Advertisement

Hybrid Electric and Fuel Cell Hybrid Electric Vehicles

  • Sheldon S. Williamson
Chapter

Abstract

Different types of alternate vehicles (AVs) exist, such as EVs, HEVs, and fuel cell vehicles (FCVs). However, HEVs are found to be the most practical and efficient substitutes for CVs in the near future. This is because the characteristics of an electric motor are found to be more favorable, compared to the characteristics of an internal combustion engine (ICE). Different combinations of energy sources exist, for example, electric and mechanical (fly-wheel) energy sources or electric and chemical (fuel cell) energy sources. However, the combination of fuel energy and electric energy sources is found to be the most acceptable, due to the combined usage of mature ICE techniques and well-established modern power electronics.

Keywords

Fuel Cell Fuel Economy Internal Combustion Engine Drive Train Fuel Cell System 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

4.1 HEV Fundamentals and Concepts

4.1.1 Concept of HEV

Different types of alternate vehicles (AVs) exist, such as EVs, HEVs, and fuel cell vehicles (FCVs). However, HEVs are found to be the most practical and efficient substitutes for CVs in the near future. This is because the characteristics of an electric motor are found to be more favorable, compared to the characteristics of an internal combustion engine (ICE). Different combinations of energy sources exist, for example, electric and mechanical (fly-wheel) energy sources or electric and chemical (fuel cell) energy sources. However, the combination of fuel energy and electric energy sources is found to be the most acceptable, due to the combined usage of mature ICE techniques and well-established modern power electronics.

An HEV is defined as a vehicle whose propulsion energy is usually acquired from more than 2 types of energy sources, one of them being electric. In addition, an HEV electric drive train employs bidirectional power follow to re-capture the heat losses occurring during braking events, which would otherwise be lost in case of a CV. The history of HEVs is surprisingly found to be as old as the automobile itself. However, the initial purpose of employing an electric motor was not to reduce fuel consumption, but to merely help the ICE propel the vehicle. More recently, the purpose of using hybrid drive trains are plentiful:
  • To provide sufficient energy to satisfy the required driving range;

  • To supply sufficient torque to meet the needs of vehicle performance;

  • To achieve higher efficiency compared to CVs and to reduce fuel consumption and GHG emissions as much as possible.

4.1.2 Working Principle of an HEV Drive Train

As mentioned in the above section, the electric drive train of an HEV usually illustrates bidirectional power flow. Fig. 4.1 depicts the concept and power flow of a typical HEV. As is clear, the HEV can choose a particular path, in order to combine power flows liberally, to meet the required load demands. The control strategy of an HEV can be designed for different purposes, based on the varied combinations of power flows.
Fig. 4.1

Illustration of power flow within the hybrid drive train

As illustrated in the above figure, considering the drive train is a combination of fuel energy and electric energy, the HEV can work in the following pattern [1]:
  • Fuel drive train propels the load alone;

  • Electric drive train propels the load alone;

  • Both fuel and electric drive trains propel the load at the same time;

  • Electric drive train is being charged from load (regenerative braking);

  • Electric drive train obtains power from fuel drive train (ICE charging battery);

  • Electric drive train is charged by ICE and regenerative braking;

  • ICE delivers power to electric drive train, to charge the battery, and propels the vehicle at the same time;

  • Fuel drive train deliver power to electric drive train and the electric drive train propels the vehicle (series HEV);

  • Fuel drive train propels the load, and load delivers power back to electric drive train.

This freedom of choosing a suitable combination of power flows creates enormous flexibility compared to a single drive train, which has been used so far in CVs. However, such an operational characteristic introduces an interesting series of efficiency issues, which entail properly designing the fuel drive train as well as the electric drive train. In essence, the most appropriate and favourable design of the overall system control strategy is of paramount importance.

4.2 Efficiencies of Series and Parallel HEV Drive Trains

4.2.1 Introduction

In order to assess, analyze, and cross-compare efficiencies of HEVs, a true drive train analysis needs to be executed. Generally, the drive train efficiency can be simply yielded out by calculating the losses at each power stage in a series or parallel drive train structure. However, the power component stage-based analysis is a practically deficient method. In order to have a fair efficiency comparison, some parameters that directly affect fuel consumption, such as drive train mass and control strategy should be taken into consideration. This chapter aims at modeling both the series and parallel HEV drive trains, and computing their definite drive train efficiencies, which can be used as a comparative scale for various other HEV topologies. The ADVISOR software is used for modeling, simulation, and parametric analysis of series and parallel HEV drive trains.

In recent years, research results state that vehicle drive train efficiency is considered as an accepted measurement to evaluate and analyse the fuel economy among different types of HEV [2, 3, 4]. There exist two popular approaches to calculate drive train efficiencies. One of the methods focuses on the losses occurring at individual drive train components. This loss-oriented analysis is termed as power component stage based analysis. On the other hand, few other analyses concentrate on the influence of critical drive train parameters, such as vehicle mass (glider weight), control strategy, and drive train mass, on the overall fuel economy. These comprehensive analyses are imperative, in order to perform accurate advanced vehicle research, and present a fair comparison between different types of HEVs.

In this chapter, efficiency studies are carried out based on series and parallel HEVs. In order to meet the needs for the required torque, series HEVs usually employ a relatively powerful on-board battery pack and electric motor, which tends to increase the overall drive train mass. However, such an electric-intensive structure liberates the ICE, and allows it to operate at its optimal efficiency points, as an on-board generator. The parallel HEV configuration combines the electric traction with the combustion traction; therefore, the system has the freedom to choose the appropriate propulsion system combination, in order to achieve best efficiency as well as satisfy the load requirement.

This chapter analyzes and compares the drive train efficiencies of series and parallel HEVs by using two different concepts; power component stage based analysis and parametric analysis. Thus, the drive train efficiencies for series and parallel HEV arrangements are fully analyzed and contrasted from the overall efficiency and system performance standpoint.

4.2.2 Component Stage Based Efficiency Analysis

As the name suggests, the power component stage-based drive train efficiency calculation is based on the number of power component stages in a particular drive train. As depicted in Figs. 4.2 and 4.3, the dark arrows indicate the direction of power flow transmitted along the drive train. Bidirectional arrows stand for two power component stages; therefore, there are six power component stages for a series HEV drive train and nine stages for a parallel HEV drive train.
Fig. 4.2

Approximate calculation of maximum theoretical drive train efficiency for series HEVs based on power component stage analysis

Fig. 4.3

Approximate calculation of maximum theoretical drive train efficiency for parallel HEVs based on power component stage analysis

For a commercially available battery and power electronic converter, the maximum efficiency is typically around 80 % each [5, 6, 7]. The approximate maximum theoretical drive train efficiency for a typical series HEV is about 25 %, which is yielded out by simply multiplying the efficiencies of the corresponding power component stages. Fig. 4.2 shows the losses occurring at each stage of a parallel HEV drive train.

