基于CSSA-LSTM的IGBT模块退化趋势预测

doi:10.16180/j.cnki.issn1007-7820.2024.08.009

摘要/Abstract

摘要：

针对逆变器中绝缘栅双极型晶体管(Insulated Gate Bipolar Transistor,IGBT)模块失效率高且易损伤老化以及器件退化过程难以预测的问题,文中提出一种结合长短期神经网络(Long Short-Term Memory,LSTM)和混沌麻雀的神经网络预测模型。通过引入二维皮尔逊相关系数法获取组合退化特征,构建基于LSTM的电压退化预测模型。利用模型自适应提取退化特征内部相关性,实现对关键信息筛选,挖掘深层次退化特征。在麻雀搜索算法的可行域中引入高斯变异的正态分布随机数和Tent映射对应的混沌序列,提升预测的精度和稳定性。对模型的学习率、神经元个数、batch-size进行寻优,寻找最优值匹配网络拓扑。采用最优结构参数的LSTM对各原始数据分别预测,得到最终的退化预测值。以NANS实验中心的加速退化数据集进行算例分析,并与常规预测算法对比,验证所提算法的有效性和准确性。

关键词: 混沌麻雀搜索算法, LSTM, 参数优化, 退化趋势预测, IGBT, 高斯变异, 预测模型, Tent映射

Abstract:

In view of the problem of high failure efficiency of IGBT(Insulated Gate Bipolar Transistor) modules in inverters, which are most prone to damage and aging, and the device degradation process is difficult to predict, a neural network prediction model combining LSTM(Long Short-Term Memory) and chaotic sparrow is proposed. By introducing the two-dimensional Pearson correlation coefficient method to obtain the combined degradation features, the LSTM-based voltage degradation prediction model is constructed. The model is used to adaptively extract the internal correlations of degradation features to realize the screening of key information and digging deep degradation features. In the feasible domain of sparrow search algorithm, Gaussian random numbers with normal distribution and chaotic sequence corresponding to Tent mapping are introduced to improve the accuracy and stability of prediction. The learning rate, number of neurons and batch-size of the model are optimized to find the optimal value to match the network topology. The LSTM with the optimal structural parameters is used to predict each original data separately and obtain the final degradation prediction value. The accelerated degradation data set of NANS experimental center is analyzed and compared with the conventional prediction algorithm to verify the effectiveness and accuracy of the proposed algorithm.

Key words: chaos sparrow search algorithm, LSTM, parameter optimization, prediction of degradation trends, IGBT, Gaussian variation, predictive models, Tent mapping

中图分类号:

TN312

柳行青, 赵国帅, 韩素敏. 基于CSSA-LSTM的IGBT模块退化趋势预测[J]. 电子科技, 2024, 37(8): 60-67.

LIU Hangqing, ZHAO Guoshuai, HAN Sumin. Prediction of Degradation Trend of IGBT Modules Based on CSSA-LSTM[J]. Electronic Science and Technology, 2024, 37(8): 60-67.

