- Email: [email protected]
The effects of urban driving conditions on the operating characteristics of conventional and hybrid electric city buses
Applied Energy 135 (2014) 472–482
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
The effects of urban driving conditions on the operating characteristics of conventional and hybrid electric city buses Seref Soylu ⇑ Sakarya University, Faculty of Engineering, 54187 Sakarya, Turkey
h i g h l i g h t s Operating characteristics of conventional and hybrid electric buses were examined. Recovery of braking energy offers an excellent opportunity to improve fuel economy. Speed and altitude profiles of routes have dramatic impacts on the energy recovery. Capacity of the auxiliary power source has a dramatic impact on the energy recovery. Round-trip efficiency of the regenerative braking system was calculated to be 27%.
a r t i c l e
i n f o
Article history: Received 29 May 2014 Received in revised form 25 August 2014 Accepted 27 August 2014
Keywords: Hybrid bus Regenerative braking Real-world driving Energy efficiency Traction energy Braking energy
a b s t r a c t The basic operating characteristics of a conventional bus (CB) and a hybrid electric bus (HEB) were examined under urban driving conditions. To perform this examination, real-time operating data from the buses were collected on the Campus-Return route of the Sakarya Municipality. The main characteristics examined were the traction, braking, engine, engine generator unit (EGU), motor/generator (M/G), and ultracapacitor (Ucap) energies and efficiencies of the buses. The route elevation profile and the frequency of stop-and-go operations of the buses were found to have dramatic impacts on the braking and traction energies of the buses. The declining profile of the Campus-Return route provided an excellent opportunity for energy recovery by the regenerative braking system of the HEB. However, owing to the limits on the capacities and efficiencies of the hybrid drive train components and the Ucap, the bus braking energies were not recovered completely. Braking energies as high as 2.2 kW h per micro-trip were observed, but less than 1 kW h of braking energy per micro-trip was converted to electricity by the M/G; the rest of the braking energy was wasted in frictional braking. The maximum energy recovered and stored in the Ucap per micro-trip was 0.5 kW h, but the amount of energy recovered and stored per micro-trip was typically less than 0.2 kW h for the entire route. The cumulative braking energy recovered and stored in the Ucap for the Campus-Return route was 52% of the available brake energy, which was 13.02 kW h. Consequently, the round-trip efficiency of the regenerative braking system, between the wheels and Ucap, was determined to be 27%. Finally, although the brake engine energy (BEE) of the CB was 1.18 times higher than its positive traction energy (PTE), the BEE of the HEB was only 1.07 times higher than its PTE. In fact, it is normal to expect the BEE to be higher than the PTE owing to power train losses, but the energy recovered by the regenerative braking system was found to cover most of the power train losses and even improve the energy efficiency of the HEB. Ó 2014 Elsevier Ltd. All rights reserved.
Abbreviations: CB, conventional bus; HEB, hybrid electric bus; EGU, enginegenerator unit; M/G, motor/generator; Ucap, ultracapacitor; PE, power electronics; Gen, generator; FE, fuel energy; BEE, brake engine energy; GE, generator energy; PTE, positive traction energy; NTE, negative traction energy; EC, European Commission; USDOE, US Department of Energy; CO2, carbon dioxide; GHG, greenhouse gases; VTP, vehicle technology program; PEMS, portable emission measurement system; GPS, global positioning system. ⇑ Tel.: +90 5359788982. E-mail addresses: [email protected], [email protected] http://dx.doi.org/10.1016/j.apenergy.2014.08.102 0306-2619/Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Both the European Commission (EC) and the United States Department of Energy (USDOE) are taking serious actions to drastically reduce CO2 emissions, with the goal of keeping climate change below 2 °C. The EC has suggested that to reach that goal,
S. Soylu / Applied Energy 135 (2014) 472–482
the transport sector needs to reduce its greenhouse gas (GHG) emissions by at least 60% by 2050, compared to the levels of 1990 [1,2]. Similarly, the USDOE is taking actions to reduce CO2 emissions through its vehicle technology program (VTP), in partnership with automotive industry leaders, national laboratories, and universities [3,4]. The main goals of the VTP are to enable the US to significantly reduce fossil fuel consumption, GHG emissions, and local emissions. GHG emissions are expected to be reduced from 2005 levels by over 40% by 2030 and by over 80% by 2050. Improving the energy efficiency of vehicle power trains by hybridizing them appears to be one of the most promising and cost-effective approaches to achieving these goals [5–8]. It has been reported in literature that more than 30% fuel saving can be expected from commercial vehicles when coupled with hybrid electric power trains [9,10]. The energy efficiency and emissions of city buses operating under urban driving conditions are highly dependent on their operating conditions [11–16]. If operated at cruising conditions on flat routes without auxiliaries (e.g., heating and air conditioning), buses need only enough energy to compensate for rolling and aerodynamic losses, but real-world operations of buses are far from such idealized conditions. Depending on the traffic and road conditions, which may require many short micro-trips with accelerations, decelerations and various road grades, braking energy losses and auxiliaries can increase energy consumption dramatically and reduce fuel economy. Electrical hybridization of city buses, on the other hand, can be a remedy for excessive fuel consumption under urban driving conditions. In comparison to CBs, HEBs can easily save fuel and reduce emissions, for two main reasons. First, the vehicle kinetic energy, which is normally wasted during braking in the case of CBs, can be recovered and stored in the form of electricity during braking in the case of HEBs. Second, since HEBs do not necessarily have mechanical links from their engines to their wheels, their engines can always operate in the optimum region of the fuel consumption map. Therefore, they have very high potential to minimize both fuel consumption and emissions. Computer modeling and simulation programs are highly effective and economical tools for use in examining the effects of design alternatives and energy management strategies on hybrid vehicles before construction of a prototype begins [17–20]. Barrero et al. [21] simulated and compared several power flow management strategies for hybrid city buses using a quasi-static ‘‘backward/forward-looking’’ simulation program. The simulation results indicated that the energy savings that can be achieved are in the range of 32.6% (when using the kinetic strategy and a 0.3-kW h energy storage system) to 40% (when using the ICE on–off strategy and a 0.65-kW h energy storage system). Xiong et al. [22] developed an energy management strategy for a series–parallel hybrid city bus using a forward-facing simulation program based on the Matlab/Simulink software. The simulation results indicated that the engine operation can be maintained in the high-efficiency range and that a theoretical improvement in fuel economy 30.3% can be achieved, compared to that of a conventional bus, over the driving cycle of a transit bus. Ahn et al. [23] simulated a regenerative braking system in a hybrid electric vehicle under various driving conditions using Matlab/Simulink and observed that hybrid electric vehicles with regenerative braking can improve fuel economy by 20–50%. All these modeling efforts have indicated that electrical hybridization of city buses provides various degrees of benefits in terms of energy efficiency and emissions. In order to confirm and quantify the benefits under real-world urban driving conditions, a research project entitled ‘‘Measurement and Modeling of Hybrid City Bus Emissions under Real-World Operating Conditions’’ was initiated at Sakarya University, with the support of the Turkish
473
Ministry of Science Technology and Industry and TEMSA R&D, which is the research and development department of a local bus manufacturer. Confirmation of vehicle emissions levels under real world driving conditions is a requirement by EC regulation [24]. In the first phase of the project, real-time data on the basic operating characteristics of a conventional 12-m city bus were collected under real-world urban driving conditions and analyzed. In the second phase, similar data were collected and analyzed for hybrid electric versions of the bus. Initial tests and reports on the project indicated that the energy efficiencies of the buses are highly dependent on the elevation and speed profiles of the bus routes [25–27]. In the present study, the benefits and barriers of HEBs were further clarified by examining the PTE, NTE, EGU, M/G, and Ucap energies and efficiencies of the HEB under real world urban driving conditions. The results are expected to be a reference for researchers who design and simulate the next generation of city buses.
2. Experimental setup The test vehicles were TEMSA-brand city buses. The buses were 12 m in length and weighed 15 tons when loaded. The CB was powered by a 6.7-liter Euro 4 CUMMINS diesel engine. It has a ZF 6 HP 504 C gear box and a ZF AV 132/80 rear axle. The HEB, which was designed based on body of the conventional bus, had a SIEMENS ELFA hybrid drivetrain system. The hybrid bus was powered by a 6.7-liter Euro 5 CUMMINS diesel engine. The Euro 4 and 5 engines have the same power and torque characteristics. Figs. 1a and 1b are schematic diagrams of the conventional and HEB drivetrains. The basic specifications of the ELFA drivetrain components and the engines are given in Table 1. There are mainly series and parallel power train configuration alternatives for HEBs. However, in order to increase braking energy recovery rate with regenerative braking, series hybrids are generally preferred for city bus applications due to their frequent stop and go operations. As the schematics illustrate, the HE drivetrain (Series Hybrid) does not have a mechanical link from the engine to the wheels. Instead, the engine drives the generator that feeds the M/G. Therefore, the engine can operate in the most efficient regions of its fuel consumption map. Additionally, the Ucap, which can be charged by regenerative braking or by the generator, feeds the M/G, especially during acceleration of the bus.
Fig. 1a. Schematics of the conventional drivetrain and its instrumentation (1. Computer, 2. GPS, 3. SEMTECH DS, 4. Engine).
474
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 1b. Schematics of the HE drivetrain and its instrumentations (PE: Power electronics, Gen: Generator).
The tests were carried out on the Campus-Return route of the Sakarya Municipality. Test data for the vehicle speed, location, engine operation, exhaust emissions, and environmental conditions were sampled every second using a SEMTECH DS manufactured by Sensor, Inc. The Semtech DS is a portable emissions measurement system (PEMS) that is capable of real-time monitoring of gaseous emissions and collection of environmental data such as the ambient pressure, temperature, and humidity. The Semtech DS also has a vehicle interface module (SAE-J1708) that collects engine and vehicle operation data. The data such as the speed, torque, power, and fuel consumption were directly collected from the engine ECU. Test data for the hybrid power train components such as generator power, M/G power, speeds, Ucap charge and discharge powers were collected using the SIADIS interface of the ELFA. In order to minimize adverse effects of temperature on the component performance of the hybrid power train, the component temperatures were kept in a limited range. In order to compute energies, the relevant power data were integrated with a MATLAB program for the entire route. 3. Route and operating characteristics of the buses The speed and elevation profiles of city buses on urban routes are very specific because of the traffic, the frequency of bus stops, and the road grade. Note that the road grade is the percent change in the route elevation per unit of horizontal distance. Even if the same bus is tested repeatedly on the same route with the same driver, the individual tests may have different speed profiles under real-world driving conditions. This is simply the nature of realworld driving. Fig. 2 shows speed and elevation profiles, along with acceleration histograms, for the Campus-Return route. The speed profiles involve many micro-trips, which are the trips between two complete stops. A typical micro-trip is shown in Fig. 3. As the profiles show, the buses experience highly frequent stopand-go events and dramatic elevation changes on this route. The micro-trips on this route typically have accelerations in the range of 2 to 2 m/s2. It should be noted that the elevations were divided by 5 for better presentation of the figure. To determine the real
value of the elevations it must be multiplied by 5. The similarities of the speed profiles can easily be compared with their basic characteristics, which are given in Table 2. Although the buses were driven over the same route, the measurement results differ slightly. A check of the dates of the tests shows that more than two years passed between the tests. During that period, the route was slightly modified because of construction activities. For this reason, the travel distances are slightly different, although the starting and ending points are the same. A comparison of the number of micro-trips and the average micro-trip maximum (Vmax_ave) and mean speeds (Vmean_ave) shows that the CB and HEB speed profiles have similar characteristics. The power consumptions of the buses for the auxiliaries, such as air conditioner, could be slightly different, but they were not measured. It was assumed that they don’t have significant impact on the brake engine energies of the buses. 4. Modeling of the bus traction power If operated at a constant speed over a flat road, a bus will run with the minimum fuel consumption per km travel while its engine providing traction force for the rolling and aerodynamic resistances, only:
F roll ¼ m g C roll
ð1Þ
F aero ¼ ð1=2Þ q C d Af v 2
ð2Þ
where Froll is the force required to overcome rolling resistance, m is the mass of the bus, g is the acceleration of gravity, Croll is the coefficient of the rolling resistance, Faero is the force required to overcome aerodynamic resistance, q is the density of air, Cd is the drag coefficient, Af is the frontal area of the bus, and v is the speed. However, as shown by the speed and altitude profiles in Fig. 2, a typical city bus operation under urban driving conditions involves frequent changes in speed, which is a measure of kinetic energy (KE), and changes in elevation, which is a measure of potential energy (PE). For these conditions, the bus engine must provide enough traction force to increase the KE and PE as well.