In order to compare the efficiencies of two different HEV configurations, all the power components used in the representative parallel HEV are assumed to be the same as those used in the series configuration. Considering there are two independent power flow channels (via electric traction and via mechanical traction) for the parallel configuration, the calculated drive train efficiency can be theoretically multiplied by a factor of 2. Thus, based on the power component stage analysis, the approximate maximum theoretical drive train efficiency for a representative parallel HEV is around 45 %.

4.3 Varied Driving Patterns and Regenerative Braking Efficiency Analysis

4.3.1 Introduction

It is a well-understood fact that power electronic converters and electric propulsion motors are extremely critical for every HEV system. It is essential that the traction motor must meet demands of varied driving schedules, and at the same time, it should run at its most optimal operating points to achieve higher drive train efficiency. Therefore, modeling the motor-inverter losses/efficiencies over typical city and highway driving schedules is the key to observe and analyze practical drive train efficiency and vehicle performance.

Currently, land vehicles are transiting from pure gasoline-fuelled to gasoline-electric combined HEV. In fact, there exist about 1,500,000 satisfied HEV users around the world. This number is constantly increasing, based on a yearly rate of roughly about 40 %. A major increase is noticeable in light-duty hybrids, such as passenger cars and SUVs [8, 9]. Majority of these vehicles are used for personal transportation, typically driven in the urban area or on a highway. Therefore, HEVs are not just run in particular routes, but in varied driving patterns. As a complex system, the automobile features vast mobility and variation. It is hard to measure its efficiency by using a simple measurement [10]. Thus, well-rounded measurements, such as well-to-wheel efficiency and tank-to-wheel (drive-train) efficiency are employed to fairly evaluate its efficiency.

However, it is a fact that HEVs cannot be optimally designed for all driving patterns. Various control parameters and efficiency data over particular driving schedules are required to optimize the control strategies for HEVs. Fortunately, because of commuters’ regular routes, most of the driving patterns are predictable. In addition, automobile manufactures are tending to design HEVs that can use the electric propulsion motor over all range of load demands. Thus, a motor with high torque density, high efficiency, excellent controllability, and accuracy is needed. Therefore, the efficiency of traction motor-inverter drive system needs to be specifically studied and analysed on the basis of varied load demands [11].

Keeping the above-mentioned constraints in mind, the major focal point of this chapter is to model the traction inverter and motor losses/efficiencies over typical driving patterns. Efficiency maps are usually used to describe total efficiency of traction motors with respect to certain speed/torque combinations. The overall HEV drive train efficiencies are determined by the resultant traction motor-controller efficiency maps during simulation. Consequently, by using efficiency maps, an HEV motor drive can be represented as a “black box” that provides a known output when certain input is applied. The modeling concept is shown in Fig. 4.4. Thus, the traction motor can be studied at all possible torque/speed combinations within the motor’s operating envelope by analyzing its operating efficiency.
Fig. 4.4

HEV motor drive efficiency modeling concept based on operating efficiency maps

In this chapter, optimal control strategies are used for parallel HEVs, based on the UDDS driving pattern, which generally represents an average urban driving model. Based on the simulation results, a comparative analysis is carried out for 6 selected different driving patterns, which show varying results in terms of overall drive train efficiency, vehicle performances, and overall emissions.

4.3.2 Vehicle Specification and Modeling

In this chapter, a typical mid-sized SUV is selected to represent the most popular vehicle in the market. The major dimensions and weights of the tested SUV chassis are summarized and revised for simulation purpose based on current commercially available SUVs. The physical parameters of the tested SUV are shown in Table 4.1.
Table 4.1

Physical parameters of tested parallel hybrid SUV

Parameters

Value

Dimensions

Coefficient of drag

0.34

Frontal area (m2)

3.15

Wheelbase (m)

2.72

Overall length (m)

4.71

Vehicle mass

Vehicle weights (kg)

2,701

Cargo weight (kg)

156

As aforementioned, a parallel HEV drive train configuration is chosen for simulation purposes. Parallel HEV drive train configuration in Fig  2.1b shows the block diagram of the modeled parallel SUV. It is clear that a parallel HEV structure is a combined traction source arrangement [12, 13]. Thus, the hybridization factor (HF), which is defined as the ratio of the total electric power to the total propulsion power, plays an important role in the overall efficiency.

The fuzzy logic drive train control strategy is selected for simulation purpose and it is optimized based on UDDS, by adjusting its SOC and HF to have better city-driving efficiency. The tested drive train components were optimized by using the auto-size routine, which employs a bisection method to optimize the component size, based on the required vehicle performance. For the simulated parallel SUV, first the entire energy storage system size is minimized, and then its minimum fuel converter (FC) size is determined, according to the vehicle performance criteria. Finally, the FC size is fixed by a suitable HF, based on the following equation.
$$ fc\_pwr = fc_{\_pwr\_\hbox{min} } + HF * (fc_{\_pwr\_\hbox{max} } - fc_{\_pwr\_\hbox{min} } ) $$
(4.1)

Here, fc_pwr = new fuel converter size; min_fc_pwr = required minimum fuel converter size to meet the vehicle performance; max_fc_pwr = required maximum fuel converter size to meet the vehicle performance; HF = hybridization factor.

Once the FC size has been determined, the routine resizes the energy storage once again, to meet the acceleration requirements. The UDDS driving pattern, which is equivalent to the first 2 cycles of the Federal Test Procedure (FTP-75) driving schedule, represents general city driving conditions. Thus, UDDS is selected as a reference, for optimizing the drive train components [14]. The optimized drive train components are listed in Table 4.2.
Table 4.2

Summary of parallel HEV drive train components

Parameters

Value

Fuel converter

Max. power

17kw @ 4,000 rpm

Max. torque

45Nm @ 4,000 rpm

Fuel converter mass

70 kg

Battery (Nickel-Metal Hydride)

Single module voltage

12 V

Number of modules

25 modules

Nominal capacity

60 Ah

Peak power (10 s pulse @ 50 % DOD @ 35 deg. C)

4.9 kW

Mass

290 kg

Motor-controller (PM)

Continuous power

48 kW

Peak torque

370 Nm

Motor-controller mass

58 kg

Maximum speed

4000 rpm

The selected drive train was tested over 6 different driving schedules, which include 3 stop-and-go type low-speed driving patterns and 3 high-speed driving patterns. The driving patterns include the West Virginia Suburban (WVU-SUB), the Urban Dynamometer Driving Schedule (FUDS), 10–15 Japan driving schedule, the Highway Fuel Economy Test (HWFET), US06 Highway Driving Schedule (US06 HWY), and the Extra-Urban Drive Cycle (EUDC) [15]. The speed profile versus time for each of the above-mentioned 6 driving schedules and their detailed characteristics are summarized in Fig. 4.2 and Table 4.3, respectively.
Table 4.3

Summary of the 6 different driving schedules

Driving Schedule

Dist. (miles)

Stops

Max. Spd. ( mph)

Ave. Spd. (mph)

Max. Accel. (ft/s2)

UDDS

7.45

17

56.7

19.58

4.84

WVU-SUB

7.44

9

44.8

16.7

4.25

10–15 Japan

2.61

14

43.96

14.24

3.89

HWFET

10.26

1

59.9

48.2

4.69

US06 HWY

6.24

1

80.3

60.84

10.12

EUDC

4.32

1

74.56

38.8

3.46

The modeled mid-sized hybrid SUV is simulated over 3 reiterations of each driving schedule. The overall efficiency and vehicle performance will be compared in the ensuing sections, which are directly influenced by the different driving patterns.

