Search

The recent advancements in artificial intelligence have promoted deep reinforcement learning (DRL) as the preferred method for developing energy management strategies (EMSs) for fuel cell vehicles (FCVs). However, the development of DRL-based EMSs is a time-consuming process, requiring repetitive training when encountering different vehicle types or learning tasks. To surmount this technical barrier, this paper develops a transferable EMS rooted in heterogeneous deep transfer reinforcement learning (DTRL) across both FCV types and optimization tasks. Firstly, a simple source EMS based on the soft actor-critic (SAC) algorithm is pre-trained for a fuel cell sedan, solely focusing on hydrogen saving. After that, a heterogeneous DTRL framework is developed by integrating SAC with transfer learning, through which both heterogeneous deep neural networks and experience replay buffers can be transferred. Subsequently, the source EMS is transferred to the target new EMS of a fuel cell bus (FCB) to be reused, with additional consideration of the fuel cell (FC) longevity. Experimental simulations reveal that the heterogeneous DTRL framework expedites the development of the new EMS for FCB by 90.28 %. Moreover, the new EMS achieves a 7.93 % reduction in hydrogen consumption and suppresses FC degradation by 63.21 %. By correlating different energy management tasks of FCVs, this article both expedites the development and facilitates the generalized application of DRL-based EMSs. • A heterogeneous DTRL framework is designed by integrating TL with SAC algorithm. • A transferable EMS across FCV types and learning tasks is proposed based on DTRL. • Both heterogeneous DNNs and experience replay buffer are transferred and reused. • Fuel economy performance and fuel cell longevity are collaboratively optimized. • The proposed EMS shows great potential for real-time application via online test. [ABSTRACT FROM AUTHOR]

The propulsion and power systems are critical to the performance of unmanned aerial vehicles (UAVs). The rapid development of electrified propulsion systems powered by proton exchange membrane fuel cells (PEMFCs) has driven technological innovation of propulsion and power systems for UAVs. In this work, a 1.5 kW PEMFC integrated propulsion system with direct airflow intake design (Beihang Hydrogen-1) is developed and verified by flight test on a UAV with a wingspan of 3.6 m. The propulsion system can directly use the airflow behind the propeller to feed the PEMFC cathode for cooling and reaction and the power generated by the PEMFC is used to drive the propeller. A DC-DC module and a lithium battery pack are also added to the propulsion system to stabilize the output voltage and increase the instantaneous power for takeoff, transition, and landing stages. The "dynamic distribution" energy management strategy for the propulsion system is proposed to increase the system dynamic response ability. The detailed data of the PEMFC integrated propulsion system during the flight are measured and analyzed. The flight test continued for 18 min, with an average cruising altitude and an average cruising speed of 101 m above sea level and 23 m/s, respectively. During the cruise stage, the average PEMFC power is 1688 W, with an average single-cell voltage of 0.65 V. The average charging power of PEMFC for the lithium battery is 41 W. The successful flight test verifies the feasibility of the design of the PEMFC integrated propulsion system, provides valuable data for the following optimal design and flight test of the PEMFC integrated propulsion systems, and opens up a new direction for the development and application of PEMFC-powered electrified propulsion systems. • A 1.5 kW PEMFC integrated propulsion system with direct airflow intake design is developed. • The "dynamic distribution" energy management strategy is proposed for the PEMFC integrated propulsion system. • The first flight verification of the PEMFC integrated propulsion system is conducted based on a UAV. • The average PEMFC power is 1688 W with an average charging power of 41 W for the battery in the cruise stage of the flight test. • The PEMFC integrated propulsion system shows excellent dynamic response ability and energy conversion efficiency. [ABSTRACT FROM AUTHOR]

Generalization to unseen environments is still a challenge for deep reinforcement learning (DRL)-based energy management strategies (EMSs). This paper proposes a real-time EMS with high generalization for a light-duty hybrid electric vehicle (HEV) from two perspectives: enhancing the generalization of the DRL algorithm and improving the accuracy of application scenario representation in the training environment. The enhanced DRL algorithm named ATSAC can adjust the update frequency and learning rate of SAC automatically to improve the generalization. With the advancement of naturalistic driving big data (NDBD) and machine learning, a specific training cycle is synthesized based on NDBD to reflect an urban-suburban real-world driving scenario more accurately. By the comprehensive comparison with SAC and TD3 based EMSs applied to unseen driving scenarios, the proposed algorithm achieves significant improvement in computational efficiency, optimality, and generalization. The results show that the computational efficiency of ATSAC is increased by 52.32% compared to SAC. The negative total reward (NTR) of ATSAC is decreased by 18.22% and 69.81% compared to SAC and TD3, respectively. Further analysis shows that the EMS trained through the synthetic driving cycle obtains 18.37% lower NTR than WLTC which demonstrates that the synthetic method can reflect the state transition probability of real-world driving scenarios better than WLTC. • A novel DRL algorithm with high generalization is researched for energy management. • An SODC is synthesized through big data and machine learning as the training cycle. • Iteration dropout can avoid overfitting to improve the generalization of DRL. • Adaptive learning rate can balance the exploration and exploitation. [ABSTRACT FROM AUTHOR]

The power transients caused by switching from drive mode to brake mode in fuel cell hybrid electric vehicles (FCHEV) can result in significant degradation cost losses to the fuel cell. To address this issue, this study proposes a self-learning Markov prediction algorithm based aging-oriented gradient drop power control strategy. First, a real-time self-learning Markov predictor (SLMP) based on the traditional offline training Markov improvement is designed to predict the demand power and combined with the sequential quadratic programming (SQP) optimization algorithm to solve for the inner optimal demand power based on its global cost function minimization characteristic. On this basis, the fuel cell gradient drop power (FGDP) strategy is proposed to optimize the operating state of the vehicle powertrain under vehicle mode switching. This involves establishing a power gradient drop step based on considering the fuel cell hydrogen consumption cost and its lifetime degradation cost to further obtain the outer fuel cell demand power at the optimal step. And three execution modes are designed to trigger the FGDP strategy. Finally, by combining the above efforts, the SLMP-FGDP optimization control strategy is constructed. The numerical verification and hardware in loop experiments results show that the proposed improved SLMP can predict the vehicle demand power more accurately. Compared with the non-FGDP system, the SLMP-FGDP strategy can effectively near-eliminate the fuel cell power transient due to any braking scenario, thus effectively controlling the fuel cell lifetime degradation cost in a lower range and realizing a reduction of up to 52.21% of the fuel cell usage costs without significantly sacrificing the hydrogen fuel economy. • The fuel cell gradient drop power strategy is formulated to suppress power transients. • A self-learning Markov predictor for real-time prediction is proposed. • Balancing the multi-objective cost function to minimize the global cost. • The numerical validations and HIL are conducted to validate the proposed strategy. [ABSTRACT FROM AUTHOR]

This paper proposes an improved "prediction + optimal control" method for energy management in hybrid electric vehicles equipped with planetary gears. A differentiable predictor and a differentiable optimal controller are developed using supervised learning and reinforcement learning approaches, respectively. Three training steps are performed for the initial predictor, the optimal controller, and the final predictor. This method improves the traditional energy management predictive optimal control approach by incorporating an additional step of retraining the differentiable predictor. This adjustment ensures that the predictor does not blindly improve its performance based on evaluation criterion irrelevant to energy management, which was commonly used in previous studies. Instead, it focuses on enhancing the overall performance of energy management under the "prediction + optimal control" framework. The approach introduced in this paper is compared with the globally optimal dynamic programming results and traditional predictive optimal control methods on the Next Generation Simulation (NGSIM) data. Our method outperforms traditional approaches in energy management on both the training dataset and the test dataset. This further illustrates that the conventional practice of presumptuously optimizing predictors in "prediction + optimal control" methods can be improved using the proposed method. • Optimizing the Energy Efficiency of Hybrid Electric Vehicles with Planetary Gears. • Incorporating Energy Minimization Objectives into Predictor Evaluation Criteria. • Both the Predictor and the Optimal Controller are Differentiable. • The Optimal Controller is Obtained by Reinforcement Learning Method. • The Improved Method is Trained and Tested on the NGSIM Dataset. [ABSTRACT FROM AUTHOR]