图/表 10

图1

图2

图3

表1

图4

图5

图6

图7

图8

表2

参考文献 22

[1]	Long Y Y, Twiefel J, Wallaschek J. A review on the mechanisms of ultrasonic wedge-wedge bonding[J]. Journal of Materials Processing Technology, 2017, 24(5): 241-258.
[2]	徐盛友. 功率模块 IGBT 状态监测及可靠性评估方法研究[D]. 重庆: 重庆大学, 2013:2-3.
	Xu Shengyou. Study on condition monitoring and reliability assessment of IGBT module[D]. Chongqing: Chongqing University, 2013:2-3.
[3]	白梁军, 黄萌, 饶臻, 等. 基于GARCH模型的IGBT寿命预测[J]. 中国电机工程学报, 2020, 40(18):5787-5796.
	Bai Liangjun, Huang Meng, Rao Zhen, et al. Lifetime prediction of IGBT based on GARCH model[J]. Proceedings of the CSEE, 2020, 40(18):5787-5796.
[4]	Ribrant J, Bertling L. Survey of failures in wind power systems with focus on Swedish wind power plants during 1997-2005[J]. IEEE Transactions Energy Conversion, 2007, 22(1):167-173.
[5]	王新, 黄冲, 许翔. PC三电平逆变器电解电容故障特征分析[J]. 电子科技, 2022, 35(5):81-86.
	Wang Xin, Huang Chong, Xu Xiang. Parametric fault characteristics analysis of electrolytic capacitor in NPC three level inverter[J]. Electronic Science and Technology, 2022, 35(5):81-86.
[6]	Li K J, Tian G Y, Cheng L, et al. State detection of bond wires in IGBT modules using eddy current pulsed ther-mography[J]. IEEE Transactions on Power Electronics, 2013, 29(9):5000-5009.
[7]	Qin F, Bie X R, An T, et al. A lifetime prediction method for IGBT modules considering the self-accelerating effect of bond wire damage[J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2021, 9(2):2271-2284.
[8]	Pedersen K B, Pedersen K. Dynamic modeling method of electro-thermo-mechanical degradation in IGBT modules[J]. IEEE Transactions on Power Electronics, 2016, 31(2):975-986.
[9]	Hung T Y, Liao L L, Chiang K N, et al. Life prediction of high-cycle fatigue in aluminum bonding wires under power cycling test[J]. Device and Materials Reliability, 2014, 14(1):484-492.
[10]	Eleffendi M A, Johnson C M. Application of Kalman filter to estimate junction temperature in IGBT power modules[J]. IEEE Transactions on Power Electronics, 2015, 31(2):1576-1587.
[11]	Patil N, Das D, Pecht M. A prognostic approach for non-punch through and field stop IGBTs[J]. Microelectronics Reliability, 2012, 52(3):482-488.
[12]	Alghassi A, Perinpanayagam S, Samie M. Stochastic RUL calculation enhanced with TDNN-based IGBT failure modeling[J]. IEEE Transactions on Reliability, 2015, 65(2):558-573.
[13]	Sreenucht T, Alghassi A, Perinpanayagam S, et al. Probabilistic Monte-Carlo method for modelling and prediction of electronics component life[J]. International Journal of Advanced Computer Science and Applications, 2014, 5(1):1-10.
[14]	Alghassi A, Perinpanayagam S, Samie M, et al. Computationally efficient, real-time, and embeddable prognostic techniques for power electronics[J]. IEEE Transactions on Power Electronics, 2015, 30(5):2623-2634.
[15]	Ahsan M, Stoyanov S, Bailey C. Data driven prognostics for predicting remaining useful life of IGBT[C]. Pilsen: The Thirty-nineth IEEE International Spring Seminar on Electronics Technology, 2016:273-278.
[16]	高金武, 贾志桓, 王向阳, 等. 基于PSO-LSTM的质子交换膜燃料电池退化趋势预测[J]. 吉林大学学报(工学版), 2022, 52(9):2192-2202.
	Gao Jinwu, Jia Zhiheng, Wang Xiangyang, et al. Degradation trend prediction of proton exchange membrane fuel cell based on PSO-LSTM[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(9):2192-2202.
[17]	赵婧宇, 池越, 周亚同. 基于SSA-LSTM模型的短期电力负荷预测[J]. 电工电能新技术, 2022, 41(6):71-79. doi: 10.12067/ATEEE2107053
	Zhao Jingyu, Chi Yue, Zhou Yatong. Short-term load forecasting based on SSA-LSTM model[J]. Advanced Technology of Electrical Engineering and Energy, 2022, 41(6):71-79. doi: 10.12067/ATEEE2107053
[18]	Lai W, Chen M Y, Li R, et al. Low ΔT_j stress cycle effect in IGBT power module die-attach lifetime modeling[J]. IEEE Transactions on Power Electronics, 2016, 31(9):6575-6585.
[19]	Wu W C, Held M, Jacob P, et al. Investigation on the long term reliability of power IGBT modules[C]. Yokohama: Proceedings of the Seventh International Symposium on Power Semiconductor Devices and IC’s, 1995:443-448.
[20]	周鹏, 董朝轶, 陈晓艳, 等. 基于阶梯式Tent混沌和模拟退火的樽海鞘群算法[J]. 电子学报, 2021, 49(9):1724-1735. doi: 10.12263/DZXB.20200593
	Zhou Peng, Dong Chaoyi, Chen Xiaoyan, et al. A salp swarm algorithm based on stepped tent chaos and simulated annealing[J]. Acta Electronica Sinica, 2021, 49(9):1724-1735. doi: 10.12263/DZXB.20200593
[21]	王万良, 金雅文, 陈嘉诚, 等. 多角色多策略多目标粒子群优化算法[J]. 浙江大学学报(工学版), 2022, 56(3):531-541.
	Wang Wanliang, Jin Yawen, Chen Jiacheng, et al. Multi-objective particle swarm optimization algorithm with multi-role and multi-strategy[J]. Journal of Zhejiang University(Engineering Science), 2022, 56(3):531-541.
[22]	王磊, 何江涛, 张振国, 等. 基于信息筛选和拉依达准则识别地下水主要组分水化学异常的方法研究[J]. 环境科学学报, 2018, 38(3):919-929.
	Wang Lei, He Jiangtao, Zhang Zhenguo, et al. Research on the method to identify the outliers of the main components of groundwater based on the information screening coupled with Pauta criterion[J]. Acta Scientiae Circumstantiate, 2018, 38(3):919-929.

相关度分析	I_CE	T₁	T₂	V_GE(_ON₎
IGBT₁	0.022 00	0.688	0.844	0.953
IGBT₂	0.015 60	0.765	0.965	0.991
IGBT₃	0.003 66	0.472	0.912	0.985

器件编号	预测模型	E_RMSE/V	E_MAE/V	E_MAPE/%
IGBT₁	LSTM	0.004 70	0.003 70	0.075 20
	SSA-LSTM	0.004 12	0.003 27	0.066 30
	PSO-LSTM	0.004 53	0.003 36	0.070 40
	CSSA-LSTM	0.003 55	0.002 74	0.055 50
IGBT₂	LSTM	0.004 34	0.003 46	0.063 90
	SSA-LSTM	0.004 85	0.003 91	0.072 20
	PSO-LSTM	0.004 63	0.003 65	0.064 30
	CSSA-LSTM	0.003 67	0.002 78	0.051 30
IGBT₃	LSTM	0.004 80	0.003 53	0.071 65
	SSA-LSTM	0.004 10	0.002 98	0.060 60
	PSO-LSTM	0.004 36	0.003 15	0.062 30
	CSSA-LSTM	0.003 81	0.002 67	0.054 20