Multi-mode predictive energy management for fuel cell hybrid electric vehicles using Markov driving pattern recognizer

Zhou, Yang; Ravey, Alexandre; Péra, Marie‐Cécile

doi:10.1016/j.apenergy.2019.114057

Cited by 160 publications

(47 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• In our previous works, an Elman NN predictor [17] and a fuzzy C-means enhanced Markov predictor [19] are built using the offline historical driving database. Nevertheless, their prediction quality would be greatly challenged if the discrepancy between the realistic driving scenarios and the historical ones were significant [27].…”

Section: Motivations and Innovationsmentioning

confidence: 99%

Multi-objective energy management for fuel cell electric vehicles using online-learning enhanced Markov speed predictor

Zhou

Ravey

Péra

2020

Energy Conversion and Management

Self Cite

View full text Add to dashboard Cite

As one of promising solutions towards future cleaner transportation, fuel cell electric vehicles have been widely regarded as an attractive technology in both academia and industry. To enhance the vehicle's operation efficiency, this paper proposes a multi-criteria power allocation strategy for a fuel cell/batterybased plug-in hybrid electric vehicle. Firstly, an adaptive online-learning enhanced Markov velocityforecast approach is proposed. Its predictive behaviors can be adjusted accordingly under various driving scenarios through the real-time-identified transition probability matrices. Subsequently, based only on the previewed trip duration information and the speed prediction results, a state-of-charge (SOC) reference planning approach is designed to guide the allocation of battery energy. Combining with the velocity-forecast results and the reference SoC, model predictive control derives the optimal powerallocation decision through minimizing the multipurpose objective function in a finite time horizon. It has been verified that (1) the presented power allocation strategy can reduce over 12.05% H2 consumption and over 94.40% fuel cell power spikes against the commonly used Charge-Depleting/Charge-Sustaining strategy; (2) despite the existence of mission time estimation errors, the presented control strategy could still bring performance enhancement over the benchmark strategy, thus demonstrating its feasibility for real-world implementations.

show abstract

Section: Motivations and Innovationsmentioning

confidence: 99%

Multi-objective energy management for fuel cell electric vehicles using online-learning enhanced Markov speed predictor

Zhou

Ravey

Péra

2020

Energy Conversion and Management

Self Cite

View full text Add to dashboard Cite

show abstract

“…The vehicle speed dataset is collected from the real driving conditions. To reduce the training time and improve the control accuracy, the vehicle velocity is divided into three-speed intervals that are [0-12] m/s, [12][13][14][15][16][17][18][19][20][21][22][23][24] m/s, [24][25][26][27][28][29][30][31][32][33][34][35][36] m/s, and they represent low, medium, and high speed, respectively. Then, the classified speed intervals are adopted to train the DDPG algorithm separately until the algorithm converges, the trained neural network is stored, as depicted in Fig.…”

Section: A a Bi-level Frameworkmentioning

confidence: 99%

“…In [23], better power distributions between the battery and the ultracapacitor of PHEVs were obtained through the RL-based method, and 16.8% energy loss reduction was achieved. The researchers in [24] and [25] adopted the predicted EMS to improve the HEVs' performance. However, the discrete state space and action space of RL hinder its further application in EMS of HEVs, and the emergence of deep reinforcement learning (DRL) has bridged over this difficulty.…”

Section: Introductionmentioning

confidence: 99%

Transfer Deep Reinforcement Learning-Enabled Energy Management Strategy for Hybrid Tracked Vehicle

Guo

Tang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

This paper proposes an adaptive energy management strategy for hybrid electric vehicles by combining deep reinforcement learning (DRL) and transfer learning (TL). This work aims to address the defect of DRL in tedious training time. First, an optimization control modeling of a hybrid tracked vehicle is built, wherein the elaborate powertrain components are introduced. Then, a bi-level control framework is constructed to derive the energy management strategies (EMSs). The upper-level is applying the particular deep deterministic policy gradient (DDPG) algorithms for EMS training at different speed intervals. The lower-level is employing the TL method to transform the pre-trained neural networks for a novel driving cycle. Finally, a series of experiments are executed to prove the effectiveness of the presented control framework. The optimality and adaptability of the formulated EMS are illuminated. The founded DRL and TL-enabled control policy is capable of enhancing energy efficiency and improving system performance.

show abstract

“…3, in the longitudinal direction, the major external forces acting on the vehicle includes tire rolling resistance ( ), aerodynamic drag ( ), slope resistance ( ), and the tractive effort of the vehicle ( ). The dynamic equation of vehicle motion along the longitudinal direction is expressed by [23,24,32,36]:…”

Section: Powertrain System Modelingmentioning

confidence: 99%

“…where is the battery pack power, is the capacity, and is the coulombic efficiency only during charge (0.98) [36].…”

Section: Battery Modelingmentioning

confidence: 99%

Investigating the impact of ageing and thermal management of a fuel cell system on energy management strategies

et al. 2020

View full text Add to dashboard Cite

This paper studies the impact of two significant aspects, namely fuel cell (FC) degradation and thermal management, over the performance of an optimal and a rule-based energy management strategy (EMS) in a fuel cell hybrid electric vehicle (FCHEV). To do so, firstly, a vehicle model is developed in simulation environment for a low-speed FCHEV composed of a FC stack and a battery pack. Subsequently, deterministic dynamic programming (DP), as an optimal strategy, and bounded load following strategy (BLFS), as a common rule-based strategy, are utilized to minimize the hydrogen consumption while respecting the operating constraints of the power sources. The performance of the EMSs is assessed in different scenarios. The first objective is to clarify the effect of FC stack degradation on the performance of the vehicle. In this regard, each EMS determines the required current from the FC stack for two FCs with different levels of degradation. The second objective is to evaluate the thermal management contribution in improving the performance for the new FC compared to the considered cases in scenario one. In this respect, each strategy deals with determining two control variables (FC current and cooling fan duty cycle). The results of this study indicate that negligence of adapting to the PEMFC health state, as the PEMFC gets aged, can increase the hydrogen consumption up to 24.8% in DP and 12.1% in BLFS. Moreover, the integration of temperature dimension into the EMS can diminish the hydrogen consumption by 4.1% and 5.3% in DP and BLFS respectively.

show abstract

Multi-mode predictive energy management for fuel cell hybrid electric vehicles using Markov driving pattern recognizer

Cited by 160 publications

References 35 publications

Multi-objective energy management for fuel cell electric vehicles using online-learning enhanced Markov speed predictor

Multi-objective energy management for fuel cell electric vehicles using online-learning enhanced Markov speed predictor

Transfer Deep Reinforcement Learning-Enabled Energy Management Strategy for Hybrid Tracked Vehicle

Investigating the impact of ageing and thermal management of a fuel cell system on energy management strategies

Contact Info

Product

Resources

About