4.3.3 Overall Efficiency Comparison Based on Varied Driving Patterns

As aforementioned, the ADVISOR software is used in order to determine the overall efficiency based on above-mentioned driving schedules. The comparative overall efficiencies of different driving schedules are shown in Fig. 4.6.

As explained earlier, a parallel HEV is the combination of different traction sources, and hence, the system has the freedom to choose the suitable propulsion system combination. For city driving, the number of stops and starts is 10 times more than that in case of highway driving. Thus, the ICE is more frequently used in city driving patterns than highway driving patterns, which results in a lower efficiency. Moreover, steep decelerations are harmful for regenerative braking energy recovery. Especially in the case of 10–15 Japan, there are 14 stops-and-starts in only 2.61 miles, as shown in Table 4.3, and the average deceleration is the greatest amongst the 3 city driving patterns. Therefore, less energy is recovered from regenerative braking and the overall drive train efficiency over 10–15 Japan is the lowest, although the driving distance and acceleration is much less compared to other city driving schedules [16].

The overall fuel economy over the designated 6 driving schedules is shown in Fig. 4.5. It can be easily observed that the overall drive train efficiency is not necessarily proportional to fuel economy. For example, the fuel economy under 10–15 Japan driving conditions is higher than that under UDDS conditions, but the overall drive train efficiency is lower. The higher fuel economy of 10–15 Japan is because it possesses lower average speed as well as maximum acceleration, which leads to less usage of the ICE and leads to higher motor-controller efficiency. In the case of drive train efficiency, UDDS has a higher efficiency, because of the efficient usage of regenerative braking. As is clear from Table 4.3, there are a total of 51 stops in UDDS, but only 42 stops in 10–15 Japan. The simulation data also shows that the energy generated by regenerative braking under UDDS conditions is 2 % higher than that of 10–15 Japan, which indicates that more energy is saved in the drive train, as shown in Fig. 4.6. In addition, the electric motor is more frequently used in UDDS, which leads to a higher efficiency.
Fig. 4.5

Simulated driving schedules for test purposes a City driving patterns, b Highway driving patterns

Fig. 4.6

Overall drive train efficiency over different driving schedules

Moreover, it is easy to notice that a single cycle distance of UDDS is 3 times greater than that of 10–15 Japan. Since the simulation is carried out over 3 repetitions for each driving schedule, the simulated distance of UDDS is approximately 9 times longer than that of the 10–15 Japan. On the other hand, the tested SUV runs as an electric vehicle because the high (80 %) initial SOC. The ICE starts working when the SOC decreases to a designated value. For these reasons, the SUV tested over 10–15 Japan is more likely to stay in electric mode longer than UDDS, which contributes a higher fuel economy. This result also suggests that a higher fuel economy can be obtained by maintaining the ESS in its high SOC range and driving the vehicle for shorter distances (Figs. 4.7, 4.8).
Fig. 4.7

Fuel economy over different driving schedules

Fig. 4.8

Comparison of available power into the motor under UDDS and 10–15 driving patterns

According to the simulation results, it is easy to see that the parallel drive train system is more suitable for highway driving. Both the fuel economies as well as the overall drive train efficiencies, under highway driving, are comparatively higher than in case of city driving patterns. This is mainly because the ICE nearly reaches its maximum efficiency (around 40 %). When the vehicle is running on the highway, the smooth driving patterns allow either the ICE or the electric motor to operate at its respective optimal operating point. Although the recovered energy from regenerative braking is much less compared to city drive cycles, as a low-efficiency power component, the ICE running at its most efficient operating points is a significant factor in improving the drive train efficiency.

4.4 Regenerative Braking Efficiency Analysis

According to the definition of regenerative braking efficiency in the previous section, the regenerative braking efficiency for each control strategy is calculated by retrieving the current that flows into motor for regenerative braking and the energy used by accessory loads, during regenerative braking events. The results are calculated based on equation \( \eta_{REGEN} = \frac{{E_{regen} }}{{E_{neg.trac} }} \bullet 100\% \) and summarized in Fig. 4.9.
Fig. 4.9

Comparative electric drive train regenerative braking efficiencies

The considerable improvement of regenerative braking efficiency in electric assist mode indicates the flexibility of this control strategy, as it is originally designed to use electric traction, when needed. Therefore, increase of the possibility of using electric traction is allowed. As shown in Fig. 4.10, the modification of regenerative braking control look-up table increases the percentage of regenerative braking usage. A cumulative amount of 20 % increment was observed compared to the original case. For the fuzzy efficiency control strategy, due to the preset ZEV condition, the regenerative braking efficiency does not increase. The compromise between regenerative braking efficiency and motor-controller efficiency is made by sacrificing regenerative braking efficiency. This is done principally because in the fuzzy efficiency mode, the motor-controller efficiency has greater priority compared to regenerative braking efficiency, as explained in the previous section.
Fig. 4.10

Diagram of braking control logic

Moreover, it is clear that the fuzzy efficiency mode is better at using electric traction and regenerative braking than the electric assist control strategy. In addition, high regenerative efficiency does not necessarily mean desirable overall drive train efficiency.

4.5 Overall Electric Drive Train Efficiency Analysis

The ultimate purpose of control strategy optimization is to improve the overall drive train efficiency by enhancing motor-controller efficiency. Also, as a key contributor to the overall drive train efficiency, regenerative braking efficiency is also appropriately optimized. Fig. 4.11 shows the comparative overall electric drive train efficiencies over the 4 different control strategies. As is evident, the improvement is approximately 2 times greater than the untreated strategies, which justifies the suitable arrangement and correct proportion of efficiency improvement between the motor-controller efficiency and regenerative braking efficiency.
Fig. 4.11

Comparative overall electric drive train efficiencies

In general, the fuel economy is proportional to the overall drive train efficiency. However, it is important to note that the rates of increase of fuel economy for the 2 proposed control strategies are slightly different. Consequently, it is easy to see from Fig. 4.12 that the fuel economy increasing rate, when using electric assist control strategy, is larger than that when using the fuzzy efficiency mode. By retrieving the simulation data from the fuel converter and motor-controller, it is straightforward to explain the above-mentioned ambiguity. Since the driving patterns for the 2 control strategies are same, the energy used during the simulated 5 driving cycles is equal as well. However, the employment of electric traction and the effectiveness of regenerative braking are different from each other. Based on the previous analyses, the regenerative braking efficiency, when using the optimized electric assist control strategy, is found to be much higher than when using the fuzzy efficiency mode. Nearly 6 to 7 % of total fuel consumption is saved due to regenerative braking in the case of electric assist control strategy compared to when using fuzzy efficiency mode.
Fig. 4.12

