|
|
|
| Degradation Prediction Method of Proton Exchange Membrane Fuel Cell Based on Online Optimization of Model Parameters |
| Han Yunfei1, Gao Fengyang1, Zhang Jiangang2, Tao Caixia1, Huang Jiaojiao3 |
1. Automation and Electrical Engineering Lanzhou Jiaotong University Lanzhou 730070 China;
2. College of Mathematics and Physics Lanzhou Jiaotong University Lanzhou 730070 China;
3. Bailie Mechanical Engineering Lanzhou City University Lanzhou 730070 China |
|
|
|
Abstract Durability is a key limiting factor for the large-scale commercialization of proton exchange membrane fuel cells (PEMFC), and predicting performance degradation has become a current research hotspot. PEMFC aging data exhibit highly nonlinear, random, and periodic characteristics, making it difficult for prediction models to extract features from historical data effectively. Therefore, improving the accuracy of PEMFC voltage degradation prediction is an urgent problem. Combined with convolutional neural networks (CNN), bidirectional long short-term memory networks (BiLSTM), and an attention mechanism, this paper proposes a hybrid method for PEMFC voltage degradation prediction.
Firstly, historical voltage data for PEMFC are processed using the median absolute deviation method, the Gaussian moving-average filter, and the maxmin normalization algorithm during subsequent data training. This process also ensures the temporal continuity and representativeness of the data. Then, a variant particle swarm optimization algorithm is introduced into the CNN-BiLSTM-Attention model to address the high nonlinearity in voltage data. At the same time, parameters, such as the learning rate, convolution kernel size, number of BiLSTM neurons, and the number of attention mechanism keys, are dynamically optimized. Thus, the method effectively prevents poor fitting and overfitting, thereby improving prediction accuracy and robustness. Finally, the proposed method is validated using steady-state and dynamic voltage aging datasets.
The CNN-BiLSTM-Attention, CNN-BiLSTM, CNN, BiLSTM, and Transformer prediction models are compared. The main conclusions are as follows. (1) The proposed method achieves more accurate voltage prediction accuracy, generalization, and robustness under different working conditions and step sizes of training datasets. (2) Under the same step size of training datasets and working conditions, the proposed method exhibits lower dependency on historical voltage data. In steady-state conditions, the cumulative prediction error of the CNN-BiLSTM-Attention model optimized by the AsyLnCPSO algorithm is reduced by up to 45.3% compared to the benchmark models. In contrast, the cumulative prediction error of the same model optimized by the SAPSO algorithm is reduced by up to 41.3%. In dynamic conditions, the cumulative prediction error of the CNN- BiLSTM-Attention model optimized by the AsyLnCPSO algorithm is reduced by up to 42.2%. The cumulative prediction error of the same model optimized by the SAPSO algorithm is reduced by up to 24.8% compared to the benchmark models. Based on the original voltage-aging data, 10% Gaussian noise is added. The proposed model's predictive performance remains excellent. (3) The proposed method has a simple structure, which is easy to implement and suitable for online applications in real vehicles.
|
|
Received: 28 March 2025
|
|
|
|
|
|
[1] 赵洪山, 潘思潮, 马利波, 等. 基于分类回放双延迟贝叶斯深度确定性策略梯度的燃料电池温度控制[J]. 电工技术学报, 2024, 39(13): 4240-4256.
Zhao Hongshan, Pan Sichao, Ma Libo, et al.Control of fuel cell temperature based on classified replay twin delayed Bayesian deep deterministic policy gradient[J]. Transactions of China Electrotechnical Society, 2024, 39(13): 4240-4256.
[2] 汪建锋, 王荣杰, 林安辉, 等. 质子交换膜燃料电池退化预测方法[J]. 电工技术学报, 2024, 39(11): 3367-3378.
Wang Jianfeng, Wang Rongjie, Lin Anhui, et al.Degradation prediction method of proton exchange membrane fuel cell[J]. Transactions of China Electrotechnical Society, 2024, 39(11): 3367-3378.
[3] 高锋阳, 高翾宇, 张浩然, 等. 全局与瞬时特性兼优的燃料电池有轨电车能量管理策略[J]. 电工技术学报, 2023, 38(21): 5923-5938.
Gao Fengyang, Gao Xuanyu, Zhang Haoran, et al.Management strategy for fuel cell trams with both global and transient characteristics[J]. Transactions of China Electrotechnical Society. 2023, 38(21): 5923-5938.
[4] 马小勇, 王议锋, 王萍, 等. 燃料电池用交错并联型Boost变换器参数综合设计方法[J]. 电工技术学报, 2022, 37(2): 397-408.