Energy management strategies (EMSs) based on speed prediction are widely used in fuel cell hybrid electric vehicles (FCHEVs). However, two fundamental issues must be addressed: short-term speed prediction and power allocation across multi-scales. To address these issues, a hierarchical EMS for FCHEV based on multi-scale information fusion is developed in this study. In this strategy, a novel speed predictor that incorporates visual information and the motion states of a vehicle is used to predict the short-term speed. At the global level, the average power demand of each road segment is predicted based on real-time traffic information, which is used to plan the global reference state of charge (SOC) trajectory before departure. At the local level, fuzzy rules are utilized to determine the short-term reference SOC. The short-term optimal co-state of the Pontryagin minimal principle-based strategy is then updated. The results demonstrate that the proposed visual information speed predictor (VISP) enhances the prediction accuracy by 12%–32.5% and leads to a 3.36% reduction in the total cost compared with the EMS utilizing the conventional predictor. In addition, the proposed EMS makes the terminal SOC close to the target value and reduces the total cost by 20.33% compared to the benchmark strategy. • A hierarchical energy management strategy based on multi-scale information fusion is proposed. • A novel speed predictor that incorporates visual information and vehicle motion states is designed. • Global reference SOC trajectory method based on the real-time temporal and spatial traffic information of road segments is proposed. • Adaptive fuzzy rules are used to determine the shot-term reference SOC for each horizon. • Real urban driving cycle is used to verify the performance of the energy management strategies. [ABSTRACT FROM AUTHOR]

Large-scale power-to‑hydrogen (P2H) stations with multi-stack configurations, are emerging as valuable flexible resources for the power grid. The energy management strategy (EMS) determines multi-stack operation statuses. Nonetheless, existing EMS focus on production without adequately addressing the implications for grid-side power factor (PF) and potential security concerns. This paper addresses this limitation by presenting a model that characterizes the PF of a multi-stack P2H system across varying operation statuses defined by current and temperature. Through this model, we highlight a clear trade-off between the PF constraint and production targets in multi-stack scheduling. Subsequently, we introduce an improved EMS framework for multi-stack P2H that seeks a balance between PF, production, and security. This EMS is organized as a two-layer execution structure to guarantee control accuracy and tractability, which includes a model-based robust multi-stack scheduling programming and a rule-based real-time increment correction algorithm in series. Case studies compare multi-stack scheduling strategies under the proposed EMS with the traditional production-oriented strategy. The effectiveness of the extended PF and security dimensions is verified to comprehensively improve the responsiveness to power instructions. Furthermore, we outline five representative cluster-level scheduling strategies aligned with different load scenarios, offering insights for practical industrial implementations. • A general framework of energy management strategy for large-scale power-to-hydrogen (P2H) stations. • The tradeoff between the power factor constraint and the production target in determining multi-stack scheduling strategies. • A two-layer EMS is executed by a model-based hour-ahead robust scheduling and a rule-based real-time increment correction. • Outlining five cluster-level scheduling strategies aligned with various load scenarios of multi-stack P2H stations. [ABSTRACT FROM AUTHOR]

Deep reinforcement learning (DRL) is now a research focus for the energy management of fuel cell vehicles (FCVs) to improve hydrogen utilization efficiency. However, since DRL-based energy management strategies (EMSs) need to be retrained when the types of FCVs are changed, it is a laborious task to develop DRL-based EMSs for different FCVs. Given that, this article introduces transfer learning (TL) into DRL to design a novel deep transfer reinforcement learning (DTRL) method and then innovatively proposes an intelligent transferable energy management framework between two different urban FCVs based on the designed DTRL method to achieve the reuse of well-trained EMSs. To begin, an enhanced soft actor-critic (SAC) algorithm integrating prioritized experience replay (PER) is formulated to be the studied DRL algorithm in this article. Then, an enhanced-SAC based EMS of a light fuel cell hybrid electric vehicle (FCHEV) is pre-trained by using massive real-world driving data. After that, the learned knowledge stored in the FCHEV's well-trained EMS is captured and then transferred into the EMS of a heavy-duty fuel cell hybrid electric bus (FCHEB). Finally, the FCHEB's EMS is fine-tuned in a stochastic environment to ensure adaptability to real driving conditions. Simulation results indicate that, compared to the state-of-the-art baseline EMS, the proposed DTRL-based EMS accelerates the convergence speed by 91.55% and improves the fuel economy by 6.78%. This article contributes to shortening the development cycle of DRL-based EMSs and improving the utilization efficiency of hydrogen energy in the urban transport sector. • An enhanced SAC algorithm combined with standard SAC and PER is formulated. • A novel DTRL method is designed by integrating enhanced SAC with transfer learning. • A transferable energy management framework is proposed based on the DTRL method. • PER buffer and SumTree together with DNN parameters are transferred and reused. • Both fuel economy and real-time performance are evaluated through online testing. [ABSTRACT FROM AUTHOR]

Proton exchange membrane fuel cells (PEMFCs) will greatly shorten their lifespan due to platinum (Pt) catalyst degradation during operation. This paper proposes an optimization-based energy management method considering Pt degradation, which is an improvement over the traditional strategy that only focuses on fuel optimization to consider both the minimal fuel and the minimum fuel cell life decay. Firstly, a one-dimensional (1D) Pt degradation model is established to comprehend how various voltage situations affect Pt deterioration. Then, various strategies to suppress Pt degradation are designed using Pontryagin's minimum principle (PMP) optimization algorithm in light of the influence analysis results, and the effects of the PMP algorithm under different strategies are tested on the hardware-in-the-loop (HIL) simulation platform. The results demonstrate that the performance of the PMP algorithm in real-time strategy is extremely near to the global optimal solution generated by the offline dynamic programming (DP) algorithm. After adding the tendency to limit high potential and voltage variation in the PMP algorithm, hydrogen consumption increases by only 2%. In comparison, the stack's degradation is decreased by nearly 50%, considerably extending the stack's service life and reducing the system's comprehensive use cost. • Prior literature verified the construction of a one-dimensional Pt degradation model. • Frequent voltage fluctuations and high voltage will exacerbate degradation via model analysis. • Developed two optimization-based strategies for reducing stack performance degradation. [ABSTRACT FROM AUTHOR]

Grid-Connected Wind Turbine is a comprehensive dynamic system comprising cross-coupling characteristics in various time scales. This paper aims to construct a real-time semi-physical co-simulation of Variable-Speed-Variable-Pitch Wind Turbines (VSVP-WTs) equipped with a permanent magnet synchronous generator (PMSG). As for the deployment configuration, software such as OpenFAST and LabVIEW are used for system modeling, while the hardware, mainly NI devices, is adopted for real-time operation and strategy implementation. The physical behaviors of energy conversion are discussed separately, especially on aeroelastic dynamics and electronic power switching. Additionally, operational strategy for the whole production is established through an embedded industrial controller, which also serves as a bi-directional interaction path. An Ethernet network-based communication interface is constructed as an information channel in which a discrete model is extracted and discussed. The synchronization and stability among hardware are guaranteed with a proper period. Several cases demonstrate detailed dynamics of the whole WT, brought on by command adjustment, environmental factors, and power system faults. The proposed platform would be considered a critical stage in digital twin application of engineering. • An aero-elastic–electrical hybrid real-time platform is constructed for PMSG-WT. • OpenFAST-NI HIL co-simulation experimental is implemented via stable communication. • A combination of mechanical–electrical operational strategy is established for power regulation. • Accuracy is guaranteed through experimental cases with three kinds of power scenarios. [ABSTRACT FROM AUTHOR]

• Dynamic transmission efficiency model of planetary coupling transmission system. • Dynamic efficiency optimization strategy considering fatigue life of powertrain. • Optimized vibration-induced energy management strategy designed and tested. This paper presents a comprehensive dynamic efficiency model for the overall powertrain efficiency that considers the internal combustion engine and electric motor generators in depth as well as a planetary coupling transmission system of PHEVs example. In particular, the dynamic forces of planetary coupling transmission system with an effect on the dynamic transmission efficiency are taken into consideration. Then a dynamic efficiency optimization strategy is proposed to improve the fuel economy, and an improved dynamic efficiency optimization strategy, consisting in adding to the dynamic efficiency optimization strategy a penalty coefficient, is applied to reach the trade-off between energy consumption and fatigue life of planetary coupling transmission system. Eventually, the dynamic efficiency optimization strategy and improved dynamic efficiency optimization strategy are validated by hardware-in-the-loop experiments. The main contribution of this study is to explore a novel way to optimize the comprehensive dynamic efficiency of whole powertrain, and also improve ride comfort by avoiding planetary coupling transmission system working in the inefficient resonance region. [ABSTRACT FROM AUTHOR]