Comparative fuel economies over 4 proposed control strategeis

Fig. 4.13 summarizes the efficiency improvement in terms of motor-controller efficiency, in both powering mode as well as regenerative braking mode. Fig. 4.13 also depicts the respective regenerative braking efficiency, overall drive train efficiency, and fuel economy improvements. There are only 2 negative increments, motor-controller efficiency in regenerative mode and the regenerative braking efficiency in fuzzy efficiency mode, which are mainly restricted by the ZEV setting in powering mode, as explained in the previous section. Nevertheless, the increment of overall drive train efficiency and fuel economy, respectively, is more than 100 % higher, which justifies that the decrease is correct and proves to be a necessary trade-off.
Fig. 4.13

Comparative efficiency improvment

4.6 Fuel Cell HEV: Modelling and Control

4.6.1 Modeling Environment

The baseline vehicle and control strategy is modeled and analyzed in the Advanced Vehicle Simulator (ADVISOR) software, which is developed in the MATLAB/Simulink environment [17]. ADVISOR is composed of a group of models, experimentally verified data, and script files. It not only allows the designer to obtain a quick analysis of the performance and fuel economy of conventional, electric, hybrid electric, and fuel cell vehicles, but it also provides detailed simulations and analysis of user-defined power train components, by taking advantage of the modeling flexibility of Simulink and the analytical power of Matlab [18].

ADVISOR uses 3 primary graphical user interface (GUI) screens to guide the user through the simulation process. The GUI facilitates interaction with the raw input and output data that is present in the MATLAB workspace. The vehicle model is depicted graphically using Simulink block diagrams, to define the connections between components, as shown in Fig. 4.14. The component models can be inserted into a vehicle model and then connected to define the flow of torque or speed and power from one component to the next. The arrows entering the top input of a component block in the fuel cell vehicle model, shown in Fig. 4.14, represent a torque and speed or a power demand from one component to the next upstream component. The power demand is based on the vehicle speed requirements and the losses of each component. Arrows entering the bottom input port of each block represent what the upstream component is able to achieve.
Fig. 4.14

Overall vehicle Simulink diagram in ADVISOR

In general, individual component models are a combination of algorithms programmed in Simulink and data files that store various tuning parameters for the algorithms. By incorporating various vehicle performance and control information into a modular environment within Matlab and Simulink, ADVISOR allows the user to interchange and design a variety of components, vehicle configurations, and control strategies. It also allows quick analysis of the vehicle performance, emissions, and fuel economy of conventional, electric, and hybrid electric vehicles.

4.6.2 Modeling and Selection of Power Components

In this section, the sizing and modeling of the Fuel Cell-Hybrid Electric Vehicle (FC-HEV) power train system are introduced. The power components mainly include the fuel cell system, battery system, and the motor-controller system.

4.6.3 Fuel Cell System

As aforementioned, fuel cells are electrochemical devices that convert the energy of a chemical reaction between hydrogen and oxygen directly into electrical energy. Various types of fuel cells exist, but as stated in  Chap. 1, the PEM fuel cell is regarded as the most promising option for automotive application, due to its high power density, low operating temperature of about 80 °C, and high overall efficiency [19].

ADVISOR includes 2 options for modeling the fuel cell. The first one is based on look-up tables, indexed to the polarization curves, which characterize the fuel cell stack performance. The key assumption is that the system can provide a specific net power, while consuming a set amount of fuel, regardless of how complex the system may be [18]. A used net power vs. efficiency data for PEM fuel cell stack built in ADVISOR is shown in Fig. 4.15.
Fig. 4.15

Net power versus. efficiency map for a 50 kW fuel cell system model

The performance of the auxiliary systems, such as air compressor and fuel pump, can be also characterized with polarization curves, from experimental data in ADVISOR. The power delivered by the fuel cell system is the difference between the power produced by the fuel cell stack and the power consumed by the auxiliary system. The second option is to model a fuel cell stack in a much more complete manner through a co-simulation link between ADVISOR and General Computational Toolkit (GCtool). In such a case, the electrochemistry, thermal characteristics, and mass transfer characteristics can also be incorporated. It must be pointed out, though, that such a detailed model is not necessary for overall vehicle system-level performance analysis.

4.6.4 Battery System

A suitable energy storage system (ESS) is required to assist the fuel cell system, to meet the power demand from the drive train. Currently, lead-acid batteries are employed in conventional cars, because of their low price and rugged structure. On the other hand, for recent HEV applications, nickel metal-hydride (Ni-MH) batteries are commercially used in the market. Compared to lead-acid batteries, Ni-MH batteries generally have much longer lifespan, higher power output, and increased charge and discharge efficiency. Besides, they are also safely recyclable [20]. Ni-MH batteries have been employed successfully in vehicles in the state of California, and demonstrated promise to meet the power and endurance requirements for electric vehicle (EV) propulsion. Meanwhile, Lithium-ion (Li-ion) batteries are likely to become serious competition for Ni-MH in EV/HV applications, but their operating life is still limited. In addition, ultra-capacitors are also currently under investigation in several research programs, but their energy density is much lower than those of batteries. The main advantage of ultra-capacitors is their high power density, which make them great options for hybridizing with battery systems, for supplying short bursts of power during acceleration, or receiving short bursts of regenerative currents, during quick decelerations. Fig. 4.16 shows the energy and power densities comparison of common energy sources [20].
Fig. 4.16

Energy and power densities of various energy storage components [20]

The battery model type used for the FC-HEV under study is the Ovonic 45Ah Ni-MH battery. The main performance characteristics of this battery are summarized in Table 4.4. The battery is modeled in ADVISOR based on the internal resistance model, as shown in Fig. 4.17. The circuit determines the output voltage and current based on the load, while estimating the rate at which this power level depletes the resistor through the internal model calculation.
Table 4.4

Ni-MH Battery parameters

Nominal Voltage

12 V

Nominal capacity (C/3)

45 Ah

Nominal energy (C/3)

598 Wh

Peak power (10 s pulse @ 50 %DOD @ 35 deg. C)

3.3 kW

Weight

8 kg

Volume (modules only)

3.2 L

Fig. 4.17

Internal resistance battery model electrical schematic

Due to the non-linear behavior of batteries, the parameters of the simulation circuit are determined from experimental data collected by the Battery Thermal Management Laboratory, manufacturer data sheets, as well as lab tests [21]. At each time step, the net battery current is then used to estimate the change in State of Charge (SOC) of the battery. Fig. 4.18 shows the internal resistance of the battery at 40 °C.
Fig. 4.18

Resistance of the Ni-MH battery at 40 deg. C in ADVISOR

4.6.5 Motor-Controller System

The electric traction motor system plays an important role in the performance of a FC-HEV. The main requirements for motor selection include: high torque density and power density; wide speed range, including constant torque and constant power operations; high efficiency over wide speed range, high reliability, and robustness; a reasonable cost [22]. There are 3 motor types suitable for HEV applications: permanent magnet motors, induction motors, and switched reluctance motors. The permanent magnet machines possess high efficiency, high torque, and high power density. However, they inherently have a short constant power range, due to limited field weakening capability. In addition, the back EMF can also be a problem at high speeds, because the inverter must be able to withstand the maximum back EMF generated by the stator winding. The switched reluctance motor (SRM) is a promising candidate for HEVs, due to of its simple construction, simple control, and good extended speed performance. However, since the SRM is not yet widely produced as a standard motor in the market, the overall electric propulsion system cost may be higher than other motor options.