Ma Xiaoyong, Wang Yifeng, Wang Ping, et al.Comprehensive parameter design method of interleaved Boost converter for fuel cell applications[J]. Transactions of China Electrotechnical Society, 2022, 37(2): 397-408.
[5] Li Qi, Wang Tianhong, Li Shihan, et al.Online extremum seeking-based optimized energy management strategy for hybrid electric tram considering fuel cell degradation[J]. Applied Energy, 2021, 285: 116505.
[6] 唐钧涛, 戚志东, 裴进, 等. 基于电荷泵的燃料电池有源网络升压变换器[J]. 电工技术学报, 2022, 37(4): 905-917.
Tang Juntao, Qi Zhidong, Pei Jin, et al.An active network DC-DC Boost converter with a charge pump employed in fuel cells[J]. Transactions of China Electrotechnical Society, 2022, 37(4): 905-917.
[7] Ma Jian, Liu Xue, Zou Xinyu, et al.Degradation prognosis for proton exchange membrane fuel cell based on hybrid transfer learning and intercell differences[J]. ISA Transactions, 2021, 113: 149-165.
[8] 李奇, 刘嘉蔚, 陈维荣. 质子交换膜燃料电池剩余使用寿命预测方法综述及展望[J]. 中国电机工程学报, 2019, 39(8): 2365-2375, 19.
Li Qi, Liu Jiawei, Chen Weirong.Review and prospect of remaining useful life prediction methods for proton exchange membrane fuel cell[J]. Proceedings of the CSEE, 2019, 39(8): 2365-2375, 19.
[9] Futter G A, Latz A, Jahnke T.Physical modeling of chemical membrane degradation in polymer electrolyte membrane fuel cells: influence of pressure, relative humidity and cell voltage[J]. Journal of Power Sources, 2019, 410: 78-90.
[10] Chen Kui, Laghrouche S, Djerdir A.Fuel cell health prognosis using unscented Kalman filter: postal fuel cell electric vehicles case study[J]. International Journal of Hydrogen Energy, 2019, 44(3): 1930-1939.
[11] Ma Rui, Yang Tao, Breaz E, et al.Data-driven proton exchange membrane fuel cell degradation predication through deep learning method[J]. Applied Energy, 2018, 231: 102-115.
[12] Zuo Jian, Cadet C, Li Zhongliang, et al.A deteriorationaware energy management strategy for the lifetime improvement of a multi-stack fuel cell system subject to a random dynamic load[J]. Reliability Engineering & System Safety, 2024, 241: 109660.
[13] 申亚涛, 王子鉴, 谷紫文, 等. 数据驱动的低压有源配电网电压在线调控策略[J/OL]. 电力系统自动化, 1-15[2025-07-16]. http://kns.cnki.net/kcms/detail/ 32.1180.TP.20250711.0922.002.html.
Shen Yatao, Wang Zijian, Gu Ziwen, et al. Datadriven online voltage control strategy for low-voltage active distribution networks[J/OL]. Automation of Electric Power Systems, 1-15[2025-07-16]. http://kns. cnki.net/kcms/detail/32.1180.TP.20250711.0922.002. html.
[14] 杜伟, 王圣, 李健, 等. 基于CNN-LSTM-AM模型的储能锂离子电池荷电状态预测[J]. 电工技术学报, 2025, 40(9): 2982-2995.
Du Wei, Wang Sheng, Li Jian, et al.Prediction of state of charge for energy storage lithium-ion batteries based on CNN-LSTM-AM model[J]. Transactions of China Electrotechnical Society, 2025, 40(9): 2982-2995.
[15] 孙静, 翟千淳. 基于融合特征和OOA-BiGRU的锂离子电池剩余使用寿命预测方法[J]. 电工技术学报, 2025, 40(9): 2996-3012.
Sun Jing, Zhai Qianchun.A novel remaining useful life prediction method based on fusion feature and OOA-BiGRU for lithium-ion batteries[J]. Transactions of China Electrotechnical Society, 2025, 40(9): 2996-3012.
[16] 杨淇, 陈景文, 华志广, 等. 基于集成型极限学习机的氢燃料电池寿命预测[J]. 电工技术学报, 2025, 40(3): 964-974.
Yang Qi, Chen Jingwen, Hua Zhiguang, et al.Research on PEMFC lifetime prediction based on ensemble extreme learning machine[J]. Transactions of China Electrotechnical Society, 2025, 40(3): 964-974.
[17] Deng Zhihua, Chen Qihong, Zhang Liyan, et al.Degradation prediction of PEMFCs using stacked echo state network based on genetic algorithm optimization[J]. IEEE Transactions on Transportation Electrification, 2022, 8(1): 1454-1466.
[18] Tang Xingwang, Qin Xiaojin, Wei Kai, et al.A novel online degradation model for proton exchange membrane fuel cell based on online transfer learning[J]. International Journal of Hydrogen Energy, 2023, 48(36): 13617-13632.