Improving plug-in hybrid electric vehicles (PHEVs) fuel economy requires a proper energy management strategy (EMS). Efforts to enhance the energy-saving performance of predictive EMSs have concentrated on advanced speed prediction methods. However, the impact of battery models on the predictive EMS hasn't been investigated. This paper aims to fill the research gap by suggesting a model predictive control-based (MPC) EMS framework that uses a hybrid speed predictor. Firstly, six control-oriented battery electro-thermal-aging models with various terminal voltage and temperature simulation accuracies have been developed and verified. Secondly, it also examines the effects of different battery models on MPC-based EMS through quantitative analysis, including battery dynamics, calculation time, and resultant operating costs, which leads to the suggestions of model selection for the design of the EMS under low- and room-temperature driving scenarios. Finally, a novel hybrid speed prediction model is proposed, where the historical speed is decomposed into strongly periodic intrinsic mode functions (IMFs) by variational mode decomposition (VMD), and backpropagation (BP) neural network is utilized to learn the feature parameter and mapping relationship of each IMF. In addition, a case study is conducted by applying the proposed speed prediction model in an MPC-based EMS method. The simulation results highlight that the proposed hybrid speed prediction model can achieve preferable speed prediction. The operating cost errors (compared with the MPC-based EMS with 100% speed prediction) are reduced to 0.4% and 0.98% at −20 °C and 25 °C driving scenarios, respectively. • Control-oriented battery electro-thermal-aging models for a wide temperature range. • Analyze the impact of battery modeling accuracy on MPC-based EMS for the first time. • A novel speed prediction model is proposed combining VMD and BP neural networks. • The effectiveness of the proposed speed prediction method is verified. [ABSTRACT FROM AUTHOR]

The powertrain system of modern PEMFC based hybrid tramways typically contains a PEMFC system and a hybrid energy storage subsystem when combing a lithium-ion battery (LIB) modules with a supercapacitor (SC) bank. Based on the detailed analysis of stochastic uncertainties in tramway operation, a suboptimal real-time power sharing strategy considering operation uncertainties as well as fuel economy and system durability is proposed in this paper. The proposed energy management strategy consists of three modules, namely the fundamental real-time penalty power sharing module, the fuzzy-logic based differential power compensation module, and the Rainflow-based predictive SOC balancing module. Firstly, suboptimal real-time power sharing among different energy sources is achieved in the fundamental real-time penalty power sharing module. Secondly, a fuzzy-logic based differential power compensation module is designed to achieve the performance degradation balancing between PEMFCs and LIBs. Furthermore, a Rainflow-based predictive SOC balancing module is developed to realize adaptive updating concerning key parameters of the above two modules based on historical SOC information identification of SC subsystem and enhance the robustness to stochastic uncertainties. Detailed simulation results demonstrate that the proposed energy management strategy can guarantee operation stabilization of PEMFC based hybrid topologies throughout the simulated driving cycle. The influence of the proposed energy management strategy on the service life of the PEMFC subsystem and fuel economy of hybrid tramway is discussed in detail. Finally, the proposed energy management strategy with optimized PEMFC and HESS both decoupled topology is verified to be more suitable for PEMFC-based hybrid tramway applications with minimum equivalent hydrogen consumption and performance degradation balancing among hybrid energy sources, compared with other reductant hybrid configuration-based energy management strategies. [ABSTRACT FROM AUTHOR]

Reinforcement learning is a new research hotspot in the artificial intelligence community. Q learning as a famous reinforcement learning algorithm can achieve satisfactory control performance without need to clarify the complex internal factors in controlled objects. However, discretization state is necessary which limits the application of Q learning in energy management for hybrid electric bus (HEB). In this paper the deep Q learning (DQL) is adopted for energy management issue and the strategy is proposed and verified. Firstly, the system modeling of bus configuration are described. Then, the energy management strategy based on deep Q learning is put forward. Deep neural network is employed and well trained to approximate the action value function (Q function). Furthermore, the Q learning strategy based on the same model is mentioned and applied to compare with deep Q learning. Finally, a part of trained decision network is analyzed separately to verify the effectiveness and rationality of the DQL-based strategy. The training results indicate that DQL-based strategy makes a better performance than that of Q learning in training time consuming and convergence rate. Results also demonstrate the fuel economy of proposed strategy under the unknown driving condition achieves 89% of dynamic programming-based method. In addition, the technique can finally learn to the target state of charge under different initial conditions. The main contribution of this study is to explore a novel reinforcement learning methodology into energy management for HEB which solve the curse of state variable dimensionality, and the techniques can be adopted to solve similar problems. [ABSTRACT FROM AUTHOR]

The traditional fuel locomotive is the primary type of locomotive currently in operation on non-electrified railways; however, it presents certain disadvantages including a low efficiency and high fuel consumption. Therefore, in this study, a multimode hybrid locomotive configuration scheme is designed to improve the system efficiency and reduce fuel consumption during locomotive operation; further, the power flow state under different modes is analyzed, the mathematical model of the hybrid locomotive is established, and a system optimal efficiency calculation method is developed. Moreover, an energy management strategy based on the optimal system efficiency and with a hierarchical architecture is proposed. In the upper layer of this proposed energy management strategy, the optimal efficiency of all operating points and operational state of each component are determined offline using the optimal efficiency calculation method. The lower layer selects and coordinates the mode online by identifying the condition of the wheel and distributes the torque of each power source and the state of each transmission system component. The simulation results indicate that the fuel economy of the proposed energy management strategy is improved by 27.22% compared with that of the traditional fuel locomotive, and the fuel economy is only 7.21% lower than the global optimization result obtained via dynamic programming. In addition, no frequent clutch switching is observed under this strategy, and the electric motor does not operate under non-rated conditions for prolonged periods; these advantages ensure the rationality of control and the reliability of component operation. Finally, the results of a hardware-in-the-loop simulation test confirm that the proposed energy management strategy demonstrates good real-time performance. • A hybrid-electric locomotive configuration with multiple modes is designed. • A system efficiency calculation method is proposed. • An optimal system-efficiency-based energy management strategy is developed. • The proposed strategy ensures real-time performance and fuel economy. • The strategy ensures rationality of mode selection and reliability of components. [ABSTRACT FROM AUTHOR]

The decay process of fuel cell (FC) and battery is highly inconsistent. The existing research focuses on the energy management strategy (EMS) aiming at minimizing the lifespan loss of a single power source. However, this EMS cannot guarantee the optimal overall durability of dual power source systems. Based on this, this paper proposes an EMS for lifespan decay coordination of dual power sources for fuel cell vehicle (FCV). Firstly, a continuous characterization model for FC lifespan decay was developed. Secondly, the influence of control parameters such as FC response speed, filtering order, equivalent consumption minimum strategy (ECMS) equivalent factor on dual power source decay rate is analyzed. Then, the proposed energy management strategy (PEMS) is based on the cooperative control of lifespan decay of two power sources based on the condition identification. Finally, the advantages of PEMS are verified by hardware in-loop simulation. The results show that: under the working conditions of CWC1 and CWC2, compared with ECMS, PEMS reduces the difference of lifespan decay rate of dual power source by 25.62 times and 32.25 times, respectively, and the lifespan decay of fuel cell and power battery tends to be consistent. It is proved that PEMS can realize the decay cooperation of dual power source of FCV under different characteristic working conditions, and significantly improve the overall durability of dual power source system. • A continuous characterization model for FC lifespan decay was developed. • Prove the influence of control parameters on the decay rate of dual power source. • Propose an EMS based on the lifespan decay synergy of dual power source of FCV. [ABSTRACT FROM AUTHOR]

