and 9pm.
6.Conclusion Thanks to the decreasing of the cost of PV installations, the grid parity comes a reality. Nowadays, it is not profitable as before to sale the produced photovoltaic electricity. It is more interesting for the PV owners to use it for their own needs and try to reach their self-consumption. In this article, as a first step an energy flow management algorithm which directs the energy flow according to priorities set was presented. It appears that even if the energy flow algorithm is necessary it is not sufficient. To go further and to use the PV electricity produced as much as possible, a Home Energy Management (HEM) is needed. This last will allow the balancing between the demand and the generation, by controlling the deferrable loads, reduce the energy consumption for example by using natural light as long as possible, exchanging excess energy produced between neighbours instead of drawing off energy from the grid. Acknowledgements We would like to thank Dr A. Diaf, UDES, for his valuable advices on the writing of this article. References [1] OECD/IEA, 2013. Transition to sustainable buildings. Strategies and opportunities to 2050. [2] Asare-Bediako B, Ramirez Elizondo L.M, Ribeiro P.F, Kling W.L. Consideration of Electricity and Heat Load Profiles for Intelligent Energy Management Systems. UPEC 2011. 46th International Universities' Power Engineering Conference. Soest. Germany; 5-8th September 2011. [3] Riffonneau Y, Bacha S, Barruel F, and Ploix S. Optimal Power Flow Management for Grid Connected PV Systems with Batteries. IEEE Trans. Sustain. Energy VOL. 2, NO. 3, JULY 2011. [4] Gudi N, Wang L, Devabhaktuni V. A demand side management based simulation platform incorporating heuristic optimization for management of household appliances. Int J Elec Power 2012;43(1):185–193. [5] Di Giorgio A, Pimpinella L. An event driven smart home controller enabling consumer economic saving and automated demand side management. Appl Energ 2012; 96:92–103. [6] Chen X, Wei T, Hu S. Uncertainty-aware household appliance scheduling considering dynamic electricity pricing in smart home. IEEE Trans. Smart Grid 2013; 2:932–941. [7] Tascikaraoglu A, Boynuegri A.R, Uzunoglu M. A demand side management strategy based on forecasting of residential renewable sources: A smart home system in Turkey. Energ Buildings 2014; 80:309–320. [8] Chikh M, Mahrane A, Bouachri F. PVSST 1.0 sizing and simulation tool for PV systems. Energy Procedia 2011; 6:75–84. [9] Al-Alawi A, Islam S.M, Demand side management for remote area power supply systems incorporating solar irradiance model, Renew energ 2004; 29:2027–2036. [10] Sera D, Teodorescu R, Rodriguez P. PV panel model based on datasheet values. Proc. IEEE Int.Symp. Ind. Electron. http://dx.doi.org/10.1109/ISIE.2007.4374981 2007; 2392–2396. [11] Copetti J. B, Chenlo F. Internal resistance characterization of lead acid battery for PV rates. Proceedings of the 11th European PV Solar Energy. Conference, Montreux, 12-16 October 1992; 1116-1119. [12] Copetti J. B, Chenlo F. A general battery model for PV system simulation. Prog Photovoltaic Res Appl 1993; 283- 292. [13] Chekired F, Mellit A, Kalogirou S.A, Larbes C. Intelligent maximum power point trackers for photovoltaic applications using FPGA chip: A comparative study. Elsevier, Sol Energy 2014; 101:83–99. [14] Chekired F, Larbes C, Rekioua D, Haddad F. Implementation of a MPPT fuzzy controller for photovoltaic systems on FPGA circuit. Elsevier, Energy Procedia 2011; 6: 541–549.
943