F ke ¼ m að1 þ eÞ
ð3Þ
F pe ¼ m g sin h
ð4Þ
where Fke is the force required to increase the KE, a is the bus acceleration, e is a factor for the rotational components of the bus power train, Fpe is the force required to increase the PE, and h is the angle of the road inclination. The traction force and power of the bus can be written as follows:
F traction ¼ F roll þ F aero þ F ke þ F pe
ð5Þ
Ptraction ¼ v F traction
ð6Þ
Integration of positive and negative traction powers over the test route provided the PTEs and NTEs, respectively. The integrations were performed with a MATLAB program. This is a vehicle only
Table 1 Specifications of ELFA drivetrain components.
Brand Rated power Rated torque Rated voltage Maximum speed Weight Capacity (Farad/kW h)
Generator
M/G
Ucap
Engine
Siemens, PM Synch 145 kW @ 4000 rpm 368 N m @ 220 A 700 V 5000 rpm 120 kg –
Siemens, AC Induction 85 kW 220 N m 650 V DC 10,000 rpm 120 kg –
ISE, Thunderpack 233 kW – 720 V DC – 180 kg 10.4/0.748
Cummins, ISB6.7 184 kW @ 2325 rpm 1020 N m @ 1200 rpm – – – –
S. Soylu / Applied Energy 135 (2014) 472–482
475
Fig. 2. Speed, elevation and acceleration profiles of the buses on the Campus-Return route.
5. Examination of the traction and braking energies of the buses
Fig. 3. A typical micro-trip from the Campus-Return route.
Table 2 Basic driving characteristics of the buses on the Campus-Return route.
Travel distance (km) Travel time (s) # of microtrips Vmax_avea (km/h) Vmean_ave.a (km/h) Test date
Conventional bus (CB)
Hybrid electric bus (HEB)
11.08 2086 40 35.48 20.80 11.06.2009
9.78 2214 35 35.16 21.23 19.09.2011
A city bus operating on an urban route must overcome resistance due to rolling, aerodynamic drag, acceleration, and elevation. The PTE needed to overcome these types of resistance is typically supplied by a diesel engine that converts chemical fuel energy (FE) into BEE. Fig. 4 compares the FE, BEE and PTE of the CB for every micro-trip on the Campus-Return route. As the figure shows, the FEs, BEEs and PTEs are highly dependent on the speed and elevation profiles of the micro-trips. As expected, the BEEs are slightly higher than the PTEs. The differences between them correspond to the frictional losses of the power train components. However, the differences between the FEs and BEEs are highly significant; the BEEs correspond to less than 1/3 of the FEs for most of the micro-trips. The ratio of the BEE to the FE is the brake engine efficiency (gbrake) for a micro-trip. As Fig. 5a shows, the gbrake is as high as 40% for the second micro-trip, which is the micro-trip with the highest traction energy, but the efficiency decreases to well below the maximum as the BEE decreases. It is well known from the literature that engine efficiency is maximized at higher engine loads and minimized at lower engine loads, mainly because of the increasing role of parasitic losses, which are pumping and frictional losses, at lower engine loads. As Fig. 5b shows, the brake-specific fuel consumption (bsfc) of the bus increases as the engine power decreases, which clearly indicates the effects of parasitic losses.
a Vmax and Vmean are calculated based on the micro-trip speed. They do not involve boarding time durations at the bus stops.
model and does not cover the components of entire power train system. Beside, since the objective of this work was not to model the coefficients of the rolling and aerodynamic drag of the buses, the coefficients were initially chosen based on Jimenez-Palacios [28] and MOVES2010 [29], but they were slightly modified later to fine-tune the model to better match the experimental data obtained from the tests. To fine-tune the rolling and drag coefficients, the magnitudes of the traction and engine power were compared at various speeds. In the comparison, the dependency of the third power of the aerodynamic resistance on the vehicle speed was particularly notable. The rolling resistance increases linearly with speed. After determining the final values of the coefficients, given below, the traction and braking powers of the CB and HEB were computed.
e ¼ 0:1; C roll ¼ 0:007; C d ¼ 0:72
Fig. 4. The FEs, BEEs and PTEs of the CB on the Campus-Return route.
476
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 5a. The BEEs and engine efficiencies of the CB on the Campus-Return route.
Fig. 7. Comparison of the PTEs and NTEs of the CB on the Campus-Return route.
Fig. 6. The PTEs and NTEs of the CB on the Campus-Return route.
Fig. 8. The FEs, GEs and PTEs of the HEB on the Campus-Return route.
bsfc, gr/kWh
Therefore, the gbrake for a typical city bus operating under urban driving conditions remains well below its maximum. Fig. 6 compares the positive and negative traction energies of the bus on the Campus-Return route. The NTE, which for CBs is the energy lost due to frictional braking, is the braking energy needed to decelerate the bus to a stop or to maintain a desired bus speed on a declining road. As Fig. 6 shows, the NTEs are not homogenous and are highly dependent on the speed and elevation profiles of the micro-trips. While the route declines sharply in the first 600 s, as was shown in Fig. 2, the absolute values of the corresponding NTEs are as high as 2.2 kW h, as shown in Fig. 6. After that time, the NTEs mostly reflect the speed characteristics of the micro-trips. The frequencies and magnitudes of the positive and negative traction energies can be better compared in a histogram, as shown in Fig. 7. The magnitudes of the NTEs are even greater than those of the PTEs because of the long downhill portion of the route. Moreover, although the NTEs and PTEs range from 2.2 kW h to 1.7 kW h, most of the values are confined to a narrower range from 0.5 to 0.5 kW h.
In summary, a CB operating on an urban route is wasting a significant amount of its FE, mainly because of parasitic and frictional braking losses. However, as mentioned earlier, hybridization of conventional buses can minimize these losses, for two reasons. First, HEBs do not necessarily have mechanical links from their engines to the wheels; therefore, their engine may operate at their most efficient conditions and minimize parasitic losses. Second, HEBs generally have regenerative braking systems with auxiliary power sources; therefore, frictional braking losses can easily be minimized as well. Fig. 8 compares the FE, generator energy (GE), and PTE of the HEB. As explained in the Experimental Setup section, there is no mechanical link between the diesel engine and the wheels in the HEB. Instead, the diesel engine drives the generator to generate electricity and feeds the M/G. Therefore, as mentioned in the preceding paragraph, there is an opportunity to operate the engine generator unit (EGU) only at higher loads and efficiencies. However, as Fig. 8 shows, the GE is directly related to the PTE; it still generates electricity primarily based on the speed and elevation profiles of the micro-trips. Actually, the limited capacity of the auxiliary power source prevents the EGU from operating at its most efficient operating conditions. Moreover, a comparison of Figs. 9 and 5a shows that the EGU efficiencies, which are the ratios of the GEs to the FEs, are generally less than the engine efficiency of the CB. Although the engine operating characteristics are the same for both of the buses, the EGU efficiency is lower primarily because of the efficiency factor of the generator. As Fig. 10 shows, the generator efficiency can be as high as 96%, but it is highly dependent on the torque and speed. The generator gear box is another factor that reduces the EGU efficiency. Fig. 8 also shows
1000 800 600 400 200 0
bsfc gr/kWh
0
20
40
60
80
100
120
140
160
180
200
Engine Power, kW Fig. 5b. Bsfc of the CB on the Campus-Return route.
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 9. The GEs and EGU efficiencies of the HEB on the Campus-Return route.
477
Fig. 11. The PTEs and NTEs of the HEB on the Campus Return route.
Fig. 10. The generator efficiency map and the torque data points. Fig. 12. Comparison of the PTEs and NTEs of the HEB on the Campus-Return route.
that the PTEs are generally higher than the GEs, even though the reverse would be expected because of the losses of the power train components. This means that the auxiliary power source is also contributing a significant amount of energy to driving the HEB. The auxiliary power source of the HEB is a 0.74 kW h Ucap that is charged mainly by the regenerative braking system. It is well known from the first law of thermodynamics that energy cannot be destroyed but can be converted into other forms of energy. In this sense, the PTE required to increase the kinetic and potential energies of a bus can be recovered during braking with a regenerative braking system. As mentioned earlier, this is one of the most important advantages of HEBs: they have the capability to convert braking energy into electrical energy and store it in auxiliary power sources, which can be a battery, a Ucap or a combination of the two. While batteries are preferred for their higher energy capacity, Ucaps are preferred for their higher specific powers, which can provide large bursts of power within a few seconds. Fig. 11 compares the PTEs and NTEs of the HEB on the CampusReturn route. As shown in Table 2, the averaged maximum and mean speeds of the HEB are slightly different from those of the CB. However, as with the CB, the absolute values of the NTEs during the first 10 micro-trips increase to as high as 2.2 kW h, and the rest of the traction energies directly reflect the speed profiles of the micro-trips. The magnitudes and frequencies of the traction energies can be better compared in a histogram. As Fig. 12 shows, with the HEB as with the CB, the absolute values of both the PTEs and NTEs are typically less than 0.5 kW h, and the absolute values of the NTEs are typically greater than those of the PTEs. This histogram compares the frequencies and magnitudes of the NTEs recovered by the regenerative braking system. If the braking energies were recovered with 100% efficiency, the bus would only need
energy to overcome rolling resistance and aerodynamic drag. A bus operating only with aerodynamic drag and rolling resistance under urban driving conditions would be highly fuel efficient. However, the real-world operations of buses are far from such idealized conditions. Optimization of the capacity and size of the auxiliary power source of a HEB is very important because of the regeneration efficiency, mass, and cost considerations involved. A histogram such as that shown in Fig. 12 is very useful in the preliminary sizing of the auxiliary power system for a HEB. Because Ucaps are massy and expensive, they must be optimized based on the operating conditions of a city bus to achieve a favorable cost/benefit ratio. As the preceding figures show, the magnitudes and frequencies of the NTEs are not constant, and they are highly dependent on the characteristics of the bus route. The NTEs shown in Fig. 12 indicate that if a Ucap with 2.0 kW h of effective capacity were chosen for braking energy recovery, almost 100% of the NTEs of the Campus-Return route would be recovered by the regenerative braking system, but the Ucap would be very heavy and expensive. If instead a Ucap with an effective capacity of 0.5 kW h were chosen, approximately 75% of the NTE would be recovered, and the rest would be lost owing to mechanical braking. However, during the recovery of the NTE, the energy must be transferred from the wheels to the Ucap over the power train components, which are mainly the M/ G and final drive components. Unfortunately, all power train components have efficiency characteristics for energy conversion and/ or transfer. Fig. 13 compares the M/G PTEs and M/G NTEs of the HEB on the Campus-Return route. The M/G PTE reflects the work that the M/G does as an electrical motor to convert electrical energy into
478
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 13. The GEs and M/G traction energies of the HEB on the Campus-Return route.