• An A3C-based EMS is proposed to reduce the total operation cost of the FCHEB. • Multi-process parallel computation is employed to achieve the A3C-based EMS. • EMSs based on single-agent A2C and multi-thread A3C are adopted as baselines. • The proposed EMS is evaluated by training and testing using different cycles. • Both adaptability and computational efficiency are evaluated by online testing. Deep reinforcement learning (DRL) has become the mainstream method to design intelligent energy management strategies (EMSs) for fuel cell hybrid electric vehicles with the prosperity of artificial intelligence in recent years. Conventional DRL algorithms are suffering from low sampling efficiency and unsatisfactory utilization of computing resources. Combined with distributed architecture and parallel computation, DRL algorithms can be more efficient. Given that, this paper proposes a novel distributed DRL-based energy management framework for a fuel cell hybrid electric bus (FCHEB) to shorten the development cycle of DRL-based EMSs while reducing the total operation cost of the FCHEB. To begin, to make full use of the limited computing resources, a novel asynchronous advantage actor-critic (A3C)-based energy management framework is designed by innovatively integrating with the multi-process parallel computation technique. Then, a promising EMS considering the extra operation cost caused by fuel cell degradation and battery aging is designed based on this novel framework. Furthermore, EMSs based on a conventional DRL algorithm, advantage actor-critic (A2C), and another conventional distributed DRL framework, multi-thread A3C, are employed as baselines, and the performance of the proposed EMS is evaluated by training and testing using different driving cycles. Simulation results indicate that compared to EMSs based on A2C and multi-thread A3C, the proposed EMS can efficiently accelerate the convergence speed respectively by 87.46% and 88.92%, and reduce the total operation cost respectively by 44.83% and 41.19%. The main contribution of this article is to explore the integration of multi-process parallel computation in a distributed DRL-based EMS for a fuel cell vehicle for more efficient utilization of hydrogen energy in the transportation sector. [ABSTRACT FROM AUTHOR]

There are several advantages in keeping motor drive DC input voltage stable around its nominal value especially when it comes to minimize losses. This paper deals with the stability enhancement of the motor drive DC input voltage of an electric vehicle with on-board hybrid energy storage system. On one hand, the natural voltage variation at the output battery pack can be avoided by using a DC/DC converter connected between the battery and the motor drive. On the other hand, the incorporation of supercapacitors in small urban electric vehicles enables handling power peaks thus reducing the battery current root-mean-square value, which in turn increases the battery lifetime. Several topologies can be considered for the coupling of supercapacitors with the vehicle energy system. In this paper, three topologies are studied: battery-only, direct hybrid coupling, and active parallel hybrid coupling of batteries and supercapacitors. A reduced-scale power level hardware-in-the-loop test-bench has been built to analyze the performance of the hybrid topologies under the ARTEMIS driving cycle. Experimental results show the effectiveness of the active parallel configuration controlled by an improved energy management strategy that dynamically regulates the supercapacitors state-of-charge. The analysis performed demonstrates that the use of this configuration coupled with an improved management strategy can increase the power transferred by 80% compared with a battery-only configuration, and by 40% compared with direct hybrid coupling or active parallel coupling configuration with a traditional management strategy, keeping the voltage stability of the DC Link. [ABSTRACT FROM AUTHOR]

Plug-in hybrid electric vehicle (PHEV) is one of the most promising products to solve the problem about air pollution and energy crisis. Considering the characteristics of urban bus route, maybe a fixed-control-parameter control strategy for PHEV cannot perfectly match the complicated variation of driving conditions, and as a result the ideal vehicle fuel economy would not be obtained. Therefore, it is of great significance to develop an adaptive real-time optimal energy management strategy for PHEV by taking the segment characteristics of driving cycles into consideration. In this study, a novel energy management strategy for Plug-in hybrid electric bus (PHEB) is proposed, which optimizes the equivalent factor (EF) of each segment in the driving cycle. The proposed strategy includes an offline part and an online part. In the offline part, the driving cycles are divided into segments according to the actual positions of bus stops, the EF of each segment is optimized by linear weight particle swarm optimization algorithm with different initial states of charge ( SOC ). The optimization results of EF are then converted into a 2-dimensional look up table, which can be used to make real-time adjustments to online control strategy. In the online part, the optimal instantaneous energy distribution is obtained in this hybrid powertrain. Finally, the proposed strategy is verified with simulation and hardware in the loop tests, and three kinds of commonly used control strategies are adopted for comparison. Results show when the initial SOC is 90%, the fuel economy with the proposed strategy can be improved by 15.93% compared with that of baseline strategy, and when the initial SOC is 60%, this value is 16.02%. The proposed strategy may provide theoretical support for control optimization of PHEV. [ABSTRACT FROM AUTHOR]

Model predictive control (MPC) can effectively solve online optimization issues, even with various constraints, when maintained at high robustness. Considering the energy management issue of plug-in hybrid electric bus (PHEB) as a constrained nonlinear optimization problem, a strategy based on stochastic model predictive control (SMPC) is put forward and verified in this paper. Firstly, Markov Chain Monte Carlo Method (MCMC) is adopted to forecast velocity sequences at every current state, in the form of multi scale single step (MSSS), with post-processing algorithms to moderate fluctuations of the prediction results like average filtering, quadratic fitting, and the like. The offline simulation results show that the optimization can effectively improve the predictive accuracy, make the following energy management feasible and reduce the fuel consumption by 1.9%. Then the SMPC-based energy management strategy is proposed. In order to prevent the driving cycle state deficiencies from interrupting the prediction for practical application, a state reconstitution method is constructed accordingly. Besides, the predictive steps are made time-varying by an online accuracy estimation method and a corresponding threshold to maintain the accuracy of forecast. Finally, the hardware-in-the-loop (HIL) experiments are conducted and the results show that the SMPC-based strategy is reasonable and the fuel consumption decreases by 3.9% further with variable predictive steps than that of fixed ones. In summary, this paper illustrates an effective SMPC-based methodology for energy management for PHEB, and techniques like MSSS prediction with post-processing, state reconstitution method, online accuracy estimation can be adopted to solve similar problems. [ABSTRACT FROM AUTHOR]

This paper proposes a simple and easily optimizable mathematical representation of an energy management strategy (EMS) for the hybrid energy storage system (HESS) in EV. The power of each device in the HESS is provided as a continuous function of load power called γ . Two strategies based on the proposed method, one incorporating fixed coefficients of the γ function (GBS) and one with coefficients optimized by a genetic algorithm (GAS) in real-time using a backward time window, are tested and compared to the rule-based strategy (RBS) and battery storage system. The calculations are made for an electric car with a LiFePO 4 battery-supercapacitor HESS. The analyzed parameters are: energy consumption, RMS and maximum current rates of the battery, and the cycle cost of an EV with HESS and a battery-powered EV. The analysis is made in dependence on drive cycle speed and an internal resistance of the battery module. The obtained results show that the GBS and the GAS are able to reduce the RMS current rate by 40% in the NEDC in comparison to battery-powered EV, as well as that maximum current rates do not exceed nominal values. The GAS aims at the minimization of energy consumption. It obtains best results in low speed cycles. [ABSTRACT FROM AUTHOR]

• Expert demonstration data of hybrid electric vehicle energy management are represented by domain adaptive meta-learning. • By embedding expert demonstration model, an apprenticeship-reinforcement learning framework is established to improve fuel economy. • It is verified under three hybrid vehicle models with different powertrain architectures and several driving cycles. • Compared to traditional reinforcement learning, the apprenticeship-reinforcement learning can effectively improve the fuel emergency and CO 2 emissions. Deep reinforcement learning (DRL) is a potential solution to develop efficient energy management strategies (EMS) for hybrid electric vehicles (HEV) that can adapt to the changing topology of electrified powertrains and the uncertainty of various driving scenarios. However, traditional DRL has many disadvantages, such as low efficiency and poor stability. This study proposes an apprenticeship-reinforcement learning (A-RL) framework based on expert demonstration (ED) model embedding to improve DRL. First, the demonstration data, calculated by dynamic programming (DP), were collected, and domain adaptive meta-learning (DAML) was used to train the ED model with the adaptive capability of working conditions. Then combined apprenticeship learning (AL) with DRL, and the ED model was used to guide the DRL to output action. The method was validated on three HEV models, and the results show that the training convergence rate increases significantly under the framework. The average increase that the apprenticeship-deep deterministic policy gradient (A-DDPG) based method applied to three HEVs achieved was 34.9 %. Apprenticeship-twin delayed twin delayed deep deterministic policy gradient (A-TD3) achieved 23 % acceleration in the power-split HEV. Because A-DDPG's EMS is more forward-looking and can mimic ED to some extent, the frequency of engine operation in the high-efficiency range has increased. Therefore, A-DDPG can improve the fuel economy of the series hybrid electric bus (HEB) by 0.2–2.7 %, and improvements averaged to about 9.6 % in the series–parallel HEV while maintaining the final SOC. This study aims to improve the sampling efficiency and optimal performance of EMS-based DRL and provide a basis for the design and development of vehicle energy saving and emission reduction. [ABSTRACT FROM AUTHOR]

