Reinforcement Learning Energy Management for Fuel Cell Hybrid Systems: A Review

Xie, Shuqi; Meng, Xiang; Gao, Fei; Zhang, Guorui; Chen, Weirong; Rajashekara, Kaushik

doi:10.1109/mie.2022.3148568

Cited by 38 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The transfer function for the computational delay of the capacitor current internal control loop and the grid current external control loop is represented by G d1 (s) and G d2 (s), respectively. These delay transfer functions can be expressed by Equation (6).…”

Section: The Configuration Of Investigated Systemmentioning

confidence: 99%

“…To obtain an online power distribution scheme with the approximately optimal operating economy, an EMS based on the double Q‐learning algorithm with state constraint and variable action space is proposed in the study by Li et al 5 Reinforcement learning (RL) is a technique for fuel cell/battery hybrid system energy management. In the study by Li et al, 6 the application of RL in EMSs from the perspective of environment construction and adopted agents is reviewed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A robust controller design procedure for LCL‐type grid‐tied proton exchange membrane fuel cell system in harmonics‐polluted network

Hosseinpour

Kholousi

Poulad

2022

Energy Science & Engineering

View full text Add to dashboard Cite

Proton exchange membrane fuel cell (PEMFC), due to its high efficiency and good stability, is one of the most efficient and effective sources of alternative renewable energy. The LCL filters are utilized to attenuate the harmonics caused by inverter‐based distributed power generation systems (DPGS). In this paper, a systematic robust control design is presented for an LCL‐type grid‐tied PEMFC inverter. The proposed nested control includes the capacitor current feedback and the outer grid current loops, considering the computational delays. The proposed control results in more accurate and realistic modeling to transfer the power of PEMFC and inject a high‐quality current into a weak and harmonics‐polluted network. Also, by utilizing the proposed method, the computational delays that affect the damping capability are eliminated, and the performance of the control system is improved. In addition, designing a virtual series and parallel impedances strengthens the phase and amplitude of the inverter output impedance. As a result, the robust operation of the inverter for significant changes in network impedance is improved, and the system well attenuates the mains voltage harmonics. The simulation results in Matlab/Simulink environment confirm the performance of the control system and the suitable quality of the injected current of the PEMFC system into the weak and harmonic‐polluted network.

show abstract

Section: The Configuration Of Investigated Systemmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A robust controller design procedure for LCL‐type grid‐tied proton exchange membrane fuel cell system in harmonics‐polluted network

Hosseinpour

Kholousi

Poulad

2022

Energy Science & Engineering

View full text Add to dashboard Cite

show abstract

“…It is referred to as a lithium-metal battery with metallic lithium as an anode. Moreover, it stands apart from other batteries in its high charge density and cost per unit [27][28][29][30][31][32][33][34][35][36]. The lithium model used in this research is based on MATLAB/Simulink model, as presented in Fig.…”

Section: B Lithium Battery Modelingmentioning

confidence: 99%

“…Furthermore, the previous work-based EMS for HPS does not place emphasis on the training environment and the setup of the reward function, and it does not differentiate between the many stages of the development of RL agents. The purpose of the study in [35] is to address this method by first introducing the notion of an RL-based EMS, then providing literature evaluations from both an RL environment and an agent, and then offering some recommendations for future research.…”

Section: Introductionmentioning

confidence: 99%

Efficient Flatness Based Energy Management Strategy for Hybrid Supercapacitor/Lithium-ion Battery Power System

et al. 2022

View full text Add to dashboard Cite

This article offers a flatness theory-based energy management strategy (FEMS) for a hybrid power system consisting of a supercapacitor (SC) and lithium-ion battery. The proposed FEMS intends to allocate the power reference for the DC/DC converters of both the battery and SC while attaining higher efficiency and stable DC bus voltage. First, the entire system model is analyzed theoretically under the differential flatness approach to reduce the model order as a flat system. Second, the proposed FEMS is validated under different load conditions using MATLAB/Simulink. Thus, this FEMS provides high-quality energy to the load and reduces the fluctuations in the bus voltage. Moreover, the performance of the FEMS is compared with the load following (LF) strategy. The obtained results show that the proposed FEMS meet the real load power under fast variations with good power quality compared to the classical LF strategy, where the maximum overshoot of the bus voltage is 5%.

show abstract

“…Pressure, 11 humidity, 12 temperature, 12 current density, 13 and gas concentration 14 are all directly related to fuel cell performance. To improve the performance of fuel cells, the optimization of flow field design, 15 control strategy, 16 water management, 17 energy management strategy, 18 and evaluation 19 was explored.…”

Section: Introductionmentioning

confidence: 99%

Methods for estimating the accumulated nitrogen concentration in anode of proton exchange membrane fuel cell stacks based on back propagation neural network

Chen

et al. 2022

Intl J of Energy Research

View full text Add to dashboard Cite

Excellent performance is an important indicator to ensure the commercial development of fuel cells. During the operation of a fuel cell system, nitrogen accumulation will occur at the anode of the fuel cell stack because of hydrogen circulation and nitrogen permeation across the membrane. Nitrogen accumulation will lead to the degradation of fuel cell performance. A suitable purge strategy can effectively reduce the excessive nitrogen concentration. The formulation of the purge strategy is closely related to the change in nitrogen concentration, so the accurate estimation of nitrogen concentration is quite important. In this paper, the effect of nitrogen concentration on fuel cell performance under different current densities is studied, and the anode nitrogen concentration in the fuel cell stack is estimated by back propagation neural network. The identification result of nitrogen concentration based on the voltage of every single cell is more accurate. Without considering aging and working condition changes, mean absolute errors of estimated results are 0.75%, 0.67%, 0.62%, and 0.73%, and root mean square errors are 1.09%, 0.97%, 0.88%, and 0.99% at different current densities of 0.6, 0.8, 1.0, and 1.2 AÁcm À2 , respectively. The results indicate that the estimation method of nitrogen concentration for fuel cell anode based on back propagation neural network has a high accuracy, which can provide a new method for formulating purge strategy for fuel cell system.

show abstract

Reinforcement Learning Energy Management for Fuel Cell Hybrid Systems: A Review

Cited by 38 publications

References 27 publications

A robust controller design procedure for LCL‐type grid‐tied proton exchange membrane fuel cell system in harmonics‐polluted network

A robust controller design procedure for LCL‐type grid‐tied proton exchange membrane fuel cell system in harmonics‐polluted network

Efficient Flatness Based Energy Management Strategy for Hybrid Supercapacitor/Lithium-ion Battery Power System

Methods for estimating the accumulated nitrogen concentration in anode of proton exchange membrane fuel cell stacks based on back propagation neural network

Contact Info

Product

Resources

About