mechanical energy. The M/G NTE reflects the work that the M/G does as a generator to convert mechanical energy into electrical energy. The M/G NTE is actually the bus braking energy that is converted into electricity by the M/G. As expected, the trends in the M/ G PTEs and M/G NTEs are similar to those shown in Fig. 11, but the absolute values of the micro-trip traction energies are significantly lower for the NTEs and significantly higher for the PTEs than those shown in Fig. 11. This means that the M/G PTEs are higher than the bus PTEs, to overcome the losses in the M/G and final drive components. Similarly, in the conversion of the bus NTEs to M/G NTEs, the losses of the final drive components and the M/G must be overcome. For these reasons, the M/G NTEs are much smaller than the bus NTEs. It should be noted that the M/G PTEs are greater than the GEs for the entire route. As mentioned earlier, because the auxiliary power source contributes to the M/G PTEs, they are greater than the GEs. The energy conversion efficiency of the M/G can be better explained using Figs. 14 and 15. Fig. 14 shows the torque and speed data points of the M/G on the Campus-Return route. As expected, the operating torques and speeds are widely scattered, which reflects the characteristics of the traction power demanded. Unfortunately, however, the M/G conversion efficiency is not constant across the operating map. As Fig. 15 shows, the efficiency of the M/G increases from 70% at lower speeds to more than 90% at higher speeds, which makes the M/G efficiency one of the most important factors for HEBs. The magnitudes and frequencies of the M/G PTEs and M/G NTEs can be better compared with a frequency histogram. As Fig. 16 shows, unlike the PTEs and NTEs shown in Fig. 12, the absolute values of the M/G NTEs are much smaller than those of the M/G PTEs.
Fig. 14. The speed and torque map of the M/G on the Campus-Return route.
Fig. 15. The M/G efficiency map and the torque data points on the Campus-Return route.
Fig. 16. The PTE and NTE histograms for the M/G on the Campus Return route.
Moreover, the absolute value of the maximum M/G NTE is less than 1 kW h. This means that some of the micro-trip braking energies (especially high-magnitude braking energies) were probably wasted by frictional braking as an energy management strategy, because of size and capacity considerations. Furthermore, as Fig. 17 shows, the absolute values of the M/G NTEs are significantly less than those of the bus NTEs, which clearly indicates the importance of the energy transfer and conversion efficiencies of the M/G and the final drive, as well as the energy management strategy of the HEB.
Fig. 17. Comparison of the magnitude and frequencies of the NTEs and M/G NTEs.
S. Soylu / Applied Energy 135 (2014) 472–482
The importance of the Ucap capacity and the energy management strategy are examined further in Figs. 18 and 19. The Ucap recovery energy is the M/G NTE to be converted to direct current and stored in the Ucap. The stored Ucap energy can optionally be used to contribute to the GE as feedback energy to feed the M/G, especially during bus accelerations. It is also important to note that, as the figures show, although the capacity of the Ucap is 0.74 kW h, the absolute values of the recovery and feedback energies typically remain less than 0.6 kW h. The main reason that the Ucap capacity is not used completely is that the Ucap voltage decreases as it is discharged, so it cannot feed the M/G. Therefore, the Ucap never becomes completely empty. As Fig. 19 shows, the frequencies of the higher-magnitude recovery and feedback energies are very low. This means that the Ucap capacity was not used effectively for the entire route. An important reason for the Ucap capacity not being used effectively is that if the state of charge (SOC) of the Ucap is at its maximum, as shown in Fig. 20, the Ucap cannot be charged any more. Once the SOC reaches its maximum, all available NTEs must be wasted because of the limited capacity of the Ucap. As Fig. 20 shows, after the 5th micro-trip, the SOC is at more than 90% and remains at that level until the 12th micro-trip. This period is the downhill driving period of the bus, as shown in Fig. 2, and there is a significant amount of NTE to be recovered, as shown in Figs. 11 and 13. However, at the beginning of this period, the Ucap SOC almost reaches its maximum, and during this period, both the feedback and recovery energy remain at very low levels of less than 0.2 kW h. Because the Ucap capacity is limited, there is no way to completely recover
479
Fig. 20. The SOC of the Ucap on the Campus-Return route.
Fig. 21. Comparison of the magnitudes and frequencies of the NTE and Ucap recovery energies.
Fig. 18. The recovery and feedback energies of the Ucap on the Campus-Return route.
the NTEs, which have higher magnitudes. Therefore, most of the NTEs available for the recovery are wasted during this period by frictional braking. On the other hand, as Figs. 13 and 18 show, most of the micro-trips on this route do not have such high NTEs to be recovered, and therefore, as can be seen from Fig. 20, the SOC generally remains at lower levels. As Fig. 21 shows, although the absolute magnitudes of the NTEs are as high as 2.2 kW h, the Ucap recovery energies are less than 0.5 kW h, mainly because of the limited capacity of the Ucap. If a higher-capacity Ucap were chosen, it would not be used effectively for most of the route, as the frequencies of the NTEs show. Optimizing the capacity of a Ucap for an urban route and developing an energy management system for effective use of the Ucap capacity are therefore challenging problems whose solutions depend on the speed and altitude profiles of the urban route, which may not be known beforehand. However, the main principle for effective use of Ucap capacity is that the SOC must be at the lowest level possible before braking so that there is enough capacity for the regenerated braking energy. If the SOC is at a very high level before braking, the NTE may not be recovered at all. 6. Examination of the cumulative traction energies of the buses
Fig. 19. The histogram for the recovery and feedback energies of the Ucap on the Campus-Return route.
Figs. 22 and 23 illustrate the cumulative energy lines for the BEE, PTE, and NTE and the speed and altitude profiles for the Campus-Return route. As these figures show, in the first 600 s, the NTE energies are very sensitive to the elevation profiles of the microtrips. During this period (the downhill driving period), the braking is so powerful that the absolute value of the braking energy increases steeply to control the speed of the bus. After that time,
480
S. Soylu / Applied Energy 135 (2014) 472–482 Table 4 Cumulative energies per km travel of the buses.
Fig. 22. Cumulative energies of the CB on the Campus-Return route.
Fig. 23. Cumulative energies of the HEB on the Campus-Return route.
both the PTEs and the absolute values of the NTEs gradually increase to their maximum values as a result of the frequent stop-and-go driving conditions. Fig. 22 shows that for the CB, the cumulative BEE, which is 14.92 kW h, is much higher than the cumulative PTE, which is 12.69 kW h. Fig. 23 shows that for the HEB, the difference between the two is not as large. In fact, it is normal to expect the BEE to be higher than the PTE owing to power train losses, although the energy recovered from braking covers most of the power train losses of the HEB. As Table 3 shows, although the cumulative PTE of the CB is 23% higher than that of the HEB, the cumulative BEE of the CB is 30% higher than that of the HEB. Comparison of the cumulative PTEs and BEEs for the buses can provide valuable information concerning the advantages of HEB technologies. However, as Table 2 shows, the driving distances and the maximum and mean speeds are slightly different for each bus in real-world testing. The traction, braking, and engine energies are known to be highly dependent on the elevation profile, speed profile, and travel distance. These dependencies must be neutralized to make the comparison of the cumulative PTEs and BEEs more meaningful. To neutralize
Energy (kW h/km)
Conventional bus
Hybrid electric bus
% Difference
PTE NTE BEE
1.15 1.45 1.35
1.01 1.33 1.07
12 8 21
the effect of the travel distance, the PTE and BEE were calculated in terms of energy per km of driving distance, as shown in Table 4. Although the PTE and NTE per km are 12% and 8% higher for the CB, respectively, the difference between the BEEs is much higher, at 21%. This means that the HEB use 21% less BEE for every km of the Campus-Return route. To neutralize the effect of speed differences, the cumulative BEE and NTE given in Table 4 were divided by their respective traction energies. The results, given in Table 5, indicate that the cumulative BEE and NTE normalized in this fashion are highly accurate comparison parameters. The BEEs are 1.18 and 1.06 times higher than their respective PTEs for the CB and HEB, respectively. As mentioned earlier, because of power train losses, it is normal to expect energy losses from the engine to the wheels. The losses for the HEB are mainly caused by the real-world performance of the generator, M/G, Ucap, and power electronics components of the power train. It is also well known from the literature that the combination of the M/G and its power electronics are approximately 80% efficient while operating under frequent stop-and-go conditions [30–36]. These losses seem to be mostly compensated for by the recovery of the braking energy of the HEB. A comparison of the cumulative PTEs and NTEs of the CB and HEB shows that the NTEs are 27% and 32% higher than the PTEs for the CB and HEB, respectively. This indicates clearly that braking energy is very significant under urban driving conditions and must not be wasted. Recovery of braking energy offers an excellent opportunity to work toward achieving the goals of the USDOE and EC to improve the fuel economy of vehicles in the future. Fig. 24 illustrates the effects of the speed and altitude profiles on the M/G and Ucap characteristics. This figure further clarifies
Table 5 The engine and braking energy per kW h of the traction energy. Energy ratio (kW h/km)/(kW h/km)
Conventional bus
Hybrid electric bus
BEE/PTE NTE/PTE
1.18 1.27
1.06 1.32
Table 3 Cumulative energies of the buses. Energy (kW h)
Conventional bus
Hybrid electric bus
% Difference
PTE NTE BEE
12.69 16.09 14.92
9.83 13.02 10.45
23 19 30
Fig. 24. Cumulative energies of power train components of the HEB on the Campus Return route.