• A general seven-layer smart city energy management strategy structure based on the cooperation between smart homes is proposed. • The key goals are flattening the smart homes power profiles and reducing their electricity bills. • This strategy supports a free charging of electric vehicles. • This energy management can be applied on any smart city equipped with a flexible number of renewable energy sources and plug-in electric vehicles. The notion of smart city is based on using both bidirectional power and data flows. For that, designing a persuasive control model for power flow becomes a big and crucial part in a smart city to optimize the power balance between production and consumption. So, a seven-layer smart city energy management strategy (SLSCEMS) for multiple home energy cooperation is presented in this paper. The optimised aims of this strategy are smoothing the smart homes power demand profiles, reducing the electricity bills (E.B), and gaining a total free charging of Plug-in Electric Vehicles (PEVs). This approach has been designed as hierarchic local and global layers. The first one is divided into three layers that aim to transfer the energy between each smart home and its own Renewable Energy sources (RES) and PEVs. The second one is split into four layers that aim to transfer the energy between each smart home and its PEVs, the neighboring homes and their RES, and the smart grid. All optimised layers are based on Particle Swarm Optimization (PSO) algorithm. The proposed strategy is evaluated in a city that contains one hundred homes classified into five categories, each category is designated by its power profile and its flexible number of RESs and PEVs. Simulation results show a decrease in the daily E.B of 26.24%, 2.42%, 60.33%, 29.51%, and 2.38% respectively of the five categories. So, these numerical results prove that the proposed SLSCEMS has considerable efficiency. [ABSTRACT FROM AUTHOR]

• Simplification of a reference battery degradation model with a formulation based on state of health successive updates and normalized exchange energy. • Integration of the model in an energy management strategy and tradeoff between storage degradation and overall system performances. • Tailored Dynamic Programming used to compute optimal battery State of Health trajectories over long period of time for given system configurations. • Optimal storage sizing and energy management of an isolated microgrid while accounting fot storage degradation and replacement. Li-ion batteries are being increasingly used in stationary applications, allowing for greater autonomy and facilitating the integration of renewable energies. However, these devices lose their capacity over time, especially during their so-called second life. This degradation affects the operation of the system and its cost, and must be taken into consideration if an optimal management is to be found. In this work, we present a framework for the integration of the battery aging in a microgrid design and energy management problem. To do so, we first propose a method to simplify a reference model for cyclic degradation of batteries. The results show that the battery loss of capacity shall be compute on a weekly to monthly basis to accurately keep track of degradation effects. Also, at a first order, we show that ageing is dependent with the initial battery State of Health at every update and the energy exchanged normalized by the nominal capacity. This simplified model formulated with Mixed Integer Linear Programming is integrated in a management problem of an isolated energy system for cost minimization. Optimization results show the necessary trade-off between storage degradation and overall system performances. Finally, a tailored dynamic programming is proposed to simulate successive years for the optimal sizing of the considered system in terms of battery capacity and its replacements. Compared to the proposed pproach, not accounting for the storage degradation leads to inaccurate results and significant cost underestimations (>20 %). [ABSTRACT FROM AUTHOR]

The advanced cruise control system has expanded the energy-saving potential of the hybrid electric vehicle (HEV). Despite this, most energy-saving researches for HEV either only optimize the energy management strategy (EMS) or integrate eco-driving through a hierarchically optimized assumption that optimizes EMS and eco-driving separately. Such kinds of approaches may lead to sub-optimal results. To fill this gap, we design a multi-agent reinforcement learning (MARL) based optimal energy-saving strategy for HEV, achieving a cooperative control on the powertrain and car-following behaviors to minimize the energy consumption and keep a safe following distance simultaneously. Specifically, a plug-in HEV model is regarded as the research object in this paper. Firstly, the HEV energy management problem in the car-following scenario is decomposed into a multi-agent cooperative task into two subtasks, each of which can conduct interactive learning through cooperative optimization. Secondly, the energy-saving strategy is designed, called the independent soft actor–critic, which consists of a car-following agent and an energy management agent. Finally, the performance of velocity tracking and energy-saving are validated under different driving cycles. In comparison to the state-of-the-art hierarchical model predictive control (MPC) strategy, the proposed MARL method can reduce fuel consumption by 15.8% while ensuring safety and comfort. [ABSTRACT FROM AUTHOR]

An appropriate energy management strategy is able to further improve the fuel economy of PHEVs. The rule-based energy management algorithms are dominated in industry due to their fast computation and ease of establishment potentials, however, their performance differ a lot from improper setting of parameters and control actions. This paper employs the dynamic programming (DP) to locate the optimal actions for the engine in PHEVs, and more importantly, proposes a recalibration method to improve the performance of the rule-based energy management through the results calculated by DP algorithm. Eventually, an optimization-based rule development procedure is presented and further validated by hardware-in-loop (HIL) simulation experiments. The HIL simulation results show that, the improved rule-based energy management strategy reduces fuel consumption per 100 km from 25.46 L diesel to 22.80 L diesel. The main contribution of this study is to explore a novel way to calibrate the existed heuristic control strategy with the global optimization result through advanced intelligent algorithms. [ABSTRACT FROM AUTHOR]

In this article, a methodology for the sizing and analysis of fuel cell/supercapacitor hybrid vehicles is presented. The proposed sizing methodology is based on the fulfilment of power requirements, including sustained speed tests and stochastic driving cycles. The procedure to generate driving cycles is also presented in this paper. The sizing algorithm explicitly accounts for the Equivalent Consumption Minimization Strategy (ECMS). The performance is compared with optimal consumption, which is found using an off-line strategy via Dynamic Programming. The sizing methodology provides guidance for sizing the fuel cell and the supercapacitor number. The results also include analysis on oversizing the fuel cell and varying the parameters of the energy management strategy. The simulation results highlight the importance of integrating sizing and energy management into fuel cell hybrid vehicles. [ABSTRACT FROM AUTHOR]

The series hybrid electric tracked bulldozer (HETB)’s fuel economy heavily depends on its energy management strategy. This paper presents a model predictive controller (MPC) to solve the energy management problem in an HETB for the first time. A real typical working condition of the HETB is utilized to develop the MPC. The results are compared to two other strategies: a rule-based strategy and a dynamic programming (DP) based one. The latter is a global optimization approach used as a benchmark. The effect of the MPC’s parameters ( e . g . length of prediction horizon) is also studied. The comparison results demonstrate that the proposed approach has approximately a 6% improvement in fuel economy over the rule-based one, and it can achieve over 98% of the fuel optimality of DP in typical working conditions. To show the advantage of the proposed MPC and its robustness under large disturbances, 40% white noise has been added to the typical working condition. Simulation results show that an 8% improvement in fuel economy is obtained by the proposed approach compared to the rule-based one. [ABSTRACT FROM AUTHOR]

Because of the limited resources of micro-controller, rule-based energy management strategies are still very popular for online control of plug-in hybrid electric vehicles, however, the control results may deviate from the optimal control results. Since the city bus routes are predetermined, the speed profiles of the certain bus route do not make much difference, this indeed creates an opportunity to design a novel energy management strategy that can reduce the micro-controller resources usage and achieve close to optimal control performance. To accomplish these goals, the single parameter of length ratio was introduced to represent trip information, and a novel efficient neural network module structure was designed to reduce the calculation time and memory usage of micro-controller. Finally, the length ratio based neural network energy management strategy was proposed for online control of plug-in hybrid electric city bus. Simulation results show that the proposed strategy can greatly decrease the total cost compared with the charge-depleting and charge-sustaining control strategy and can be regarded as an approximated global optimal energy management strategy. [ABSTRACT FROM AUTHOR]

In this paper an adaptive energy management strategy (EMS) based on fuzzy logic and the optimal sizing for a tramway with a hybrid energy storage system (ESS) combining batteries (BT) and supercapacitors (SC) are presented. The EMS applies a sliding window to estimate the forward energy consumption and adapt the instantaneous power target for BT and SC. The hybrid ESS sizing is obtained by an optimization with multi-objective genetic algorithms (GA). The fitness functions are expressed in economic terms, and correspond to the costs of the energy absorbed from the catenary as well as the operation cost of the hybrid ESS (investment and cycling cost). The selected case study is the tramway of Seville, which operates in zones with and without catenary. The aim is to minimize the daily operating cost of the tramway taking into account the BT and SC degradation approach (cycling) and fulfilling the performance of the tramway in the catenary-less zone. The proposed approach (adaptive EMS and optimal sizing) is compared with the current solution in the tramway (SC-based) and with a hybrid ESS managed by a rule-based EMS (RB-EMS) in terms of daily operating cost and energy harnessing during regenerative braking phase. The proposed approach show cost reductions up to 25.5% (from SC-based), 6.2% (from hybrid ESS with RB-EMS) and a global efficiency around 84.4%. [ABSTRACT FROM AUTHOR]