Thus, the popular induction motor (IM) is selected for FC-HEV modeling in the thesis due to its simplicity, robustness, and adequate extended speed range. Also, IMs do not have back EMF to deal with, at high speeds [22]. Field-oriented control makes an IM behave like a simple DC machine. In ADVISOR, the entire motor system is modeled based on motor efficiency maps, where the motor efficiency is determined as a function of toque and speed. Fig. 4.19 shows the efficiency map of the Westinghouse 75 kW IM. The bold lines represent the maximum torques, according to the speed of the motor.
Fig. 4.19

Efficiency map of a Westinghouse 75 kW AC induction motor [17]

Corresponding to the backward-facing vehicle modeling approach, the desired speed and torque requests, propagated from the transmission, are translated by the motor model into a power request through a series of mathematical equations.

4.6.6 Baseline Vehicle

The vehicle dynamic model is described by the typical force balance equation as shown in 4.2, from which the total driving force is computed as the sum of rolling resistance force, aerodynamic resistance force, acceleration force, and climbing resistance force. The model first calculates the required driving force, according to the required acceleration. Thereafter, the achievable acceleration is calculated, based on the output driving force. The vehicle speed is determined by the driving cycle, transmission gear ratio, and the wheel radius. In this thesis, we assume that the vehicle has a one-speed transmission.
$$ {\text{Ftotal }} = {\text{ Frolling }} + {\text{ Faero }} + {\text{ Facc }} + {\text{ Fclimb}} $$
(4.2)
The vehicle characteristics are assumed to be based on current production of baseline conventional vehicles. The vehicle parameters are selected based on 2 types of vehicles: the mid-size family sedan and mid-size SUV. Table  2.2 outlines the vehicle modeling assumptions (Table 4.5).
Table 4.5

Vehicle specifications

Vehicle type

Mid-size SUV

Mid-size Sedan

HEV glider mass

1,179 kg

636 kg

Cargo mass

136 kg

136 kg

Fuel cell vehicle goss mass

2,095 kg

1,300 kg

Rolling resistance

0.012

0.012

Frontal area

2.66 m2

2.0 m2

Coefficient of drag

0.44

0.35

4.6.7 Summary

This chapter summarized the sizing and modeling aspects of the vehicle and its main power components. The ADVISOR software as a modeling and simulation environment was introduced. The complete modeled block diagram of the FC-HEV is shown in Fig. 4.20.
Fig. 4.20

Block diagram of the modeled FC-HEV drive train

As mentioned earlier, the PEM fuel cell is used due to its high power density, low operating temperature, and high efficiency. For system level performance analysis, the fuel cell system is modeled by look-up tables, indexed to the polarization curves, which characterize the fuel cell stack performance. The nickel metal-hydride (Ni-MH) battery is used as the ESS, because of its high energy density and reasonable cost. The battery is modeled based on the internal resistance model and experimental data. The motor system used is an AC induction motor, which is modeled based on its efficiency map. The baseline vehicle parameters are selected based on current production of conventional vehicles. In the analyses performed in the ensuing chapters, 2 types of baseline vehicles are considered; they include a mid-size family sedan and a mid-size sports utility vehicle (SUV).

4.6.8 Control Fuel Cell HEV

4.6.8.1 Introduction

A typical drive train layout of a FC-HEV with control information flow and power flow is shown in Fig. 4.21. The FC-HEV utilizes the fuel cell system as the main power source to provide electricity and uses a reversible energy storage accumulator, such as a battery or an ultra capacitor, as a supplementary power source. This hybridization not only downsizes the fuel cell and fulfills transient power demand fluctuation, but also leads to significant energy savings through regenerative braking energy recovery [23].
Fig. 4.21

Main schematic of the overall system

As is the case with regular hybrid electric vehicles (HEVs), a good system level power control strategy is essentially required to solve the problem of managing the power sharing between the fuel cell and the battery. An optimal control strategy design helps achieve maximum fuel economy, system efficiency, and maximize ESS life span, while maintaining required vehicle dynamic performance. In addition, simplicity, feasibility, and robustness are also important factors to evaluate different power control strategies. Various types of power control strategies have been proposed for HEVs, which could be extended to FC-HEV applications [24, 25, 26, 27, 28, 29, 30, 31, 32].

Some of the popular FC-HEV power control strategies are reviewed in the ensuing sections. Thereafter, optimized design, modeling, and in-depth analysis are performed on 2 types of control strategies, namely the load follower control scheme and the equivalent consumption minimization strategy (ECMS). In order to investigate their control performance, and to further optimize their respective designs, detailed comparisons and analyses based on simulation tests, are also presented in this chapter.

4.6.8.2 Review of FC-HEV Power Control Strategies

The general goal of the power control strategy for a typical FC-HEV drive train is to maximize the vehicle system efficiency and enhance fuel economy, while maintaining the required vehicle performance. There are several global optimization algorithms, such as dynamic programming (DP), developed for HEVs, to find the optimal solution of power distribution [31]. The DP method is a cost function based dynamic optimization tool, which can guarantee global optimal solution up to the grid accuracy of the states. However, these kinds of strategies are based on a prior knowledge of future driving conditions. Therefore, they are not suitable for real-time control, but they can best serve as a benchmark for improving other control strategies.