S. Soylu / Applied Energy 135 (2014) 472–482
the effects of the speed and altitude profiles on the recovery and feedback energies for the Campus-Return route. As the figure shows, the M/G NTE is quite similar to that of the NTE, as shown in Fig. 23. The final values of the NTE and M/G NTE are 13.02 and 9.54 kW h, respectively. The M/G appears to provide 73% of the HEB braking energy. The difference between the M/G NTE and the NTE is accounted for by frictional braking and frictional losses from the wheels to the M/G. It would be ideal to provide all the braking energy needed by the HEB with the M/G and recover all of the braking energy, but as mentioned earlier, this would require Ucap and power electronics components of very high capacity, which may not be feasible because of weight and cost considerations. As Fig. 24 shows, during the time period from 600 to 1000 s, the Ucap recovery line deviates significantly from the M/G NTE line. This means that some of M/G NTE energy that was converted to electrical energy was not stored in the Ucap but rather was wasted as heat by an electrical resistor installed on the bus. This time period corresponds to the downhill portion of the route, as shown in Fig. 12. During this period, the altitude decreases steeply, which provides a great opportunity for energy recovery. However, the Ucap recovery trace cannot follow the M/ G NTE trace closely during this period, for the aforementioned reasons. After completing the downhill portion of the route, the Ucap starts to recover the M/G braking energy again. Consequently, the cumulative Ucap energy recovered was 6.83 kW h. This means that only 52% of the HEB braking energy was recovered and stored in the Ucap. This is actually the energy recovery efficiency of the regenerative braking system from the wheels to Ucap for the Campus-Return route. As a result, the round-trip efficiency (from the wheels to Ucap and from the Ucap to wheels) of the regenerative braking was determined to be 27% by squaring the 52%.
7. Conclusions The effects of the elevation and speed profiles of the CampusReturn route of the Sakarya Municipality on the PTE, NTE, BEE, and energy efficiency of a CB and an HEB were examined in this study. The following conclusions can be drawn from the results obtained: Because a hybrid city bus does not necessarily have a mechanical link from its engine to its wheels, there is an opportunity to operate its engine only at higher loads and efficiencies. However, it was found in this study that the limited capacity of the auxiliary power source prevents the engine from operating independently of the traction energy demanded. The operation of the engine still depends primarily on the speed and elevation profiles of the micro-trips. The NTEs of the CB and HEB were 27% and 32% higher than their respective PTEs because of the long downhill portion of the Campus-Return route. This clearly indicates that braking energy is very significant under urban driving conditions and must not be wasted. Recovery of braking energy offers an excellent opportunity to work toward achieving the goals of the USDOE and EC to improve the fuel economy of vehicles in the future. Although the PTE and NTE per km were 12% and 8% lower, respectively, for the HEB than for the CB, the BEE per km of the HEB was 21% lower than for the CB. Moreover, although the BEE of the CB was 1.18 times higher than its respective PTE, it was only 1.06 times higher for the HEB. It is not surprising that the BEE should be higher than the PTE, because of power train losses, but in the case of the HEB, the energy recovered by the regenerative braking system covers most of the losses of the power train. A frequency histogram of the NTEs is crucial for optimizing the size and capacity of the HEB power train components because it indicates the frequencies and magnitudes of the braking energies
481
that can be recovered with a regenerative braking system. It was found in this study that the micro-trip NTEs and PTEs of the buses ranged from 2.2 kW h to 1.7 kW h, but most of the micro-trip NTEs and PTEs were confined within a narrower range from 0.5 to 0.5 kW h. The Ucap recovery energy was found to be 6.83 kW h for the Campus-Return route. This corresponds to 52% of the cumulative NTE of the HEB. This finding reflects the energy recovery efficiency of the regenerative braking system for the Campus-Return route. Consequently, the round-trip efficiency of the regenerative braking system was calculated to be 27%. Because of the limited number of measurements obtained and test vehicles used, these conclusions cannot be generalized. Depending on the characteristics of bus routes and energy management strategies, hybrid city buses can achieve different energy efficiencies on different routes. However, these results make significant contributions to better understanding the benefits and barriers of HEBs. Acknowledgements The author thanks to the Ministry of Science, Technology and Industry for their financial support, TEMSA R&D for their technical and financial support, Ali Fuat Iskender for technical assistance, Necdet Ayaz for driving assistance and Muammer Soylu for Autocad drawings of the schematics. References [1] White Paper 2011. Roadmap to a single European transport area – towards a competitive and resource efficient transport system. EC staff working paper. SEC (2011) 358 final, Brussels; 28 March 2011. [2] Perujo A, Christian Thiel C, Nemry F. Electric vehicles in an urban context: environmental benefits and techno-economic barriers. In: Soylu Seref, editor. Electric vehicles – the benefits and barriers; 2011. p. 1–18 [chapter 1]. http:// dx.doi.org/10.5772/20760. ISBN 978-953-307-287-6. [3] Davis PB. Overview of the U.S. DOE vehicle technologies program. Office of the Vehicle Technologies Program Energy Efficiency and Renewable Energy. U.S. Department of Energy, DOE-ACE-2011AR, USA; 2012. [4] VTP. Multi year program plan, 2011–2015. US Department of Energy, Energy Efficiency and Renewable Energy, Vehicle Technologies Program; December 2010. [5] Al-Alawi BM, Bradley TH. Analysis of corporate average fuel economy regulation compliance scenarios inclusive of plug in hybrid vehicles. Appl Energy 2014;113:1323–37. [6] Millo F, Rolando L, Fuso R, Mallamo F. Real CO2 emissions benefits and end user’s operating costs of a plug-in hybrid electric vehicle. Appl Energy 2014;114:563–71. [7] Andress D, Nguyen TD, Das S. Reducing GHG emissions in the United States’ transportation sector. Energy Sustain Dev 2011;15:117–36. [8] Jimenez-Espadafor et al. Infantry mobility hybrid electric vehicle performance analysis and design. Appl Energy 2011;88:2641–52. [9] Singh G. FY progress report for advanced combustion engine research and developments. Energy Efficiency and Renewable Energy Vehicle Technologies Program, DOE-ACE-2011AR, USA; 2011. [10] Lucena SE. In: Soylu Seref, editor. A survey on electric and hybrid electric vehicle technology, electric vehicles – the benefits and barriers; 2011. p. 1–18 [chapter 1]. ISBN 978-953-307-287-6. [11] Soylu S. Examination of PN emissions and size distributions of a hybrid city bus under real world urban driving conditions. Int J Automot Technol 2014;15(3):369–76. [12] Erlandsson L, Almen J, Johansson H. Measurement of emissions from heavy duty vehicles meeting Euro IV/V emission levels by using on-board measurement in real life operation. In: 16th International symposium ‘‘transport and air pollution’’, Graz; 2008. [13] Cocker DR, Shah SD, Johnson K, Miller JW, Norbeck JM. Development and application of a mobile laboratory for measuring emissions from diesel engines. 1. Regulated gaseous emissions. Environ Sci Technol 2004;38:2182–9. [14] Johnson KC, Durbin TD, Cocker DR, Miller WJ, Bishnu JK, Maldonado H, et al. On-road comparison of a portable emission measurement system with a mobile reference laboratory for a heavy-duty diesel vehicle. Atmos Environ 2009;43:2877–83. [15] Durbin TD, Johnson K, Cocker DR, Miller JW. Evaluation and comparison of portable emissions measurement systems and federal reference methods for emissions from a back-up generator and a diesel truck operated on a chassis dynamometer. Environ Sci Technol 2007;41:6199–204.
482
S. Soylu / Applied Energy 135 (2014) 472–482
[16] Zhihua L, Ge Y, Johnson KC, Shah AN, Tan J, Wang C, et al. Real-world operation conditions and on-road emissions of Beijing diesel buses measured by using portable emission measurement system and electric low-pressure impactor. Sci Total Environ 2011;409:1476–80. [17] Butler KL, Ehsani M, Kamath PA. Matlab-based modeling and simulation package for electric and hybrid electric vehicle design. IEEE Trans Veh Technol 1999;48:1770–8. [18] Schaltz E. In: Soylu Seref, editor. Electrical vehicle design and modeling, electric vehicles –modelling and simulations; 2011. p. 1–25. http://dx.doi.org/ 10.5772/20271. ISBN 978-953-307-477-1. [19] Ganley JC. Design and testing of a series hybrid vehicle with an ultracapacitor energy buffer. Proc Inst Mech Eng D J Automob Eng 2012;226:869–80. [20] Jaber K, Fakhfakh A, Neji R. Modeling and simulation of high performance electrical vehicle powertrains in VHDL-AMS. In: Soylu Seref, editor. Electric vehicles –modelling and simulations. Intech Open Access Publisher; 2011. p. 1–25. http://dx.doi.org/10.5772/16650. ISBN 978-953-307-477-1. [21] Barrero R, Coosemans T, Van Mierlo J. Hybrid buses: defining the power flow management strategy and energy storage system needs. World Electr Veh J 2009;3:2032–6653. [22] Xiong W, Zhang Y, Yin C. Optimal energy management for a series–parallel hybrid electric bus. Energy Convers Manage 2009;50:1730–8. [23] Ahn JK, Jung KH, Kim DH, Jin HB, Kim HS, Hwang SH. Analysis of a regenerative braking system for hybrid electric vehicles using an electro-mechanical brake. Int J Automot Technol 2009;10:229–34. [24] EC Regulation. No 595/2009 of the European Parliament and of the Council. Official Journal of the European Union; 18 June 2009. [25] Soylu S, Bal A, Semercioglu H, Ay EF. Examination of an urban city bus operating conditions and emissions. In: Soylu Seref, editor. Urban traffic and hybrid vehicles; 2010. p. 1–12. http://dx.doi.org/10.5772/10176. ISBN 978953-307-100-8. [26] Bal A, Semercioglu H, Soylu S, Ay EF. Sehir otobüslerinin gerçek dünya kosßullarinda motor calisßma ßsartlarinin ve NOx emisyonlarinin incelenmesi,
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
Otekon 2010, 5. Otomotiv Teknolojileri Kongresi, 7–8 Haziran 2010, Bursa, Turkey; 2010. Semercioglu H, Soylu S, Bal A. Examination of real world operating conditions and emissions of a hybrid city bus. In: ICAT’10 international conference on energy and automotive technologies, Istanbul; November 5, 2010. Jimenez-Palacios JL. Understanding and quantifying motor vehicle emissions with vehicle specific power and TILDAS remote sensing. Ph.D. thesis. Massachusetts Institute of Technology, Cambridge, MA, USA; 1999. MOVES2010. Highway vehicle population and activity data. Assessment and standards division, office of transportation and air quality. USA: US Environmental Protection Agency; 2010. Baisden AC, Emadi A. ADVISOR-based model of a battery and an ultracapacitor energy source for hybrid electric vehicles. IEEE Trans Veh Technol 2004;53(1). Grbovic PJ, Delarue P, Le Moigne P. Selection and design of ultra-capacitor modules for power conversion applications: from theory to practice. In: 2012 IEEE 7th international power electronics and motion control conference – ECCE Asia, Harbin, China; June 2–5, 2012. Miller JM. Power electronics in hybrid electric vehicle applications. In: Applied power electronics conference and exposition, 2003. APEC ‘03. Eighteenth annual IEEE; 2003. Mi C. Topology, design, analysis and thermal management of power electronics for hybrid electric vehicle applications. Int J Electr Hybrid Veh 2008;1(3):2008. O’Keefe MP, Vertin K. An analysis of hybrid electric propulsion systems for transit buses. Milestone completion report, NREL/MP-540-32858, USA; October 2002. Williamson SS, Lukic SM, Emadi A. Comprehensive drive train efficiency analysis of hybrid electric and fuel cell vehicles based on motor-controller efficiency modeling. IEEE Trans Power Electr 2006;21(3). Williamson SS, Emadi A, Rajashekara K. Comprehensive efficiency modeling of electric traction motor drives for hybrid electric vehicle propulsion applications. IEEE Trans Veh Technol 2007;56(4).