This paper presents a novel control strategy for the optimal management of microgrids with high penetration of renewable energy sources and different energy storage systems. The control strategy is based on the integration of optimal generation scheduling with a model predictive control in order to achieve both long and short-term optimal planning. In particular, long-term optimization of the various microgrid components is obtained by the adoption of an optimal generation scheduling, in which a statistical approach is used to take into account weather and load forecasting uncertainties. The real-time management of the microgrid is instead entrusted to a model predictive controller, which has the important feature of using the results obtained by the optimal generation scheduling. The proposed control strategy was tested in a laboratory-scale microgrid present at the University of Seville, which is composed of an electronic power source that emulates a photovoltaic system, a battery bank and a hydrogen production and storage system. Two different experimental tests that simulate a summer and a winter day were carried out over a 24-h period to verify the reliability and performance enhancement of the control system. Results show an effective improvement in performance in terms of reduction of the microgrid operating cost and greater involvement of the hydrogen storage system for the maintenance of a spinning reserve in batteries. [ABSTRACT FROM AUTHOR]

• A co-simulation platform for Hybrid Electric Vehicles is developed for virtual assessment of Energy Management, performance, CO 2 & pollutant emissions. • CO 2 & pollutant emissions models are based on convolutional neural network for accurate transient state prediction. • A continuous deep reinforcement learning based EMS of HEV is proposed. • DDQL is trained with random generated driving cycles based on Markov Chain approach from real recordings. • Cumulative d error for predicted pollutants emissions is below 8.5% while HEV simulation is running up to 10 times faster than real time in a standard laptop. The Energy Management Strategy (EMS) in an HEV is the key for improving fuel economy and simultaneously reducing pollutant emissions. This paper presents a methodology for developing hybrid models that enable EMS testing as well as the evaluation of fuel consumption, CO 2 and pollutant emissions (CO, NO x and THC). In this context, pollutant emissions are hard to quantify with static models such as the well-known map-based approach which is mainly due to the pronounced impact of transient effects. The novelty of this paper primarily comes from the characterization of pollutant emissions through Convolutional Neural Networks (CNN), providing high accuracy for both, instantaneous and cumulative values. The input parameters are classical Internal Combustion Engine (ICE) measurements such as engine speed, air mass flow, torque and exhaust temperature. The proposed CNNs are reduced to a minimum for low complexity and fast computability. These models are developed with experimental data from chassis dyno testing of a conventional turbo-charged gasoline engine vehicle. The pollutant emission models are used in conjunction with physical models of the remaining powertrain allowing for real time simulations of the complete HEV vehicle. The Double Deep-Q learning algorithm is proposed for the EMS and compared to the Dynamic programming (DP) solution. The introduced methodology is developed in a co-simulation framework between MATLAB-Simulink and AMESIM. The resulting model runs between 8 and 10 times faster than real time in an off-the-shelf PC. This provides the capability for developing models suitable for HIL (hardware-in-the-loop) and SIL (software-in-the-loop) applications. The final error in predicted CO 2 remains below 2.5% while the final cumulative error for pollutants is below 8.5% in the case of CO and HC emissions. [ABSTRACT FROM AUTHOR]

• An online energy management is devised for multi-stack fuel cell hybrid vehicles. • Multi-agent reinforcement learning method is applied for energy management. • Response to fuel cell failure situations can be effectively dealt with. • The proposed method can be transplanted to the actual controller to run online. • The experimental results show that the expected goal can be well accomplished. Because of the high-power demand of fuel cell hybrid vehicles, a multi-stack fuel cell system (MFCS) composed of multiple low-power fuel cell stacks (FCSs) instead of a high-power one has become a satisfactory solution. This is due to the modularity of MFCS, which is more reliable and durable. However, the hybrid power system (MHPSS) of an MFCS hybrid electric vehicle possesses not only MFCS, but also batteries to improve the dynamic performance of MHPSS. Due to the difference in characteristics of MFCS and battery, the energy management strategy (EMS) of MHPSS is the key of ensuring its safe and efficient operation. However, most of the existing MHPSS EMSs are complicated in design and complex to compute online. Besides, they also do not consider the robustness of EMS, that is, EMS has the ability to guarantee the safe operation of MHPSS when the MFCS fails. To solve the above problems, this paper proposes an EMS based on independent Q-learning (IQL) which is an algorithm of multi-agent reinforcement learning to maintain battery state of charge (SOC) and minimize hydrogen consumption. The proposed EMS is not only simple in design, but also can guarantee the normal operation of the MHPSS when the MFCS fails, and can be transplanted to execute online on a microcontroller unit or a field programmable gate array. The various parts of the MHPSS model are first built, then the IQL strategy (IQLS) is trained offline in the established model environment, and finally, the IQLS is ported to the hardware-in-the-loop platform for validation. In order to verify the effectiveness of the proposed EMS, the solution of dimensionality-reduced DP is also used to calculate the fuel economy. Through the experimental verification under different initial SOC and driving cycles, it can be seen that the IQLS proposed in this paper can achieve the goal of maintaining battery SOC and minimizing hydrogen consumption, and also has good generalization ability and safe operation ability under fault conditions. [ABSTRACT FROM AUTHOR]

• An electric-hydrogen-heat hybrid energy storage system is constructed, and a closed-loop energy path of electricity-hydrogen-electricity inside the MECS is formed. • A multi-objective optimization model is constructed with the objectives of carbon emission reductions, investment costs, operating costs, and the goodness of fit. • A complex experiment which is close to the real situation was designed. In the analysis process of the calculation example, three typical days in spring, summer, autumn, and winter are selected to simulate the change throughout the year. With the effectiveness of carbon emission reduction and the trend of clean energy utilization, installed photovoltaic (PV) capacity is increasing rapidly. The multi-energy coupling system (MECS), including hybrid energy storage, can effectively reduce the volatility of PV output and reduce its reliance on the grid. An MECS model is constructed, with objectives of carbon emission reduction, economics, and reliability. The NSGA-Ⅱ algorithm is used and the Pareto frontier of optimal solutions is obtained. The calculation example proved the validity of the model. The energy storage ecosystem composed of battery (BAT), hydrogen storage (HYS), and heat storage (HS), can effectively reduce the BAT capacity configuration. The integrated heat system can increase the energy efficiency by approximately 29%. The system revenue of the reliability optimal scheme (ROS) is 3,305 thousand yuan less than the multi-criteria optimal scheme (MOS); however, its reliability is better than that of MOS, as illustrated by the standard deviation reduction of grid interaction by 4.78%.The energy management strategy has been proven to be feasible because the system can find better reliability and economy from PV, energy storage planning, arbitrage between purchase and sale price differences, and electricity-hydrogen-heat conversion processes. [ABSTRACT FROM AUTHOR]

• Stochastic co-optimization of kinematic states and EMS for CAEVs is developed. • Dynamic probabilistic constraints ensuring driving safety are formulated. • A deep learning model predicting the PDF of preceding vehicle speed is constructed. • An efficient hierarchical solver for the co-optimization problem is designed. Ameliorating energy efficiency and enhancing driving safety are both extremely concerning issues for connected and automated electric vehicles (CAEVs) driving in a random traffic environment. To enhance driving safety and fully coordinate the potential conflict between driving safety and energy efficiency, an adaptive co-optimization method of speed planning and energy management strategy (EMS) with dynamic probabilistic constraints is proposed under the framework of stochastic model predictive control. The dynamic probabilistic constraints are enabled by the proposed composite sequence generation model, which predicts the future speed distribution of the preceding vehicle according to the probability relationship among future speed sequence, historical speed sequence, and macroscopic traffic state of downstream road segments, effectively modeling the macro and micro disturbance from random traffic environment and improving the prediction accuracy by about 10% (along with an over 57% decrease in distribution divergence) compared with pure sequence generation model. Comparison with existing co-optimization methods under the same car-following tasks validates the promising performance of the proposed adaptive co-optimization method, which produces dynamic feasible regions for kinematic states according to downstream traffic state and the driving state of the preceding vehicle, raising the driving safety by 14.81% and retaining the relatively high energy efficiency. [ABSTRACT FROM AUTHOR]