Rule-based control strategies are popular for FC-HEV power management, due to their simplicity and feasible implementation. These types mainly include the thermostat scheme, load follower scheme [17, 24], and fuzzy logic scheme [26, 27, 28]. The thermostat scheme features simplicity and robustness. Under this scheme, the fuel cell will turn on and off based on the battery SOC. The fuel cell turns on when the SOC reaches the low limit and turns off when the SOC reaches the high limit. When the fuel cell is on, it will always operate at the most efficient power level, as shown in Fig. 4.22, which compares the fuel cell operating point of the thermostat scheme with load follower scheme. Although this strategy is simple and easy to control, it has some disadvantages; firstly, it cannot satisfy vehicle driving requirements, especially during acceleration or high power command. Moreover, this strategy leads to frequent charging/discharging of the battery, which is unfavorable.
Fig. 4.22

Fuel cell operating points with the load follower and the thermostat scheme

The load follower scheme, to a large extent, can solve the problems occurring in the thermostat scheme. The basis of the power follower scheme is to determine the operation state of the fuel cell, according to the power demand from the vehicle and the battery state of charge (SOC). The fuel cell output is never to be a constant value, but tends to change, following the transient power requirements in a reasonable region. A minimum and a maximum output power level (Pfc_min, Pfc_max) should be determined to avoid the fuel cell system operating in low efficiency. Meanwhile, the battery SOC should be controlled within a range where regenerative braking energy can be effectively absorbed, while ensuring battery life. In this thesis, this control strategy is selected for the purpose of optimal design and simulation test based analyses in the ensuing sections, because it can achieve better control performance compared to thermostat, and it is easy to create an acceptable design within a short time.

More recently, fuzzy logic is becoming increasingly popular in hybrid vehicle control, because it enables the development of dynamic rule-based behavior. It solves the problem that exists in static control approaches, where the parameters are normally optimal for a specific vehicle type and a specific driving condition, while becoming sub-optimal in other conditions. The main advantage of fuzzy logic control schemes is that they can be tuned and adapted to the specific driving conditions and plant dynamics, thus enhancing the degree of freedom of control [28]. Another benefit is that it does not depend on accurate mathematical modeling, which is hard to obtain for complex systems, such as FC-HEVs.

In a fuzzy logic controller, the knowledge of an expert can be coded in the form a rule-base, and can be used in decision making. A basic type of fuzzy logic based power control scheme is illustrated in Fig. 4.23 [26]. As is clear, the inputs of the fuzzy controller are battery SOC and battery current, and the fuzzy output is the required current for the fuel cell. Fig. 4.24 shows the basic analysis method for fuzzy logic control. The fuzzifier converts the crisp input value into a fuzzy value, with degrees of membership functions.
Fig. 4.23

Fuzzy logic based power control scheme

Fig. 4.24

Structure of fuzzy logic controller

The inference engine combines the fuzzy rules into a definite map, from a fuzzy set of inputs to the output, based on fuzzy logic principles. The defuzzifier then reconverts the resulting fuzzy value into a specific crisp value, as a reference variable. The heart of a fuzzy system is a set of knowledge-based IF–THEN fuzzy rules. However, the main limitation of a fuzzy logic control strategy is its complexity, which limits implementation flexibility.

To develop a cost function based optimal algorithm, which is real-time applicable, some improved strategies have been proposed. The Stochastic Dynamic Programming (SDP) has been proposed to solve the power management as a stochastic problem [32]. The basic principle of SDP problem formulation is to model the power command as a stochastic process, and an optimal controller based on the stochastic model can be designed, in order to find an optimal control policy that maps the control decision against the vehicle operation states. At the same time, the disadvantage is that it is computationally expensive to build cost tables and corresponding optimal control for complex dynamic systems. Another popular cost function based control strategy is the equivalent consumption minimization strategy (ECMS) [29], which is developed for parallel HEVs. The ECMS replaces the global cost function to a local one, which adjusts the instantaneous power split by calculating an equivalent fuel cost function for an array of power splits between 2 energy sources, and selects the split with the lowest fuel cost. This type of control strategy can often reach a nearly optimal operation set point. The ECMS strategy, as a representative of cost-function based control strategies, is selected for optimized design and simulation-based study in the ensuing sections.

4.7 Power Electronics Interface of Fuel Cell and Traction System

4.7.1 Introduction

This section will present the low level power circuit design and control of the power train system for a FC-HEV. As illustrated in Fig. 4.25 of the proposed power train system configuration, the power conditioner between the fuel cell and the battery system plays a crucial role, in order to provide protection of power components and matching the voltage levels of different power sources to the main DC bus. Meanwhile, it will also provides control of the demanded power according to the reference power value (usually transfer to reference current) from the system supervisory controller, as discussed in last chapter.
Fig. 4.25

Power train control system of a FCHEV

In the ensuing sections, the power train topology selection based on the component characteristics and power requirements will be discussed. Two popular topologies are considered based on 2 options of hybridization degree selection. Thereafter, the circuit modeling of power components in PSIM software as well as the power converter design will be introduced. Finally, the control scheme design and simulation-based analysis will be presented.

4.7.2 Power Train Configuration

An ideal topology for FC-HEV is that both the fuel cell system and the ESS are directly connected to the propulsion motor. This seems to be most efficient and economical configuration. However, this configuration is applicable only if both the fuel cell and ESS output voltages match the voltage level of the DC bus, which in turn, needs be set to a level suitable for usage with the motor system. The fact is that fuel cell output is usually lower than the DC bus requirement, and tends to depict a wide variation during operation, or the voltage between the fuel cell and battery does not match [33]. This leads to low efficiency and reliability, and proves to be extremely complicated when providing power distribution control in a hybrid configuration.

Therefore, a DC/DC converter is necessary for power conditioning between the 2 power sources and the DC voltage bus, which is in turn connected to the propulsion system. In general, there are 2 options of power train structures, based on the position of the DC/DC converter, as shown in Fig. 4.26. In topology-A, the fuel cell is connected to the high voltage DC bus through a unidirectional DC/DC converter, while the battery is directly connected to the DC bus. In this condition, the fuel cell output can be directly controlled, while the battery output voltage needs to match the DC bus voltage level. In topology-B, the fuel cell is directly connected to the DC bus, while the battery is connected to the DC bus through a bi-directional DC/DC converter.
Fig. 4.26

Power train topological options for FC-HEVs

The utilization of a bi-directional converter between the battery and DC bus allows more flexibility to the battery, because such an arrangement not only reduces the voltage requirement of the battery, but also provides the freedom to control its state of charge (SOC). Since the fuel cell is directly connect to the high voltage DC bus, a large sized stack or voltage level is required, and the control of fuel cell power can only be achieved indirectly, by controlling the battery output, or through internal fuel control. The use of 2 high power converters for both battery and fuel cell generally is not economical from the point of view of cost and size consideration.

In this section, the topology selection is considered according to the hybridization degree. The hybridization degree here is defined as the ratio between the fuel cell rated power and the peak power of the traction motor. A higher hybridization degrees leads to better fuel economy, but requires a large sized fuel cell, which in turn leads to high cost. Here, the sizing of a mid-size SUV type vehicle is considered for study, since SUVs are one of the most popular and fuel inefficient vehicle type. Moreover, SUVs provide more potential to arrange the power component size, since they have large space. Two specific cases of hybridization degrees are selected based on the peak/average power requirement of a mid-size SUV (140 kW/50 kW, in this case), as described in Table 4.6.
Table 4.6

Two cases of hybridization for a mid-size SUV

 

Fuel cell size (PEM) (kW)

Battery size

(Ni-MH)

Power train

type

Case-1

60

25 modules

Topology-A

Case-2

80

20 modules

Topology-B

For case-1, power train topology-A is chosen, because the fuel cell size is relatively small and the voltage level for a 60 kW PEM fuel cell (200–300 V) does not match the DC bus voltage (300–450 V). Therefore, the DC/DC converter allows a downsized fuel cell and can allow complete control. At the same time the rated voltage of the 25 battery pack cells can be set to be around 400 V, which represents the DC bus voltage level. For case-2, which has a larger fuel cell and smaller battery pack, the 80 kW fuel cell (300–400 V) can be directly connected to the DC bus, while a bi-directional DC/DC converter is needed for the battery, to match the voltage level as well as to control the battery charging and discharging performance.