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
The effects of urban driving conditions on the operating characteristics of conventional and hybrid electric city buses Seref Soylu ⇑ Sakarya University, Faculty of Engineering, 54187 Sakarya, Turkey
h i g h l i g h t s Operating characteristics of conventional and hybrid electric buses were examined. Recovery of braking energy offers an excellent opportunity to improve fuel economy. Speed and altitude profiles of routes have dramatic impacts on the energy recovery. Capacity of the auxiliary power source has a dramatic impact on the energy recovery. Round-trip efficiency of the regenerative braking system was calculated to be 27%.
a r t i c l e
i n f o
Article history: Received 29 May 2014 Received in revised form 25 August 2014 Accepted 27 August 2014
Keywords: Hybrid bus Regenerative braking Real-world driving Energy efficiency Traction energy Braking energy
a b s t r a c t The basic operating characteristics of a conventional bus (CB) and a hybrid electric bus (HEB) were examined under urban driving conditions. To perform this examination, real-time operating data from the buses were collected on the Campus-Return route of the Sakarya Municipality. The main characteristics examined were the traction, braking, engine, engine generator unit (EGU), motor/generator (M/G), and ultracapacitor (Ucap) energies and efficiencies of the buses. The route elevation profile and the frequency of stop-and-go operations of the buses were found to have dramatic impacts on the braking and traction energies of the buses. The declining profile of the Campus-Return route provided an excellent opportunity for energy recovery by the regenerative braking system of the HEB. However, owing to the limits on the capacities and efficiencies of the hybrid drive train components and the Ucap, the bus braking energies were not recovered completely. Braking energies as high as 2.2 kW h per micro-trip were observed, but less than 1 kW h of braking energy per micro-trip was converted to electricity by the M/G; the rest of the braking energy was wasted in frictional braking. The maximum energy recovered and stored in the Ucap per micro-trip was 0.5 kW h, but the amount of energy recovered and stored per micro-trip was typically less than 0.2 kW h for the entire route. The cumulative braking energy recovered and stored in the Ucap for the Campus-Return route was 52% of the available brake energy, which was 13.02 kW h. Consequently, the round-trip efficiency of the regenerative braking system, between the wheels and Ucap, was determined to be 27%. Finally, although the brake engine energy (BEE) of the CB was 1.18 times higher than its positive traction energy (PTE), the BEE of the HEB was only 1.07 times higher than its PTE. In fact, it is normal to expect the BEE to be higher than the PTE owing to power train losses, but the energy recovered by the regenerative braking system was found to cover most of the power train losses and even improve the energy efficiency of the HEB. Ó 2014 Elsevier Ltd. All rights reserved.
Abbreviations: CB, conventional bus; HEB, hybrid electric bus; EGU, enginegenerator unit; M/G, motor/generator; Ucap, ultracapacitor; PE, power electronics; Gen, generator; FE, fuel energy; BEE, brake engine energy; GE, generator energy; PTE, positive traction energy; NTE, negative traction energy; EC, European Commission; USDOE, US Department of Energy; CO2, carbon dioxide; GHG, greenhouse gases; VTP, vehicle technology program; PEMS, portable emission measurement system; GPS, global positioning system. ⇑ Tel.: +90 5359788982. E-mail addresses: [email protected], [email protected] http://dx.doi.org/10.1016/j.apenergy.2014.08.102 0306-2619/Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Both the European Commission (EC) and the United States Department of Energy (USDOE) are taking serious actions to drastically reduce CO2 emissions, with the goal of keeping climate change below 2 °C. The EC has suggested that to reach that goal,
S. Soylu / Applied Energy 135 (2014) 472–482
the transport sector needs to reduce its greenhouse gas (GHG) emissions by at least 60% by 2050, compared to the levels of 1990 [1,2]. Similarly, the USDOE is taking actions to reduce CO2 emissions through its vehicle technology program (VTP), in partnership with automotive industry leaders, national laboratories, and universities [3,4]. The main goals of the VTP are to enable the US to significantly reduce fossil fuel consumption, GHG emissions, and local emissions. GHG emissions are expected to be reduced from 2005 levels by over 40% by 2030 and by over 80% by 2050. Improving the energy efficiency of vehicle power trains by hybridizing them appears to be one of the most promising and cost-effective approaches to achieving these goals [5–8]. It has been reported in literature that more than 30% fuel saving can be expected from commercial vehicles when coupled with hybrid electric power trains [9,10]. The energy efficiency and emissions of city buses operating under urban driving conditions are highly dependent on their operating conditions [11–16]. If operated at cruising conditions on flat routes without auxiliaries (e.g., heating and air conditioning), buses need only enough energy to compensate for rolling and aerodynamic losses, but real-world operations of buses are far from such idealized conditions. Depending on the traffic and road conditions, which may require many short micro-trips with accelerations, decelerations and various road grades, braking energy losses and auxiliaries can increase energy consumption dramatically and reduce fuel economy. Electrical hybridization of city buses, on the other hand, can be a remedy for excessive fuel consumption under urban driving conditions. In comparison to CBs, HEBs can easily save fuel and reduce emissions, for two main reasons. First, the vehicle kinetic energy, which is normally wasted during braking in the case of CBs, can be recovered and stored in the form of electricity during braking in the case of HEBs. Second, since HEBs do not necessarily have mechanical links from their engines to their wheels, their engines can always operate in the optimum region of the fuel consumption map. Therefore, they have very high potential to minimize both fuel consumption and emissions. Computer modeling and simulation programs are highly effective and economical tools for use in examining the effects of design alternatives and energy management strategies on hybrid vehicles before construction of a prototype begins [17–20]. Barrero et al. [21] simulated and compared several power flow management strategies for hybrid city buses using a quasi-static ‘‘backward/forward-looking’’ simulation program. The simulation results indicated that the energy savings that can be achieved are in the range of 32.6% (when using the kinetic strategy and a 0.3-kW h energy storage system) to 40% (when using the ICE on–off strategy and a 0.65-kW h energy storage system). Xiong et al. [22] developed an energy management strategy for a series–parallel hybrid city bus using a forward-facing simulation program based on the Matlab/Simulink software. The simulation results indicated that the engine operation can be maintained in the high-efficiency range and that a theoretical improvement in fuel economy 30.3% can be achieved, compared to that of a conventional bus, over the driving cycle of a transit bus. Ahn et al. [23] simulated a regenerative braking system in a hybrid electric vehicle under various driving conditions using Matlab/Simulink and observed that hybrid electric vehicles with regenerative braking can improve fuel economy by 20–50%. All these modeling efforts have indicated that electrical hybridization of city buses provides various degrees of benefits in terms of energy efficiency and emissions. In order to confirm and quantify the benefits under real-world urban driving conditions, a research project entitled ‘‘Measurement and Modeling of Hybrid City Bus Emissions under Real-World Operating Conditions’’ was initiated at Sakarya University, with the support of the Turkish
473
Ministry of Science Technology and Industry and TEMSA R&D, which is the research and development department of a local bus manufacturer. Confirmation of vehicle emissions levels under real world driving conditions is a requirement by EC regulation [24]. In the first phase of the project, real-time data on the basic operating characteristics of a conventional 12-m city bus were collected under real-world urban driving conditions and analyzed. In the second phase, similar data were collected and analyzed for hybrid electric versions of the bus. Initial tests and reports on the project indicated that the energy efficiencies of the buses are highly dependent on the elevation and speed profiles of the bus routes [25–27]. In the present study, the benefits and barriers of HEBs were further clarified by examining the PTE, NTE, EGU, M/G, and Ucap energies and efficiencies of the HEB under real world urban driving conditions. The results are expected to be a reference for researchers who design and simulate the next generation of city buses.
2. Experimental setup The test vehicles were TEMSA-brand city buses. The buses were 12 m in length and weighed 15 tons when loaded. The CB was powered by a 6.7-liter Euro 4 CUMMINS diesel engine. It has a ZF 6 HP 504 C gear box and a ZF AV 132/80 rear axle. The HEB, which was designed based on body of the conventional bus, had a SIEMENS ELFA hybrid drivetrain system. The hybrid bus was powered by a 6.7-liter Euro 5 CUMMINS diesel engine. The Euro 4 and 5 engines have the same power and torque characteristics. Figs. 1a and 1b are schematic diagrams of the conventional and HEB drivetrains. The basic specifications of the ELFA drivetrain components and the engines are given in Table 1. There are mainly series and parallel power train configuration alternatives for HEBs. However, in order to increase braking energy recovery rate with regenerative braking, series hybrids are generally preferred for city bus applications due to their frequent stop and go operations. As the schematics illustrate, the HE drivetrain (Series Hybrid) does not have a mechanical link from the engine to the wheels. Instead, the engine drives the generator that feeds the M/G. Therefore, the engine can operate in the most efficient regions of its fuel consumption map. Additionally, the Ucap, which can be charged by regenerative braking or by the generator, feeds the M/G, especially during acceleration of the bus.
Fig. 1a. Schematics of the conventional drivetrain and its instrumentation (1. Computer, 2. GPS, 3. SEMTECH DS, 4. Engine).
474
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 1b. Schematics of the HE drivetrain and its instrumentations (PE: Power electronics, Gen: Generator).
The tests were carried out on the Campus-Return route of the Sakarya Municipality. Test data for the vehicle speed, location, engine operation, exhaust emissions, and environmental conditions were sampled every second using a SEMTECH DS manufactured by Sensor, Inc. The Semtech DS is a portable emissions measurement system (PEMS) that is capable of real-time monitoring of gaseous emissions and collection of environmental data such as the ambient pressure, temperature, and humidity. The Semtech DS also has a vehicle interface module (SAE-J1708) that collects engine and vehicle operation data. The data such as the speed, torque, power, and fuel consumption were directly collected from the engine ECU. Test data for the hybrid power train components such as generator power, M/G power, speeds, Ucap charge and discharge powers were collected using the SIADIS interface of the ELFA. In order to minimize adverse effects of temperature on the component performance of the hybrid power train, the component temperatures were kept in a limited range. In order to compute energies, the relevant power data were integrated with a MATLAB program for the entire route. 3. Route and operating characteristics of the buses The speed and elevation profiles of city buses on urban routes are very specific because of the traffic, the frequency of bus stops, and the road grade. Note that the road grade is the percent change in the route elevation per unit of horizontal distance. Even if the same bus is tested repeatedly on the same route with the same driver, the individual tests may have different speed profiles under real-world driving conditions. This is simply the nature of realworld driving. Fig. 2 shows speed and elevation profiles, along with acceleration histograms, for the Campus-Return route. The speed profiles involve many micro-trips, which are the trips between two complete stops. A typical micro-trip is shown in Fig. 3. As the profiles show, the buses experience highly frequent stopand-go events and dramatic elevation changes on this route. The micro-trips on this route typically have accelerations in the range of 2 to 2 m/s2. It should be noted that the elevations were divided by 5 for better presentation of the figure. To determine the real
value of the elevations it must be multiplied by 5. The similarities of the speed profiles can easily be compared with their basic characteristics, which are given in Table 2. Although the buses were driven over the same route, the measurement results differ slightly. A check of the dates of the tests shows that more than two years passed between the tests. During that period, the route was slightly modified because of construction activities. For this reason, the travel distances are slightly different, although the starting and ending points are the same. A comparison of the number of micro-trips and the average micro-trip maximum (Vmax_ave) and mean speeds (Vmean_ave) shows that the CB and HEB speed profiles have similar characteristics. The power consumptions of the buses for the auxiliaries, such as air conditioner, could be slightly different, but they were not measured. It was assumed that they don’t have significant impact on the brake engine energies of the buses. 4. Modeling of the bus traction power If operated at a constant speed over a flat road, a bus will run with the minimum fuel consumption per km travel while its engine providing traction force for the rolling and aerodynamic resistances, only:
F roll ¼ m g C roll
ð1Þ
F aero ¼ ð1=2Þ q C d Af v 2
ð2Þ
where Froll is the force required to overcome rolling resistance, m is the mass of the bus, g is the acceleration of gravity, Croll is the coefficient of the rolling resistance, Faero is the force required to overcome aerodynamic resistance, q is the density of air, Cd is the drag coefficient, Af is the frontal area of the bus, and v is the speed. However, as shown by the speed and altitude profiles in Fig. 2, a typical city bus operation under urban driving conditions involves frequent changes in speed, which is a measure of kinetic energy (KE), and changes in elevation, which is a measure of potential energy (PE). For these conditions, the bus engine must provide enough traction force to increase the KE and PE as well.