Driving cycles of a city bus is statistically characterized by some repetitive features, which makes the predictive energy management strategy very desirable to obtain approximate optimal fuel economy of a plug-in hybrid electric bus. But dealing with the complicated traffic conditions and finding an approximated global optimal strategy which is applicable to the plug-in hybrid electric bus still remains a challenging technique. To solve this problem, a novel driving-behavior-aware modified stochastic model predictive control method is proposed for the plug-in hybrid electric bus. Firstly, the K -means is employed to classify driving behaviors, and the driver models based on Markov chains is obtained under different kinds of driving behaviors. While the obtained driver behaviors are regarded as stochastic disturbance inputs, the local minimum fuel consumption might be obtained with a traditional stochastic model predictive control at each step, taking tracking the reference battery state of charge trajectory into consideration in the finite predictive horizons. However, this technique is still accompanied by some working points with reduced/worsened fuel economy. Thus, the stochastic model predictive control is modified with the equivalent consumption minimization strategy to eliminate these undesirable working points. The results in real-world city bus routines show that the proposed energy management strategy could greatly improve the fuel economy of a plug-in hybrid electric bus in whole driving cycles, compared with the popular charge depleting–charge sustaining strategy and it may offer some useful insights for realizing the approximate global optimal energy management for the plug-in hybrid electric vehicles. [ABSTRACT FROM AUTHOR]

Planetary gears (PGs) play a critical role in hybrid electric vehicles (HEVs) by combining the output torques of different powertrain components and delivering the resulting torque to the wheels. Whilst previous studies show that the number of planetary gears affects performance of HEVs, there is no prior study to systematically investigate such effects on energy consumption. This paper quantifies the energy efficiency improvement of HEVs due to increasing the number of PGs from one to two, and from two to three. This is done by comparing the minimum energy consumption for different topologies when the rest of the powertrain components – namely electric motors, batteries and engine – are the same. To calculate the minimum energy consumption, the paper proposes an optimal energy management strategy (EMS) for each topology to find the optimum sequence of clutch engagement and torque distribution. The minimum energy consumption of a vehicle with different number of PGs is then evaluated using the automotive simulation models (ASM) from dSPACE. Results show that, for the same electric motors and engine, increasing the number of PGs from one to two and from two to three reduces energy consumption by 5% and 1.5%, respectively. • Electrified powertrains with one, two and three planetary gears are compared. • Two design rules are proposed for hybrid powertrains with multiple planetary gears. • Energy management strategy optimally distributes torque demands and switches operating modes. • Higher numbers of planetary gears reduce energy consumption of hybrid powertrains. [ABSTRACT FROM AUTHOR]

• Novel hybrid dual-motor transmission system is designed for electric vehicles. • The blended optimal rule extraction method is proposed to extract energy management strategy. • SOM is utilized to classify the mixed working points from DP results. • The blended optimal strategy can be implemented in practice effectively. • HIL experiments with compound running scenario are performed to verify the configuration and strategy. To improve the performance of powertrain configuration and heuristic strategy used in electric commercial vehicles currently for complex traffic scenarios, we propose a blended optimal rule extraction method for a novel hybrid dual-motor transmission (hDMT) system. Dynamic programming (DP) is applied to determine the optimal working points on the real-world cycles. However, it is challenging to extract online rules in real-time due to overlapping working points induced by compound running conditions. To create a trade-off between optimum and practicability, a novel blended optimal rule extraction method of energy management strategy is proposed for the integrated powertrain system. To be specific, DP is applied offline to find optimal working points from compound running scenarios. Then, a self-organizing feature map (SOM) is utilized to classify overlapped working points with reference to the characteristics of sample points. To solve the over-classification issue, the recursive k-means method is used to cluster the best matching unit generated by SOM. Based on the above, the optimal strategy is developed systematically and implemented online. Furthermore, the hardware-in-the-loop experiments and simulation results demonstrate that the proposed powertrain configuration and method can improve the economic performance of EVs. The online performance of the blended optimal strategy is highly consistent with the simulation, which can reduce energy consumption by 29.02% compared to the preliminary strategy. Compared with traditional powertrain configuration, the hDMT system also has better energy-saving potential. [ABSTRACT FROM AUTHOR]

• Trip pattern recognition is using the Learning Vector Quantization Neural Network. • Deep Q-learning network trip pattern adaptive battery aging-aware method is devised. • Multi-criteria and trip pattern adaptive strategies are used as benchmark methods. • Tests results indicate the proposed method has good fuel economy and battery aging. The driving trip pattern is of great significance in hydrogen consumption and battery Longevity of the plug-in fuel cell hybrid electric vehicles (PFCHEV). However, the traditional energy management strategy failed to consider the uncertainty of driving patterns. To overcome this drawback, a deep Q-learning network based trip pattern adaptive (DQN-TPA) battery longevity-conscious strategy is proposed in this study. To begin with, the trip pattern recognition based Learning Vector Quantization Neural Network is devised for pattern identification, and the adaptive-equivalent consumption minimizes strategy (A-ECMS) is conducted to improve the hydrogen consumption. Then, a TPA longevity-conscious strategy is developed and compared with the conventional multi-criteria (MC) optimization strategy to investigate the discrepancy brought by the pattern adaptation. And finally, in combination with the above efforts, an improved DQN-TPA based battery longevity-conscious strategy has been established accordingly. The advances are confirmed by the validation results that, the A-ECMS makes an 11.76% promotion in fuel economy by taking the deviation among different driving patterns into concern. The TPA strategy shows more adaptiveness than the MC optimization strategy in which, the effective Ah-throughput is 5.17% lower than MC-based while keeping the same economy. Further improvement can be achieved by the modified DQN-TPA based approach by remedying the imperfection of TPA-based recognition delay and performing the economy and durability conscious actions with 5.87% further reduction of effective Ah-throughput without observably sacrificing the fuel economy. Furthermore, the effectiveness and adaptiveness of the proposed strategy are validated by the Hardware-in-the-Loop experiments. Both the numerical validation and semi-physical validation results indicate that the DQN-TPA based approach made it possible to develop the battery longevity-conscious strategy capable of significantly adapting various driving patterns and improving the hydrogen consumption and battery durability performance of the PFCHEV. [ABSTRACT FROM AUTHOR]

This paper deals with the real-time energy management strategies for a hybrid energy storage system (HESS), including a battery and a supercapacitor (SC), for an electric city bus. The most attractive advantage deriving from HESSs is the possibility of reducing the battery current stress to extend its lifetime. To quantitatively compare the effects of different control strategies on reducing battery degradation, a dynamic degradation model for the LiFePO 4 battery is proposed and validated in this paper. The battery size is optimized according to the requested minimal mileage, while the size of SC is optimized based on the power demand profile of the typical China Bus Driving Cycle (CBDC). Based on the optimized HESS, a novel fuzzy logic controller (FLC) and a novel model predictive controller (MPC) are proposed and compared with the existing rule-based controller (RBC) and filtration based controller (FBC), after all the controllers are tuned to their best performance along the CBDC. It turns out that FLC and RBC achieve the best performance among the four controllers, which is validated by the DP-based result. Furthermore, about 50% of the HESS life cycle cost is reduced in comparison with the battery-only configuration. In addition, the controllers are also compared along the New European Driving Cycle (NEDC), which represents another normalized driving cycle. The results show that the RBC, MPC, and FLC achieve a similar performance, and they reduce about 23% of the HESS life cycle cost when compared to the battery-only configuration. The RBC and FLC are regarded as the best choices in practical applications due to their remarkable performance and easy implementation. [ABSTRACT FROM AUTHOR]