4.7.3 Power Component Modeling

4.7.3.1 Fuel Cell System

To obtain the electrical characteristics of the fuel cell, its circuit model is represented by a look-up table and controlled voltage source, which provides the fuel cell voltage corresponding to the current drawn from the fuel cell, as shown in Fig. 4.27. The diode at the fuel cell output is to prevent the negative current going back into the stack, and an on–off controller is added, to ensure that the fuel cell operates in an acceptable area.
Fig. 4.27

Fuel cell model

The fuel cell V–I characteristic is normally portrayed in the form of a polarization curve, which is determined by the relation between cell voltage and current density, as described in  Chap. 1. Figure 4.28 illustrates the V–I curve of a 60 kW fuel cell. The stack temperature and membrane water content affect the fuel cell voltage. The voltage decreases as higher current is drawn from the fuel cell, due to the fuel cell electrical resistance, inefficient reactant gas transport, and low reaction rate [34]. Lower voltage indicates lower efficiency of the fuel cell, and the maximum current drawn from the fuel cell is defined as the current at which the maximum output power is achieved. Many cells are typically combined in a stack, to satisfy the power requirement of the target application.
Fig. 4.28

V-I polarization curve of a 60 kW PEM fuel cell stack

4.7.3.2 Battery System

The battery system is modeled based on a typical RC model, as described in  Chap. 2. This model consists of a voltage controlled voltage source in series with an internal resistor, as shown in Fig. 4.29. The battery output voltage is determined by the battery SOC, through a look-up table. The relation between battery cell-voltage and SOC is obtained from validated experimental data, as shown in Fig. 4.30. The battery SOC is calculated as the energy present in the battery divided by the maximum energy capacity (Ah) of the battery pack, as given in 4.3.
Fig. 4.29

Battery model

Fig. 4.30

SOC versus terminal voltage curve for Ni-MH battery cell

$$ SOC = \frac{{Capacity_{\hbox{max} } (Ah) - Ah\_used}}{{Capacity_{\hbox{max} } (Ah)}} $$
(4.3)

4.7.3.3 Propulsion System

An active load is used to model the demanded power from the propulsion system, which avoids the complicated modeling of motor and motor controller. The propulsion system is modeled as a controlled current source, drawing current from the system, as shown in Fig. 4.31. Various driving scenarios are translated to corresponding power requirements through a look-up table with respect to time. This relation data can be obtained from the vehicle system level simulations. The motor required current can thereby be obtained after divided by the DC bus voltage.
Fig. 4.31

Propulsion system model

4.8 Concept of Fuel Cell Plug-in HEV (FC-PHEV)

4.8.1 Fuel Cell-Hybrid Electric Vehicle Architecture

In the long term scenario, fuel cells represent one of the most appealing technologies for vehicle propulsion to further achieve high fuel efficiency, zero emissions, and low noise. Fuel cells are considered among the most promising alternative power sources, which can replace the conventional internal combustion engine (ICE). Compared to battery-powered electric vehicles (EVs), fuel cell vehicles (FCVs) have the advantage of longer driving range without a long battery charging time. In addition, compared to ICE vehicles, FCVs also depict comparatively higher energy efficiency and much lower emissions, due to direct conversion of free energy from the fuel into electric energy, without undergoing combustion. However, to fully achieve the potential energy savings of a fuel cell vehicle, it is important to recover the braking energy and ensure the operation of the FC system at maximum efficiency over the entire range of driving conditions encountered. This can be reached by a hybridization approach similar to gasoline-engine powered HEV. Furthermore, FC-HEVs present the advantages of cleaner and more efficient energy source, combined with the energy savings typical of EVs.

A typical power train of a fuel cell vehicle is as shown in Fig. 4.32. While most major automotive companies are investing in fuel cell vehicles, many challenges remain in getting fuel cell vehicles in the market. The major challenges include increasing fuel cell reliability, developing hydrogen infrastructure, improving on-board hydrogen storage capabilities, and overall cost reduction. One of the main research focuses is to develop a power control strategy and a power management system, which includes a fuel cell system, an energy storage system (ESS), and a suitable power electronic interface.
Fig. 4.32