F ke ¼ m að1 þ eÞ
ð3Þ
F pe ¼ m g sin h
ð4Þ
where Fke is the force required to increase the KE, a is the bus acceleration, e is a factor for the rotational components of the bus power train, Fpe is the force required to increase the PE, and h is the angle of the road inclination. The traction force and power of the bus can be written as follows:
F traction ¼ F roll þ F aero þ F ke þ F pe
ð5Þ
Ptraction ¼ v F traction
ð6Þ
Integration of positive and negative traction powers over the test route provided the PTEs and NTEs, respectively. The integrations were performed with a MATLAB program. This is a vehicle only
Table 1 Specifications of ELFA drivetrain components.
Brand Rated power Rated torque Rated voltage Maximum speed Weight Capacity (Farad/kW h)
Generator
M/G
Ucap
Engine
Siemens, PM Synch 145 kW @ 4000 rpm 368 N m @ 220 A 700 V 5000 rpm 120 kg –
Siemens, AC Induction 85 kW 220 N m 650 V DC 10,000 rpm 120 kg –
ISE, Thunderpack 233 kW – 720 V DC – 180 kg 10.4/0.748
Cummins, ISB6.7 184 kW @ 2325 rpm 1020 N m @ 1200 rpm – – – –
S. Soylu / Applied Energy 135 (2014) 472–482
475
Fig. 2. Speed, elevation and acceleration profiles of the buses on the Campus-Return route.
5. Examination of the traction and braking energies of the buses
Fig. 3. A typical micro-trip from the Campus-Return route.
Table 2 Basic driving characteristics of the buses on the Campus-Return route.
Travel distance (km) Travel time (s) # of microtrips Vmax_avea (km/h) Vmean_ave.a (km/h) Test date
Conventional bus (CB)
Hybrid electric bus (HEB)
11.08 2086 40 35.48 20.80 11.06.2009
9.78 2214 35 35.16 21.23 19.09.2011
A city bus operating on an urban route must overcome resistance due to rolling, aerodynamic drag, acceleration, and elevation. The PTE needed to overcome these types of resistance is typically supplied by a diesel engine that converts chemical fuel energy (FE) into BEE. Fig. 4 compares the FE, BEE and PTE of the CB for every micro-trip on the Campus-Return route. As the figure shows, the FEs, BEEs and PTEs are highly dependent on the speed and elevation profiles of the micro-trips. As expected, the BEEs are slightly higher than the PTEs. The differences between them correspond to the frictional losses of the power train components. However, the differences between the FEs and BEEs are highly significant; the BEEs correspond to less than 1/3 of the FEs for most of the micro-trips. The ratio of the BEE to the FE is the brake engine efficiency (gbrake) for a micro-trip. As Fig. 5a shows, the gbrake is as high as 40% for the second micro-trip, which is the micro-trip with the highest traction energy, but the efficiency decreases to well below the maximum as the BEE decreases. It is well known from the literature that engine efficiency is maximized at higher engine loads and minimized at lower engine loads, mainly because of the increasing role of parasitic losses, which are pumping and frictional losses, at lower engine loads. As Fig. 5b shows, the brake-specific fuel consumption (bsfc) of the bus increases as the engine power decreases, which clearly indicates the effects of parasitic losses.
a Vmax and Vmean are calculated based on the micro-trip speed. They do not involve boarding time durations at the bus stops.
model and does not cover the components of entire power train system. Beside, since the objective of this work was not to model the coefficients of the rolling and aerodynamic drag of the buses, the coefficients were initially chosen based on Jimenez-Palacios [28] and MOVES2010 [29], but they were slightly modified later to fine-tune the model to better match the experimental data obtained from the tests. To fine-tune the rolling and drag coefficients, the magnitudes of the traction and engine power were compared at various speeds. In the comparison, the dependency of the third power of the aerodynamic resistance on the vehicle speed was particularly notable. The rolling resistance increases linearly with speed. After determining the final values of the coefficients, given below, the traction and braking powers of the CB and HEB were computed.
e ¼ 0:1; C roll ¼ 0:007; C d ¼ 0:72
Fig. 4. The FEs, BEEs and PTEs of the CB on the Campus-Return route.
476
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 5a. The BEEs and engine efficiencies of the CB on the Campus-Return route.
Fig. 7. Comparison of the PTEs and NTEs of the CB on the Campus-Return route.
Fig. 6. The PTEs and NTEs of the CB on the Campus-Return route.
Fig. 8. The FEs, GEs and PTEs of the HEB on the Campus-Return route.
bsfc, gr/kWh
Therefore, the gbrake for a typical city bus operating under urban driving conditions remains well below its maximum. Fig. 6 compares the positive and negative traction energies of the bus on the Campus-Return route. The NTE, which for CBs is the energy lost due to frictional braking, is the braking energy needed to decelerate the bus to a stop or to maintain a desired bus speed on a declining road. As Fig. 6 shows, the NTEs are not homogenous and are highly dependent on the speed and elevation profiles of the micro-trips. While the route declines sharply in the first 600 s, as was shown in Fig. 2, the absolute values of the corresponding NTEs are as high as 2.2 kW h, as shown in Fig. 6. After that time, the NTEs mostly reflect the speed characteristics of the micro-trips. The frequencies and magnitudes of the positive and negative traction energies can be better compared in a histogram, as shown in Fig. 7. The magnitudes of the NTEs are even greater than those of the PTEs because of the long downhill portion of the route. Moreover, although the NTEs and PTEs range from 2.2 kW h to 1.7 kW h, most of the values are confined to a narrower range from 0.5 to 0.5 kW h.
In summary, a CB operating on an urban route is wasting a significant amount of its FE, mainly because of parasitic and frictional braking losses. However, as mentioned earlier, hybridization of conventional buses can minimize these losses, for two reasons. First, HEBs do not necessarily have mechanical links from their engines to the wheels; therefore, their engine may operate at their most efficient conditions and minimize parasitic losses. Second, HEBs generally have regenerative braking systems with auxiliary power sources; therefore, frictional braking losses can easily be minimized as well. Fig. 8 compares the FE, generator energy (GE), and PTE of the HEB. As explained in the Experimental Setup section, there is no mechanical link between the diesel engine and the wheels in the HEB. Instead, the diesel engine drives the generator to generate electricity and feeds the M/G. Therefore, as mentioned in the preceding paragraph, there is an opportunity to operate the engine generator unit (EGU) only at higher loads and efficiencies. However, as Fig. 8 shows, the GE is directly related to the PTE; it still generates electricity primarily based on the speed and elevation profiles of the micro-trips. Actually, the limited capacity of the auxiliary power source prevents the EGU from operating at its most efficient operating conditions. Moreover, a comparison of Figs. 9 and 5a shows that the EGU efficiencies, which are the ratios of the GEs to the FEs, are generally less than the engine efficiency of the CB. Although the engine operating characteristics are the same for both of the buses, the EGU efficiency is lower primarily because of the efficiency factor of the generator. As Fig. 10 shows, the generator efficiency can be as high as 96%, but it is highly dependent on the torque and speed. The generator gear box is another factor that reduces the EGU efficiency. Fig. 8 also shows
1000 800 600 400 200 0
bsfc gr/kWh
0
20
40
60
80
100
120
140
160
180
200
Engine Power, kW Fig. 5b. Bsfc of the CB on the Campus-Return route.
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 9. The GEs and EGU efficiencies of the HEB on the Campus-Return route.
477
Fig. 11. The PTEs and NTEs of the HEB on the Campus Return route.
Fig. 10. The generator efficiency map and the torque data points. Fig. 12. Comparison of the PTEs and NTEs of the HEB on the Campus-Return route.
that the PTEs are generally higher than the GEs, even though the reverse would be expected because of the losses of the power train components. This means that the auxiliary power source is also contributing a significant amount of energy to driving the HEB. The auxiliary power source of the HEB is a 0.74 kW h Ucap that is charged mainly by the regenerative braking system. It is well known from the first law of thermodynamics that energy cannot be destroyed but can be converted into other forms of energy. In this sense, the PTE required to increase the kinetic and potential energies of a bus can be recovered during braking with a regenerative braking system. As mentioned earlier, this is one of the most important advantages of HEBs: they have the capability to convert braking energy into electrical energy and store it in auxiliary power sources, which can be a battery, a Ucap or a combination of the two. While batteries are preferred for their higher energy capacity, Ucaps are preferred for their higher specific powers, which can provide large bursts of power within a few seconds. Fig. 11 compares the PTEs and NTEs of the HEB on the CampusReturn route. As shown in Table 2, the averaged maximum and mean speeds of the HEB are slightly different from those of the CB. However, as with the CB, the absolute values of the NTEs during the first 10 micro-trips increase to as high as 2.2 kW h, and the rest of the traction energies directly reflect the speed profiles of the micro-trips. The magnitudes and frequencies of the traction energies can be better compared in a histogram. As Fig. 12 shows, with the HEB as with the CB, the absolute values of both the PTEs and NTEs are typically less than 0.5 kW h, and the absolute values of the NTEs are typically greater than those of the PTEs. This histogram compares the frequencies and magnitudes of the NTEs recovered by the regenerative braking system. If the braking energies were recovered with 100% efficiency, the bus would only need
energy to overcome rolling resistance and aerodynamic drag. A bus operating only with aerodynamic drag and rolling resistance under urban driving conditions would be highly fuel efficient. However, the real-world operations of buses are far from such idealized conditions. Optimization of the capacity and size of the auxiliary power source of a HEB is very important because of the regeneration efficiency, mass, and cost considerations involved. A histogram such as that shown in Fig. 12 is very useful in the preliminary sizing of the auxiliary power system for a HEB. Because Ucaps are massy and expensive, they must be optimized based on the operating conditions of a city bus to achieve a favorable cost/benefit ratio. As the preceding figures show, the magnitudes and frequencies of the NTEs are not constant, and they are highly dependent on the characteristics of the bus route. The NTEs shown in Fig. 12 indicate that if a Ucap with 2.0 kW h of effective capacity were chosen for braking energy recovery, almost 100% of the NTEs of the Campus-Return route would be recovered by the regenerative braking system, but the Ucap would be very heavy and expensive. If instead a Ucap with an effective capacity of 0.5 kW h were chosen, approximately 75% of the NTE would be recovered, and the rest would be lost owing to mechanical braking. However, during the recovery of the NTE, the energy must be transferred from the wheels to the Ucap over the power train components, which are mainly the M/ G and final drive components. Unfortunately, all power train components have efficiency characteristics for energy conversion and/ or transfer. Fig. 13 compares the M/G PTEs and M/G NTEs of the HEB on the Campus-Return route. The M/G PTE reflects the work that the M/G does as an electrical motor to convert electrical energy into
478
S. Soylu / Applied Energy 135 (2014) 472–482
Fig. 13. The GEs and M/G traction energies of the HEB on the Campus-Return route.