This study fits in the context of collective self-consumption in energy communities, where participants within a given area can exchange and trade energy among themselves. We propose a two-stage approach that decouples the operational phase, i.e., the energy management strategy (EMS), from the settlement where the energy is contractually allocated to the participants. In particular, this decoupled approach allows testing and comparing different methods for the EMS on one side and for the energy sharing on the other, for a total of forty different investigated combinations. Rule-based and optimization-based approaches are considered, allowing each community to compare and select the most appropriate management in line with its specification. The numerical results obtained on a real test case in France show 11.7% saving on the global community bill compared to a case in which the members are not organized as a community. Also, specific allocation mechanisms proposed in this paper allow uniforming the savings in terms of individual bill reduction, which can further encourage end-users to join a community. • Four energy management strategies and ten options for energy allocation and billing. • A penalization of storage systems power variations in order to incur less degradation. • Community saving over 11% compared to a baseline of individual self-consumption only. • Uniform members' bill reduction thanks to optimization-based benefit sharing strategy. [ABSTRACT FROM AUTHOR]

• Hybridization of a long-haul truck using an electrical variable transmission; • Benchmark analysis of three electrical variable transmission powertrain topologies; • Bi-level optimization of the powertrain plant design and its control; • Impact of the component sizing on fuel consumption; • Reduction of fuel consumption and CO 2 emissions by up to 14.2%; The series–parallel architecture is the most interesting for hybrid electric vehicles, allowing the lowest fuel consumption. Unlike passenger cars, this architecture is not commercially available on the heavy-duty vehicles market. This is due to technical limitations associated with unsufficient load capability of the geartrain. To address this issue, new transmissions, such as the electrical variable transmission, have been developed. The novelty of this paper relies on the hybridization of a long-haul truck using the electrical variable transmission. This study aims to investigate the potential of using this new transmission for trucks. For that aim, fuel consumption benchmarking of three powertrain topologies is performed, considering: (a) a gearless topology; (b) a geared topology that uses one gearbox inserted between the engine and the mechanical input port of the electrical variable transmission; (c) a geared topology similar to the second one, but, with an additional multi-stage gearbox inserted to the mechanical output port of the electrical variable transmission. For a fair comparison between the different topologies, a bi-level optimization process has been used, incorporating the optimization of both components sizing and control. Results show that the fuel consumption of the gearless powertrain is higher than the engine-powered truck due to higher losses in the electrical variable transmission. While maximum fuel reduction of 14.2% was obtained by a geared topology that uses two gearboxes. Furthermore, emphasis is given to understand the effect of the powertrain component sizing on fuel consumption. Depending on the defined sizing, a possible fuel reduction is achieved from 3.3% to 14.2% for the two geared topologies. The reduction of CO 2 emissions is found to be proportional to the fuel savings. Considering a long-haul mission, the last findings prove that the electrical variable transmission exhibits potential to reduce fuel consumption, if an adequate powertrain topology and its sizing are well defined. [ABSTRACT FROM AUTHOR]

• Proposing an energy management strategy for a renewable-based microgrid. • Minimum total operating cost is achieved successfully. • The bus voltage has been stabilized employing a flat controller. • Comprehensive analysis and statistics have been provided. This research proposes an effective energy management strategy (EMS) for the economic operation under standalone and grid-connected operating modes of integrated solar renewables microgrid. The proposed microgrid is composed of a photovoltaic generator (PV), a fuel cell system (FC), and a battery storage system. The random nature of the renewable and the load power imposed some stability problems and economic problems, including operating costs. The suggested technique was based on the bald eagle search optimization algorithm (BES), which was designed for a one-day scheduling horizon. The key objectives of this paper were to satisfy the load power with the lowest operating costs under a stable direct current (DC) bus voltage, enhance the overall system efficiency and protect the battery from deep discharge and overcharge. To demonstrate the effectiveness of the proposed strategy, the obtained results were compared with other optimizers, including particle swarm optimization (PSO), salp swarm algorithm (SSA), artificial eco-system optimizer (AEO), COOT optimizer, and political optimizer (PO). The comparison confirmed the superiority of the proposed strategy in terms of minimum operating energy cost (0.1577c€/kW), high efficiency (87.395%) and final SoC (33.268%). [ABSTRACT FROM AUTHOR]

• An energy management strategy considering the effect of start-stop is proposed. • Preference of neural network can be determined by genetic algorithm. • The proportion of start-stop of fuel cell is reduced by 33%. Because of its high efficiency, no emission, low noise and many other advantages, proton exchange membrane fuel cell is considered to be able to be applied in automobiles to replace the traditional internal combustion engine. In order to improve the lifespan of fuel cell, the design of energy management strategy becomes the focus of research. This paper addresses the energy management strategy of fuel cell hybrid electric vehicle- fuel cell as the main power source, battery as the auxiliary power source. Existing researches are summarized and a new algorithm is proposed. As frequent startup, shutdown and rapid load change can reduce the lifespan of fuel cell, it is necessary to avoid this situation as far as possible. For this purpose, the reported work proposes Neural Network Optimized by Genetic Algorithm (NNOGA) as an effective strategy of the studied system. Through the optimization of genetic algorithm, the neural network can be trained pertinently, and the trained network can consciously avoid specific outputs according to the requirements. With the help of the optimization ability of Neural Network Optimized by Genetic Algorithm, which can change the preference of the trained neural network, the network can consciously avoid unnecessary start-stop and fast load change. Therefore, lifespan of fuel cell is prolonged. Simulation and comparative experiments verify the validity of the proposed algorithm. [ABSTRACT FROM AUTHOR]

• Speed prediction MPC is proposed for fuel cell electric vehicle energy management. • Fuel cell performance degradation cost is considered in the proposed MPC strategy. • Speed prediction influences on the energy management strategies are discussed. • Optimization strategies comparisons exhibit the effectiveness of the proposed MPC. • HIL test results verify the real-time performance of the proposed strategies. Due to the poor dynamic response ability of the fuel cell, the battery is normally applied to integrate with fuel cell to configure the hybrid power system in electric vehicles. In this paper, a vehicle speed prediction model predictive control (SP-MPC) energy management strategy is developed for the hybrid power system in fuel cell electric vehicles. The main principle of the proposed SP-MPC is that the future vehicle total power demand is forecasted via the Markov speed predictor and imported into the energy management system response prediction model to improve the control performance by more accurate disturbance description. The objective function is set for equivalent hydrogen consumption minimization and fuel cell degradation inhibition. As a contrast, the normal MPC strategy, the speed prediction dynamic programming (SP-DP) strategy and the DP offline strategy are formulated. Comparing with the normal MPC strategy, the SP-MPC strategy has a 3.74% reduction in the total operation cost under MANHATTAN condition. The SP-MPC strategy also has a 1.39% reduction in the total operation cost than the SP-DP strategy. Moreover, two scenarios are introduced with different disturbance prediction accuracy to verify the influences of the prediction inaccuracy on the SP-MPC and SP-DP results. For SP-DP strategy, the total operation cost under actual forecast scenario has increased by 5.03% compared with the perfect forecast scenario. The similar result can be seen in the SP-MPC, but the increase between perfect and actual forecast scenario is only 1.02%, which indicates a better robustness to the disturbance prediction inaccuracy compared with the SP-DP strategy. A DSP hardware in loop (HIL) test is conducted for real-time performance verification of the proposed SP-MPC. [ABSTRACT FROM AUTHOR]

• A comprehensive review of temperature effects on the performance of fuel cell vehicle. • Temperature effects on the fuel cell and lithium-ion battery have been analyzed. • Guidance for designing integrated energy management strategy is provided. Fuel cell hybrid electric vehicles powered by fuel cell and lithium-ion battery have been considered as an attractive and potential candidate in place of internal combustion vehicles with the advantage of high energy density and energy conversion efficiency, and have become a configuration of interest of many researchers in recent years. Energy management and thermal management are critical issues for steady output power of fuel cell hybrid power system. However, most of the research on energy management and temperature control of fuel cell hybrid electric vehicles is mutually independent. From the literatures, it is observed that existing energy management strategies without integration of fuel cell or battery's temperature dimension are more or less incapable to perform very well. This paper focuses on reviewing the temperature effects on energy management strategies for fuel cell hybrid electric vehicles. To do this, temperature effects on the performance of lithium-ion battery and proton exchange membrane fuel cell are summarized respectively at first. Then, energy management in combination with thermal management of hybrid electric vehicles and fuel cell hybrid electric vehicles are analyzed and summarized in detail. It is indicated that the solution of power distribution considering temperature effects is a significant way to improve the efficiency and service life of fuel cell hybrid electric vehicles compared to the energy management strategy without considering temperature effects. Based on the above analysis, future suggestions of EMS designing are prospected in order to achieve better performance for fuel cell hybrid electric vehicles. [ABSTRACT FROM AUTHOR]