[19] Peng Yulin, Chen Tao, Xiao Fei, et al.Remaining useful lifetime prediction methods of proton exchange membrane fuel cell based on convolutional neural network-long short-term memory and convolutional neural network-bidirectional long short-term memory[J]. Fuel Cells, 2023, 23(1): 75-87.
[20] Jin Jiashu, Chen Yuepeng, Xie Changjun, et al.Remaining useful life prediction of PEMFC based on the multi-input cycle reservoir with jump network[J]. International Journal of Hydrogen Energy, 2023, 48(34): 12844-12860.
[21] 李彬, 张雨蒙, 周照钒. 基于LSTM-Transformer的钢铁工业用户调节潜力预测与优化[J/OL]. 电力系统自动化, 1-9[2025-07-16]. http://kns.cnki.net/kcms/detail/32.1180.TP.20250708.1753.014.html.
Li Bin, Zhang Yumeng, Zhou Zhaofan, et al. An LSTM-Transformer based prediction and optimization method for user regulation potential in iron and steel industry[J/OL]. Automation of Electric Power Systems, 1-9[2025-07-16]. http://kns.cnki.net/kcms/detail/32.1180.TP.20250708.1753.014.html.
[22] Li Haolong, Liu Li, Zhang Liyan, et al.A fusion intelligent degradation interval prediction method based on hybrid health indicator for proton exchange membrane fuel cells[J]. IEEE Transactions on Transportation Electrification, 2024, 10(4): 9839-9851.
[23] Ke Chang, Han Kai, Wang Yongzhen.A hybrid method combining degradation mechanisms and deep learning for lifetime prediction of proton exchange membrane fuel cells under dynamic load cycle conditions[J]. Journal of Power Sources, 2025, 633: 236464.
[24] Li Changzhi, Lin Wei, Wu Hangyu, et al.Performance degradation decomposition-ensemble prediction of PEMFC using CEEMDAN and dual data-driven model[J]. Renewable Energy, 2023, 215: 118913.
[25] Zheng Lu, Hou Yongping, Zhang Tao, et al.Performance prediction of fuel cells using long short-term memory recurrent neural network[J]. International Journal of Energy Research, 2021, 45(6): 9141-9161.
[26] Chen Xi, Feng Wentao, Hu Yukang, et al.Power density optimization for proton exchange membrane fuel cell stack based on data-driven and improved light spectrum algorithm[J]. Energy Conversion and Management, 2025, 326: 119467.
[27] 全睿, 程功, 周宇龙, 等. 基于EWOA优化CNNBiGRU-AT模型的燃料电池衰退预测[J]. 电工技术学报, 2025, 40(19): 6342-6358.
Quan Rui, Cheng Gong, Zhou Yulong, et al.Enhanced whale optimization algorithm-based CNN-BiGRUAT model for aging prediction of fuel cell[J]. Transactions of China Electrotechnical Society, 2025, 40(19): 6342-6358.
[28] Rezk H, Aly M, Fathy A.A novel strategy based on recent equilibrium optimizer to enhance the performance of PEM fuel cell system through optimized fuzzy logic MPPT[J]. Energy, 2021, 234: 121267.
[29] Ahmadi S, Abdi S, Kakavand M.Maximum power point tracking of a proton exchange membrane fuel cell system using PSO-PID controller[J]. International Journal of Hydrogen Energy, 2017, 42(32): 20430-20443.
[30] Santos R, Borges G, Santos A, et al.A rotationally invariant semi-autonomous particle swarm optimizer with directional diversity[J]. Swarm and Evolutionary Computation, 2020, 56: 100700.
[31] Jamil M, Yang Xinshe.A literature survey of benchmark functions for global optimisation problems[J]. International Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4(2): 150.
[32] Levy A V, Montalvo A.The tunneling algorithm for the global minimization of functions[J]. SIAM Journal on Scientific and Statistical Computing, 1985, 6(1): 15-29.
[33] Arora J S, Elwakeil O A, Chahande A I, et al.Global optimization methods for engineering applications: a review[J]. Structural Optimization, 1995, 9(3): 137-159.
[34] Sun Changcheng, Gao Mingyu, Cai Hui, et al.Data-driven state-of-charge estimation of a lithiumion battery pack in electric vehicles based on realworld driving data[J]. Journal of Energy Storage, 2024, 101: 113986.
[35] Tamilarasan S, Wang Chongkai, Shih Y C, et al.Generative adversarial networks for stack voltage degradation and RUL estimation in PEMFCs under static and dynamic loads[J]. International Journal of Hydrogen Energy, 2024, 89: 66-83. |
|
|
|
2026, Vol. 41 