mechanical energy. The M/G NTE reflects the work that the M/G does as a generator to convert mechanical energy into electrical energy. The M/G NTE is actually the bus braking energy that is converted into electricity by the M/G. As expected, the trends in the M/ G PTEs and M/G NTEs are similar to those shown in Fig. 11, but the absolute values of the micro-trip traction energies are significantly lower for the NTEs and significantly higher for the PTEs than those shown in Fig. 11. This means that the M/G PTEs are higher than the bus PTEs, to overcome the losses in the M/G and final drive components. Similarly, in the conversion of the bus NTEs to M/G NTEs, the losses of the final drive components and the M/G must be overcome. For these reasons, the M/G NTEs are much smaller than the bus NTEs. It should be noted that the M/G PTEs are greater than the GEs for the entire route. As mentioned earlier, because the auxiliary power source contributes to the M/G PTEs, they are greater than the GEs. The energy conversion efficiency of the M/G can be better explained using Figs. 14 and 15. Fig. 14 shows the torque and speed data points of the M/G on the Campus-Return route. As expected, the operating torques and speeds are widely scattered, which reflects the characteristics of the traction power demanded. Unfortunately, however, the M/G conversion efficiency is not constant across the operating map. As Fig. 15 shows, the efficiency of the M/G increases from 70% at lower speeds to more than 90% at higher speeds, which makes the M/G efficiency one of the most important factors for HEBs. The magnitudes and frequencies of the M/G PTEs and M/G NTEs can be better compared with a frequency histogram. As Fig. 16 shows, unlike the PTEs and NTEs shown in Fig. 12, the absolute values of the M/G NTEs are much smaller than those of the M/G PTEs.
Fig. 14. The speed and torque map of the M/G on the Campus-Return route.
Fig. 15. The M/G efficiency map and the torque data points on the Campus-Return route.
Fig. 16. The PTE and NTE histograms for the M/G on the Campus Return route.
Moreover, the absolute value of the maximum M/G NTE is less than 1 kW h. This means that some of the micro-trip braking energies (especially high-magnitude braking energies) were probably wasted by frictional braking as an energy management strategy, because of size and capacity considerations. Furthermore, as Fig. 17 shows, the absolute values of the M/G NTEs are significantly less than those of the bus NTEs, which clearly indicates the importance of the energy transfer and conversion efficiencies of the M/G and the final drive, as well as the energy management strategy of the HEB.
Fig. 17. Comparison of the magnitude and frequencies of the NTEs and M/G NTEs.
S. Soylu / Applied Energy 135 (2014) 472–482
The importance of the Ucap capacity and the energy management strategy are examined further in Figs. 18 and 19. The Ucap recovery energy is the M/G NTE to be converted to direct current and stored in the Ucap. The stored Ucap energy can optionally be used to contribute to the GE as feedback energy to feed the M/G, especially during bus accelerations. It is also important to note that, as the figures show, although the capacity of the Ucap is 0.74 kW h, the absolute values of the recovery and feedback energies typically remain less than 0.6 kW h. The main reason that the Ucap capacity is not used completely is that the Ucap voltage decreases as it is discharged, so it cannot feed the M/G. Therefore, the Ucap never becomes completely empty. As Fig. 19 shows, the frequencies of the higher-magnitude recovery and feedback energies are very low. This means that the Ucap capacity was not used effectively for the entire route. An important reason for the Ucap capacity not being used effectively is that if the state of charge (SOC) of the Ucap is at its maximum, as shown in Fig. 20, the Ucap cannot be charged any more. Once the SOC reaches its maximum, all available NTEs must be wasted because of the limited capacity of the Ucap. As Fig. 20 shows, after the 5th micro-trip, the SOC is at more than 90% and remains at that level until the 12th micro-trip. This period is the downhill driving period of the bus, as shown in Fig. 2, and there is a significant amount of NTE to be recovered, as shown in Figs. 11 and 13. However, at the beginning of this period, the Ucap SOC almost reaches its maximum, and during this period, both the feedback and recovery energy remain at very low levels of less than 0.2 kW h. Because the Ucap capacity is limited, there is no way to completely recover
479
Fig. 20. The SOC of the Ucap on the Campus-Return route.
Fig. 21. Comparison of the magnitudes and frequencies of the NTE and Ucap recovery energies.
Fig. 18. The recovery and feedback energies of the Ucap on the Campus-Return route.
the NTEs, which have higher magnitudes. Therefore, most of the NTEs available for the recovery are wasted during this period by frictional braking. On the other hand, as Figs. 13 and 18 show, most of the micro-trips on this route do not have such high NTEs to be recovered, and therefore, as can be seen from Fig. 20, the SOC generally remains at lower levels. As Fig. 21 shows, although the absolute magnitudes of the NTEs are as high as 2.2 kW h, the Ucap recovery energies are less than 0.5 kW h, mainly because of the limited capacity of the Ucap. If a higher-capacity Ucap were chosen, it would not be used effectively for most of the route, as the frequencies of the NTEs show. Optimizing the capacity of a Ucap for an urban route and developing an energy management system for effective use of the Ucap capacity are therefore challenging problems whose solutions depend on the speed and altitude profiles of the urban route, which may not be known beforehand. However, the main principle for effective use of Ucap capacity is that the SOC must be at the lowest level possible before braking so that there is enough capacity for the regenerated braking energy. If the SOC is at a very high level before braking, the NTE may not be recovered at all. 6. Examination of the cumulative traction energies of the buses
Fig. 19. The histogram for the recovery and feedback energies of the Ucap on the Campus-Return route.
Figs. 22 and 23 illustrate the cumulative energy lines for the BEE, PTE, and NTE and the speed and altitude profiles for the Campus-Return route. As these figures show, in the first 600 s, the NTE energies are very sensitive to the elevation profiles of the microtrips. During this period (the downhill driving period), the braking is so powerful that the absolute value of the braking energy increases steeply to control the speed of the bus. After that time,
480
S. Soylu / Applied Energy 135 (2014) 472–482 Table 4 Cumulative energies per km travel of the buses.
Fig. 22. Cumulative energies of the CB on the Campus-Return route.
Fig. 23. Cumulative energies of the HEB on the Campus-Return route.
both the PTEs and the absolute values of the NTEs gradually increase to their maximum values as a result of the frequent stop-and-go driving conditions. Fig. 22 shows that for the CB, the cumulative BEE, which is 14.92 kW h, is much higher than the cumulative PTE, which is 12.69 kW h. Fig. 23 shows that for the HEB, the difference between the two is not as large. In fact, it is normal to expect the BEE to be higher than the PTE owing to power train losses, although the energy recovered from braking covers most of the power train losses of the HEB. As Table 3 shows, although the cumulative PTE of the CB is 23% higher than that of the HEB, the cumulative BEE of the CB is 30% higher than that of the HEB. Comparison of the cumulative PTEs and BEEs for the buses can provide valuable information concerning the advantages of HEB technologies. However, as Table 2 shows, the driving distances and the maximum and mean speeds are slightly different for each bus in real-world testing. The traction, braking, and engine energies are known to be highly dependent on the elevation profile, speed profile, and travel distance. These dependencies must be neutralized to make the comparison of the cumulative PTEs and BEEs more meaningful. To neutralize
Energy (kW h/km)
Conventional bus
Hybrid electric bus
% Difference
PTE NTE BEE
1.15 1.45 1.35
1.01 1.33 1.07
12 8 21
the effect of the travel distance, the PTE and BEE were calculated in terms of energy per km of driving distance, as shown in Table 4. Although the PTE and NTE per km are 12% and 8% higher for the CB, respectively, the difference between the BEEs is much higher, at 21%. This means that the HEB use 21% less BEE for every km of the Campus-Return route. To neutralize the effect of speed differences, the cumulative BEE and NTE given in Table 4 were divided by their respective traction energies. The results, given in Table 5, indicate that the cumulative BEE and NTE normalized in this fashion are highly accurate comparison parameters. The BEEs are 1.18 and 1.06 times higher than their respective PTEs for the CB and HEB, respectively. As mentioned earlier, because of power train losses, it is normal to expect energy losses from the engine to the wheels. The losses for the HEB are mainly caused by the real-world performance of the generator, M/G, Ucap, and power electronics components of the power train. It is also well known from the literature that the combination of the M/G and its power electronics are approximately 80% efficient while operating under frequent stop-and-go conditions [30–36]. These losses seem to be mostly compensated for by the recovery of the braking energy of the HEB. A comparison of the cumulative PTEs and NTEs of the CB and HEB shows that the NTEs are 27% and 32% higher than the PTEs for the CB and HEB, respectively. This indicates clearly that braking energy is very significant under urban driving conditions and must not be wasted. Recovery of braking energy offers an excellent opportunity to work toward achieving the goals of the USDOE and EC to improve the fuel economy of vehicles in the future. Fig. 24 illustrates the effects of the speed and altitude profiles on the M/G and Ucap characteristics. This figure further clarifies
Table 5 The engine and braking energy per kW h of the traction energy. Energy ratio (kW h/km)/(kW h/km)
Conventional bus
Hybrid electric bus
BEE/PTE NTE/PTE
1.18 1.27
1.06 1.32
Table 3 Cumulative energies of the buses. Energy (kW h)
Conventional bus
Hybrid electric bus
% Difference
PTE NTE BEE
12.69 16.09 14.92
9.83 13.02 10.45
23 19 30
Fig. 24. Cumulative energies of power train components of the HEB on the Campus Return route.