Typical power train layout of FCHEV

References

  1. 1.
    M. Ehsani, Y. Gao, S. E. Gay, A. Emadi, Modern Electric, Hybrid Electric, and Fuel Cell Vehicles: Fundamentals, Theory and Design, 1st edn. (CRC Press, New York, 2004)Google Scholar
  2. 2.
    W.D. Jones, Putting electricity where the rubber meets the road. IEEE Spectr. 44(7), 18–20 (2007)CrossRefGoogle Scholar
  3. 3.
    S. Williamson, M. Lukic, A. Emadi, Comprehensive drive train efficiency analysis of hybrid electric and fuel cell vehicles based on motor-controller efficiency modeling. IEEE Trans. Power Electron. 21(3), 730–740 (2006)CrossRefGoogle Scholar
  4. 4.
    S. Imai, N. Takeda, Y. Horii, Total efficiency of a hybrid electric vehicle, in Proceedings of IEEE Power Conversion Conference, Nagaoka, Japan, Aug 1997, pp. 947–950Google Scholar
  5. 5.
    G. Maggetto, J. Van Mierlo, Electric and Hybrid Electric Vehicle: A Survey, IEE Seminar on Electric, Hybrid Electric, and Fuel Cell Vehicles (Durham, UK, 2005), pp. 1–11Google Scholar
  6. 6.
    S.S. Williamson A. Emadi, Comparative assessment of hybrid electric and fuel cell vehicles based on comprehensive well-to-wheels efficiency analysis. IEEE 54(3), 856–862 (2005)Google Scholar
  7. 7.
    F. An, D. Santini, Assessing tank-to-wheel efficiencies of advanced technology vehicles, in Proceedings SAE World Congress, Detroit, MI, March 2003Google Scholar
  8. 8.
    L. Eudy, J. Zuboy, Overview of advanced technology transportation, 2004 update, Technical Report on Advanced Vehicle Testing Activity, Energy Efficiency and Renewable Energy, U.S. Department of Energy, DOE/GO-102004-1849, Aug 2004Google Scholar
  9. 9.
    A. Burk, Present Status and Marketing Prospects of the Emerging Hybrid-Electric and Diesel Technologies to Reduce CO2 Emissions of New Light-Duty Vehicles in California (Institute of Transportation Studies, University of California, Davis) [Online]. Available: http://repositories.cdlib.org/itsdavis/UCD-ITS-RR-07-2
  10. 10.
    J. Burns, T. Cors, B. Knight, B. Thelen, Evaluating advanced automotive energy technologies: a multivariate mobility contribution metric. Int. J. Energy Techn. Policy 2(3), 262–271 (2004)CrossRefGoogle Scholar
  11. 11.
    S.S. Williamson, A. Emadi, Fuel cell vehicles: opportunities and challenges, in Proceedings of IEEE Power Engineering Society (PES) General Meeting, vol. 2, Denver, CO, June 2004, pp. 1640–1645Google Scholar
  12. 12.
    X. Li, S.S. Williamson, Comparative investigation of series and parallel hybrid electric vehicle (HEV) efficiencies based on comprehensive parametric analysis, in Proceedings of IEEE Vehicle Power and Propulsion Conference, Arlington, TX, Sept 2007Google Scholar
  13. 13.
    G. Rizzoni, L. Guzzella, B.M. Baumann, Unified modeling of hybrid electric vehicle drivetrains, AAAS 4(3), 246–257 (1999)Google Scholar
  14. 14.
    M. Amrhein, S. Member, P.T. Krein, Dynamic simulation for analysis of hybrid electric vehicle system and subsystem interactions, including power electronics. IEEE Trans. Veh. Technol. 54(3), 825–836 (2005)CrossRefGoogle Scholar
  15. 15.
    J. Wu, A. Emadi, M.J. Duoba, T.P. Bohn, Plug-in hybrid electric vehicles: testing, simulations and analysis, in Proceedings of IEEE Vehicle Power and Propulsion Conference, Arlington, TX, Sept 2007Google Scholar
  16. 16.
    F. Stodolsky, L. Gaines, C.L. Marshall, F. An, J.J. Eberhardt, Total Fuel Cycle Impacts of Advanced Vehicles, SAE Technical Paper Series, no. 724, 1999Google Scholar
  17. 17.
    N.R.E.L. NREL, Advanced Vehicle Simulator (ADVISOR) Documentation, http://www.ctts.nrel.gov/analysis/advisor_doc
  18. 18.
    T. Markel, A. Brooker, T. Hendricks, V. Johnson, K. Kelly, B. Kramer, M. O’Keefe, S. Sprik, K. Wipke, ADVISOR: a systems analysis tool for advanced vehicle modeling. J. Power Sources 110(2), 255–266 (2002)CrossRefGoogle Scholar
  19. 19.
    D. Boettner, M. Moran, Proton exchange membrane (PEM) fuel cell-powered vehicle performance using direct-hydrogen fueling and on-board methanol reforming. Energy 29(12–15), 2317–2330 (2004)CrossRefGoogle Scholar
  20. 20.
    J. Park, B. Raju, A. Emadi, Effects of An ultra-capacitor and battery energy storage system in a hybrid electric vehicle, Society of Automotive Engineers (SAE) Journal, Paper No. 2005-01-3452, 2005; and, in Proceedings of SAE 2005 Future Transportation Technology Conference, Chicago, IL, Sept 2005Google Scholar
  21. 21.
    V.H. Johnson, Battery performance models in ADVISOR. J. Power Sources 110(2), 321–329 (2002)CrossRefGoogle Scholar
  22. 22.
    C.C. Chan, The State of the Art of Electric, Hybrid, and Fuel Cell Vehicles. proc. IEEE 95(4), 704–718 (2007)CrossRefGoogle Scholar
  23. 23.
    A.P. Roussean, R. Ahluwalia, Q. Zhang, Energy storage requirements for fuel cell vehicles, in Proceeding of SAE World Congress, Detroit, MI, March 2004, Paper No. 2004-01-1302Google Scholar
  24. 24.
    H.Y. Cho, W. Gao, H. Ginn, A new power control strategy for hybrid fuel cell vehicles, in Proceedings of IEEE Workshop on Power Electronics in Transportation, Detroit, MI, Oct 2004, pp. 159–166Google Scholar
  25. 25.
    C. Liang, W. Qingnian, Energy management strategy and parametric design for fuel cell family sedan, in Proceedings of SAE Future Transportation Technology Conference and Exposition., Costa Mesa, CA, Aug 2001, Paper No. 2001-01-2506Google Scholar
  26. 26.
    H.S. Ahn, N.S. Lee, Power distribution control law for FCHEV-A fuzzy logic-based approach, in Proceedings of IEEE International Conference on Control and Automation, Budapest, Hungary, June 2005, pp. 486–0490Google Scholar
  27. 27.
    N.S. Lee, G.M. Jeong, H.S. Ahn, Improvement of fuel economy using fuzzy logic-based power distribution control strategy for a FCHEV, in Proceedings of IEEE International Conference on Computational Intelligence and Security, vol. 1, Guangzhou, China, 2006, pp. 891–894Google Scholar
  28. 28.
    J.S. Lai, D.J. Nelson, Energy management power converters in hybrid electric and fuel cell vehicles,in Proceedings of the IEEE, vol. 95, April 2007, pp. 766–777Google Scholar
  29. 29.
    G. Paganelli, Y. Guezennec, G. Rizzoni, Optimizing control strategy for hybrid fuel cell vehicles, in Proceedings of SAE International Congress, Detroit, MI, March 2002, Paper No. 2002-01-0102Google Scholar
  30. 30.
    P. Rodatz, G. Paganelli, A. Sciarretta, L. Guzzella, Optimal power management of an experimental fuel cell/supercapacitor-powered hybrid vehicle. Control Eng. Practice 13(1), 41–53 (2005)CrossRefGoogle Scholar
  31. 31.
    J. Liu, H. Peng, Control optimization for a power-split hybrid vehicle, in Proceedings of IEEE American Control Conference, Minneapolis, MN, June 2006, pp. 14–16Google Scholar
  32. 32.
    A. Schell, H. Peng, D. Tran, E. Stamos, C.C. Lin, Modeling and control strategy development for fuel cell electric vehicles. Annual Reviews Control 29, 159–168 (2005)CrossRefGoogle Scholar
  33. 33.
    K. Rajashekara, Power conversion and control strategies for fuel cell vehicles, in Proceedings of 29th Annual Conference of the IEEE Industrial Electronics Society, vol. 3, Roanoke, VA, Nov 2003, pp. 2865–2870Google Scholar
  34. 34.
    J.H. Hirschenhofer, D.B. Stauffer, R.R. Engleman, M.G. Klett, Fuel Cell Handbook, 4th edn. (Department of Energy, Federal Energy Technology Center, USA, 1998) DOE/FETC-99/1076Google Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  1. 1.Department of Electrical and Computer EngineeringConcordia UniversityMontrealCanada

Personalised recommendations