Improving fuel economy and performance of a fuel-cell hybrid electric vehicle (fuel-cell, battery, and ultra-capacitor) using optimized energy management strategy

Highlights

• The structure of Fuel-cell/Battery/Ultra-Capacitor is designed as FCHEV.

• A novel optimized fuzzy logic control is proposed for FCHEV supervisory system.

• The proposed strategy contributes to better performance in different aspects.

• Results are compared based on energy consumption, efficiency and storage level.

• The battery charge level changes less than 2% over 22 different driving cycles.

Abstract

Fuel-Cell System (FCS) is the primary energy supply of a Fuel-Cell Vehicle (FCV). Battery or Ultra-Capacitor (UC), as a secondary power source, is used along the FCS to improve the FCV’s power response. Battery and UC composition, as a hybrid power source presenting the term of Fuel-Cell Hybrid Electric Vehicle (FCHEV), provides the FCV with the advantages of high energy density and high dynamic response. The supervisory system of the FCHEV could be managed efficiently to exploit the benefits of battery and UC at the same time. As a matter of fact, in such a combination, the performance of the hybrid powertrain largely depends on how to distribute the requested power through different types of energy sources.

In this paper, we design the powertrain elements of an FCHEV in advance, with FCS/Battery/UC considerations. The energy management strategy (EMS) is achieved by presenting a novel power sharing method and by implementing an intelligent control technique constructed based on Fuzzy Logic Control (FLC). The control parameters are accurately adjusted by the genetic algorithm (GA) while considering targets and restrictions within a multi-objective optimization function over a combined city/highway driving cycle. This optimized supervisory system is examined by Advanced Vehicle Simulator (ADVISOR) to evaluate the performance of the proposed EMS over 22 different driving cycles and some specific performance tests. The results of simulation show that the presented strategy progressively affects the vehicle characteristics. Fuel economy enhancement, vehicle performance improvement, battery charge-sustaining capability, and optimal energy distribution are some of the significant outcomes achieved by the optimized FLC-based EMS.

Introduction

Fuel-Cell Vehicle (FCV) is known as an electric vehicle equipped with FCS [1]. Integrating FCS with battery or UC is a well-known method to mitigate FCS limitations. Battery/UC composition as a hybrid power source which presents the term of Fuel-Cell Hybrid Electric Vehicle (FCHEV), provides the FCV with the advantages of high energy density and high dynamic response. In such a combination, designing an optimal energy management strategy (EMS) plays a vital role in the success of the FCHEV supervisory system [2], [3].

There are various EMSs designed and optimized for the hybrid supervisory system [3], [4], [5], [6], [7]. Linear programming and PID controller [8], [9], [10], state flow algorithms and multiple operation mode control [11], [12], [13], [14], [15], dynamic programming techniques [16], [17], [18], fuzzy logic control (FLC) [14], [19], [20], [21], convex programming [22], model predictive control [23], [24], and optimal control theory [25], [26] are some of the applied strategies. To have an optimal EMS, we need both of control methods and optimization techniques. The EMS deals with hybrid power sources to meet commanded power whereas optimization procedure tries to have a more efficient power balance. In other words, considering powertrain condition, the requested power should be distributed by the EMS while achieving the best fuel economy and vehicle performance [14].

In the FCHEV configuration, the battery is a well-known secondary power source. In the recent studies, Ettihir et al. [26] proposed two adaptive EMSs to be used in the FCS/Battery supervisory system: hysteresis and optimal power splitting. The first strategy tried to keep the battery charge level around its reference value while the second one uses the FCS current as a control variable to distribute power between FCS and battery pack. These strategies were compared based on consumed hydrogen energy and battery energy in a sample cycle. In Ref. [27] three operation modes, including traction/braking/stopping, were presented with an FCS/Battery hybrid vehicle. Ensuring the feasibility of FCV power production, their proposed EMS limited the battery load while the battery charge occurred under traction and braking modes.

UC is another type of power source used in the hybrid configuration mainly due to its high power density [4]. UC plays a vital role in providing instantaneous power, particularly in acceleration and regenerative braking. In fact, its power density, durability, and efficiency in charge/discharge cycles give more advantages in comparison with battery and FCS [28], [29]. In the recent studies, Sami et al. [10] presented an EMS based on two main modes for an FCV integrated with UC. Despite the first mode in which the FCV operated with both FCS and UC, the UC was the sole power unit in the second operating mode in case of fuel limitation. During the first mode operation, a PI controller was applied to preserve the optimum performance of FCS and UC. Based on their experimental results, UC can meet load requirements in both modes. In Ref. [25], Li et al. employed an optimal control theory to find the optimal control laws used in their proposed hybrid configuration. The objective of this research was to minimize the hydrogen fuel consumption while considering FCS durability and charge level of UC. In the presented strategy, FCS provided more power which led to a smaller change in UC charge level.