S. Soylu / Applied Energy 135 (2014) 472–482
the effects of the speed and altitude profiles on the recovery and feedback energies for the Campus-Return route. As the figure shows, the M/G NTE is quite similar to that of the NTE, as shown in Fig. 23. The final values of the NTE and M/G NTE are 13.02 and 9.54 kW h, respectively. The M/G appears to provide 73% of the HEB braking energy. The difference between the M/G NTE and the NTE is accounted for by frictional braking and frictional losses from the wheels to the M/G. It would be ideal to provide all the braking energy needed by the HEB with the M/G and recover all of the braking energy, but as mentioned earlier, this would require Ucap and power electronics components of very high capacity, which may not be feasible because of weight and cost considerations. As Fig. 24 shows, during the time period from 600 to 1000 s, the Ucap recovery line deviates significantly from the M/G NTE line. This means that some of M/G NTE energy that was converted to electrical energy was not stored in the Ucap but rather was wasted as heat by an electrical resistor installed on the bus. This time period corresponds to the downhill portion of the route, as shown in Fig. 12. During this period, the altitude decreases steeply, which provides a great opportunity for energy recovery. However, the Ucap recovery trace cannot follow the M/ G NTE trace closely during this period, for the aforementioned reasons. After completing the downhill portion of the route, the Ucap starts to recover the M/G braking energy again. Consequently, the cumulative Ucap energy recovered was 6.83 kW h. This means that only 52% of the HEB braking energy was recovered and stored in the Ucap. This is actually the energy recovery efficiency of the regenerative braking system from the wheels to Ucap for the Campus-Return route. As a result, the round-trip efficiency (from the wheels to Ucap and from the Ucap to wheels) of the regenerative braking was determined to be 27% by squaring the 52%.
7. Conclusions The effects of the elevation and speed profiles of the CampusReturn route of the Sakarya Municipality on the PTE, NTE, BEE, and energy efficiency of a CB and an HEB were examined in this study. The following conclusions can be drawn from the results obtained: Because a hybrid city bus does not necessarily have a mechanical link from its engine to its wheels, there is an opportunity to operate its engine only at higher loads and efficiencies. However, it was found in this study that the limited capacity of the auxiliary power source prevents the engine from operating independently of the traction energy demanded. The operation of the engine still depends primarily on the speed and elevation profiles of the micro-trips. The NTEs of the CB and HEB were 27% and 32% higher than their respective PTEs because of the long downhill portion of the Campus-Return route. This clearly indicates that braking energy is very significant under urban driving conditions and must not be wasted. Recovery of braking energy offers an excellent opportunity to work toward achieving the goals of the USDOE and EC to improve the fuel economy of vehicles in the future. Although the PTE and NTE per km were 12% and 8% lower, respectively, for the HEB than for the CB, the BEE per km of the HEB was 21% lower than for the CB. Moreover, although the BEE of the CB was 1.18 times higher than its respective PTE, it was only 1.06 times higher for the HEB. It is not surprising that the BEE should be higher than the PTE, because of power train losses, but in the case of the HEB, the energy recovered by the regenerative braking system covers most of the losses of the power train. A frequency histogram of the NTEs is crucial for optimizing the size and capacity of the HEB power train components because it indicates the frequencies and magnitudes of the braking energies
481
that can be recovered with a regenerative braking system. It was found in this study that the micro-trip NTEs and PTEs of the buses ranged from 2.2 kW h to 1.7 kW h, but most of the micro-trip NTEs and PTEs were confined within a narrower range from 0.5 to 0.5 kW h. The Ucap recovery energy was found to be 6.83 kW h for the Campus-Return route. This corresponds to 52% of the cumulative NTE of the HEB. This finding reflects the energy recovery efficiency of the regenerative braking system for the Campus-Return route. Consequently, the round-trip efficiency of the regenerative braking system was calculated to be 27%. Because of the limited number of measurements obtained and test vehicles used, these conclusions cannot be generalized. Depending on the characteristics of bus routes and energy management strategies, hybrid city buses can achieve different energy efficiencies on different routes. However, these results make significant contributions to better understanding the benefits and barriers of HEBs. Acknowledgements The author thanks to the Ministry of Science, Technology and Industry for their financial support, TEMSA R&D for their technical and financial support, Ali Fuat Iskender for technical assistance, Necdet Ayaz for driving assistance and Muammer Soylu for Autocad drawings of the schematics. References [1] White Paper 2011. Roadmap to a single European transport area – towards a competitive and resource efficient transport system. EC staff working paper. SEC (2011) 358 final, Brussels; 28 March 2011. [2] Perujo A, Christian Thiel C, Nemry F. Electric vehicles in an urban context: environmental benefits and techno-economic barriers. In: Soylu Seref, editor. Electric vehicles – the benefits and barriers; 2011. p. 1–18 [chapter 1]. http:// dx.doi.org/10.5772/20760. ISBN 978-953-307-287-6. [3] Davis PB. Overview of the U.S. DOE vehicle technologies program. Office of the Vehicle Technologies Program Energy Efficiency and Renewable Energy. U.S. Department of Energy, DOE-ACE-2011AR, USA; 2012. [4] VTP. Multi year program plan, 2011–2015. US Department of Energy, Energy Efficiency and Renewable Energy, Vehicle Technologies Program; December 2010. [5] Al-Alawi BM, Bradley TH. Analysis of corporate average fuel economy regulation compliance scenarios inclusive of plug in hybrid vehicles. Appl Energy 2014;113:1323–37. [6] Millo F, Rolando L, Fuso R, Mallamo F. Real CO2 emissions benefits and end user’s operating costs of a plug-in hybrid electric vehicle. Appl Energy 2014;114:563–71. [7] Andress D, Nguyen TD, Das S. Reducing GHG emissions in the United States’ transportation sector. Energy Sustain Dev 2011;15:117–36. [8] Jimenez-Espadafor et al. Infantry mobility hybrid electric vehicle performance analysis and design. Appl Energy 2011;88:2641–52. [9] Singh G. FY progress report for advanced combustion engine research and developments. Energy Efficiency and Renewable Energy Vehicle Technologies Program, DOE-ACE-2011AR, USA; 2011. [10] Lucena SE. In: Soylu Seref, editor. A survey on electric and hybrid electric vehicle technology, electric vehicles – the benefits and barriers; 2011. p. 1–18 [chapter 1]. ISBN 978-953-307-287-6. [11] Soylu S. Examination of PN emissions and size distributions of a hybrid city bus under real world urban driving conditions. Int J Automot Technol 2014;15(3):369–76. [12] Erlandsson L, Almen J, Johansson H. Measurement of emissions from heavy duty vehicles meeting Euro IV/V emission levels by using on-board measurement in real life operation. In: 16th International symposium ‘‘transport and air pollution’’, Graz; 2008. [13] Cocker DR, Shah SD, Johnson K, Miller JW, Norbeck JM. Development and application of a mobile laboratory for measuring emissions from diesel engines. 1. Regulated gaseous emissions. Environ Sci Technol 2004;38:2182–9. [14] Johnson KC, Durbin TD, Cocker DR, Miller WJ, Bishnu JK, Maldonado H, et al. On-road comparison of a portable emission measurement system with a mobile reference laboratory for a heavy-duty diesel vehicle. Atmos Environ 2009;43:2877–83. [15] Durbin TD, Johnson K, Cocker DR, Miller JW. Evaluation and comparison of portable emissions measurement systems and federal reference methods for emissions from a back-up generator and a diesel truck operated on a chassis dynamometer. Environ Sci Technol 2007;41:6199–204.
482
S. Soylu / Applied Energy 135 (2014) 472–482
[16] Zhihua L, Ge Y, Johnson KC, Shah AN, Tan J, Wang C, et al. Real-world operation conditions and on-road emissions of Beijing diesel buses measured by using portable emission measurement system and electric low-pressure impactor. Sci Total Environ 2011;409:1476–80. [17] Butler KL, Ehsani M, Kamath PA. Matlab-based modeling and simulation package for electric and hybrid electric vehicle design. IEEE Trans Veh Technol 1999;48:1770–8. [18] Schaltz E. In: Soylu Seref, editor. Electrical vehicle design and modeling, electric vehicles –modelling and simulations; 2011. p. 1–25. http://dx.doi.org/ 10.5772/20271. ISBN 978-953-307-477-1. [19] Ganley JC. Design and testing of a series hybrid vehicle with an ultracapacitor energy buffer. Proc Inst Mech Eng D J Automob Eng 2012;226:869–80. [20] Jaber K, Fakhfakh A, Neji R. Modeling and simulation of high performance electrical vehicle powertrains in VHDL-AMS. In: Soylu Seref, editor. Electric vehicles –modelling and simulations. Intech Open Access Publisher; 2011. p. 1–25. http://dx.doi.org/10.5772/16650. ISBN 978-953-307-477-1. [21] Barrero R, Coosemans T, Van Mierlo J. Hybrid buses: defining the power flow management strategy and energy storage system needs. World Electr Veh J 2009;3:2032–6653. [22] Xiong W, Zhang Y, Yin C. Optimal energy management for a series–parallel hybrid electric bus. Energy Convers Manage 2009;50:1730–8. [23] Ahn JK, Jung KH, Kim DH, Jin HB, Kim HS, Hwang SH. Analysis of a regenerative braking system for hybrid electric vehicles using an electro-mechanical brake. Int J Automot Technol 2009;10:229–34. [24] EC Regulation. No 595/2009 of the European Parliament and of the Council. Official Journal of the European Union; 18 June 2009. [25] Soylu S, Bal A, Semercioglu H, Ay EF. Examination of an urban city bus operating conditions and emissions. In: Soylu Seref, editor. Urban traffic and hybrid vehicles; 2010. p. 1–12. http://dx.doi.org/10.5772/10176. ISBN 978953-307-100-8. [26] Bal A, Semercioglu H, Soylu S, Ay EF. Sehir otobüslerinin gerçek dünya kosßullarinda motor calisßma ßsartlarinin ve NOx emisyonlarinin incelenmesi,
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
Otekon 2010, 5. Otomotiv Teknolojileri Kongresi, 7–8 Haziran 2010, Bursa, Turkey; 2010. Semercioglu H, Soylu S, Bal A. Examination of real world operating conditions and emissions of a hybrid city bus. In: ICAT’10 international conference on energy and automotive technologies, Istanbul; November 5, 2010. Jimenez-Palacios JL. Understanding and quantifying motor vehicle emissions with vehicle specific power and TILDAS remote sensing. Ph.D. thesis. Massachusetts Institute of Technology, Cambridge, MA, USA; 1999. MOVES2010. Highway vehicle population and activity data. Assessment and standards division, office of transportation and air quality. USA: US Environmental Protection Agency; 2010. Baisden AC, Emadi A. ADVISOR-based model of a battery and an ultracapacitor energy source for hybrid electric vehicles. IEEE Trans Veh Technol 2004;53(1). Grbovic PJ, Delarue P, Le Moigne P. Selection and design of ultra-capacitor modules for power conversion applications: from theory to practice. In: 2012 IEEE 7th international power electronics and motion control conference – ECCE Asia, Harbin, China; June 2–5, 2012. Miller JM. Power electronics in hybrid electric vehicle applications. In: Applied power electronics conference and exposition, 2003. APEC ‘03. Eighteenth annual IEEE; 2003. Mi C. Topology, design, analysis and thermal management of power electronics for hybrid electric vehicle applications. Int J Electr Hybrid Veh 2008;1(3):2008. O’Keefe MP, Vertin K. An analysis of hybrid electric propulsion systems for transit buses. Milestone completion report, NREL/MP-540-32858, USA; October 2002. Williamson SS, Lukic SM, Emadi A. Comprehensive drive train efficiency analysis of hybrid electric and fuel cell vehicles based on motor-controller efficiency modeling. IEEE Trans Power Electr 2006;21(3). Williamson SS, Emadi A, Rajashekara K. Comprehensive efficiency modeling of electric traction motor drives for hybrid electric vehicle propulsion applications. IEEE Trans Veh Technol 2007;56(4).