All the strategies mentioned above, in general, tried to minimize FCS/Battery or FCS/UC energy consumption. Despite benefits of these configurations, there are some other distinctive features giving advantages to the FCHEV by considering FCS/Battery/UC composition. There are several techniques used in the literature to manage the power sources in this hybrid structure [29], [30], [31], [32], [33], [34]. A large amount of power density of the UC and energy density of the battery provide FCHEV with the opportunity to respond to high power and energy demands such as commanded power in acceleration or uphill [4]. Despite the mentioned benefits, this hybrid scheme makes the FCHEV powertrain more complex, and correspondingly it needs an advance EMS. In Ref. [30], the author presented an operation mode control for a typical FCHEV. Battery and UC charging and discharging modes occurred with a simple relation between load power and fuel-cell power. Equivalent consumption minimization strategy is one of the strategies used in the recent literature [29], [32], [34]. In the current studies to distribute the power demand to the FCS/Battery/UC hybrid tramway, a multi-mode strategy based on the equivalent consumption minimization strategy was proposed by the authors in Ref. [34] and Ref. [32]. Odeim et al. in Ref. [35] proposed a real-time strategy to minimize the hydrogen consumption and battery contribution based on an offline algorithm as a benchmark. In order to have battery current limited, authors in Ref. [33] presented an algorithm evaluated in a drive cycle to control the energy flux in the FCS/Battery/UC hybrid vehicle. On the other hand, considering the vehicle’s main targets, an optimization method should be employed to ensure the optimality of the proposed strategy during an indexed drive cycle. Having an optimized EMS, authors in Ref. [36] and Ref. [35] employed Multi-objective optimization method to minimize the defined cost functions while considering fuel economy and system durability. Moreover, in Ref. [35], GA was employed by the authors to find the best values for FCHEV control parameters, which led to improvement in the battery lifetime.

Whereas finding an optimal EMS for FCVs through conventional control techniques is a well-researched topic, there has been far less work on the still challenging tasks of optimal EMS designing problems, which faces FCHEVs. Having our experience in FCV power-train developing in hand, we try to propose an EMS to work out the issues associated with FCHEV power sharing as a constrained multi-objective problem while integrating fuel economy improvement and vehicle targets. In this paper, we are going to present a new optimal EMS based on an intelligent control method and power track control (PTC) technique. PTC helps to keep the battery charge level and Fuzzy Logic control, as an intelligent control method, determines the required power for FCS and UC. Another distinctive feature of this EMS is its optimization procedure which is entirely different in comparison with those applied in the literature. Despite the EMSs optimized within a cycle for better fuel economy, the control parameters of this EMS are tuned by the GA during a combined City/Highway driving cycle while considering vehicle constraints, initial conditions, and performance goals. Moreover, in order to approve the performance of the presented EMS, instead of a specific drive cycle used in the literature, we use 22 different driving cycle to compare different strategies. We also perform powertrain calculations as the primary step of FCHEV designing. The Battery/UC composition is designed as a secondary power unit of a typical FCV to meet the commanded power. While considering hybrid power units, firstly, a novel EMS is presented to share the requested power through the power sources as well as efficiently improving battery durability using power track control and FLC simultaneously. Secondly, the proposed EMS is to be optimized based on vehicle targets and fuel economy through a multi-objective optimization function and Genetic Algorithm (GA). This method of optimization is chosen to tune the parameters of FLC strategy based upon the results of a simulation in combined urban and highway drive cycles along with main performance tests. In this study, for the first time, key designing targets such as FCS and ESS efficiency, start-up conditions, final energy storage level, and acceleration requirements are considered during the evaluation of the defined objective function to have optimized EMS in both vehicle performance and fuel economy. Finally, to examine the EMS thoroughly, rather than a specific driving cycle, the proposed strategy is to be evaluated in more than 20 driving cycles by using ADVISOR, the widely-known advanced vehicle simulation software.

The rest of the paper is structured as follows. The section of FCHEV Configuration and Calculations illustrates powertrain of the FCHEV and power source calculations. In section EMS Modelling and Simulation, the proposed control strategy is thoroughly investigated. We introduce our FLC strategy in this section. EMS Optimization section determines the optimization algorithm and fitness function. Defining EMS targets over a driving cycle is done in this part. Results’ analysis is given in section Simulation Results and Discussion. Finally, the conclusion and future works are presented in the last chapter.