作者：蒋闯, 艾红, 陈雯柏

作者单位：北京信息科技大学自动化学院，北京 100192

关键词：绝缘栅双极晶体管;长短时记忆网络;注意力机制;主成分分析;退化预测

摘要：

绝缘栅双极晶体管(IGBT)在可靠性分析任务中时间信息难以充分利用，导致预测精度不高。文中提出一种基于多维时域特征和注意力机制的深度学习方法，该方法结合主成分分析（PCA）技术、长短时记忆网络（LSTM）和注意力（Attention）机制。首先，采用时域分析来手动提取原始数据中的多维时间特征，并利用PCA技术对其进行特征融合处理；然后，利用LSTM网络从样本数据中自动学习序列特征，引入的Attention机制能够对更重要的特征和时间步长赋予更大的权值。最后，使用NASA Ames实验室加速老化数据库进行实验，结果表明所提方法优于最新方法。手动提取的时间特征在经过特征融合后，可以作为序列数据预测任务中的有效退化特征，并结合Attention机制大大提高预测精度。

IGBT degradation prediction method based on time domain characteristics and LSTM-Attention
JIANG Chuang, AI Hong, CHEN Wenbai
College of Automation, Beijing Information Science & Technology University, Beijing 100192, China
Abstract: Aiming at the problem that it is difficult to make full use of time information in reliability analysis task of insulated gate bipolar transistor (IGBT), resulting in low prediction accuracy, a deep learning method based on multi-dimensional features and attention mechanism is proposed. This method combines principal component analysis (PCA), long and short-term memory network (LSTM) and attention mechanism. Firstly, time domain analysis is used to manually extract multi-dimensional time features from the original data, and PCA technology is used for feature fusion. Then, the LSTM network is used to automatically learn sequence features from sample data. The introduced attention mechanism can learn the importance of features and time steps, and give greater weights to more important features. Finally, the prediction accuracy of the model is improved by combining the manually extracted features with the automatically learned features. Experiments are carried out using NASA Ames laboratory accelerated aging database, and the results show that the proposed method is better than the latest method.
Keywords: insulated gate bipolar transistor;long and short-term memory network;attention mechanism;principal component analysis;degradation prediction
2023, 49(8):8-14  收稿日期: 2021-12-19;收到修改稿日期: 2022-02-26
基金项目: 国家自然科学基金资助项目（61973041）；北京市自然科学基金资助项目（4202026）
作者简介: 蒋闯(1993-),男,河南驻马店市人,硕士研究生,专业方向为机器学习和剩余寿命预测研究
参考文献
[1] ZHANG J, HU J, YOU H, et al. Characterization method of IGBT comprehensive health index based on online status data[J]. Microelectronics Reliability, 2021, 116: 114023
[2] FANG X, LIN S, HUANG X, et al. A review of data-driven prognostic for IGBT remaining useful life[J]. Chinese Journal of Electrical Engineering, 2018, 4(3): 73-79
[3] AHSAN M, STOYANOV S, BAILEY C. Data driven prognostics for predicting remaining useful life of IGBT[C]//2016 39th International Spring Seminar on Electronics Technology (ISSE). IEEE, 2016: 273-278.
[4] ISMAIL A, SAIDI L, SAYADI M, et al. Gaussian process regression remaining useful lifetime prediction of thermally aged power IGBT[C]//IECON 2019-45th Annual Conference of the IEEE Industrial Electronics Society. IEEE, 2019, 1: 6004-6009.
[5] LI W, WANG B, LIU J, et al. IGBT aging monitoring and remaining lifetime prediction based on long short-term memory (LSTM) networks[J]. Microelectronics Reliability, 2020, 114: 113902
[6] 高伟, 张琼洁, 李长留, 等. 基于LSTM网络的牵引变流器IGBT故障预测方法研究[J]. 电子器件, 2020, 43(4): 804-808
[7] GE J, HUANG Y, TAO Z, et al. RUL predict of IGBT based on deepAR using transient switch features[C]//PHM Society European Conference. 2020, 5(1): 11-11.
[8] ISMAIL A, SAIDI L, SAYADI M, et al. Remaining useful lifetime prediction of thermally aged power insulated gate bipolar transistor based on Gaussian process regression[J]. Transactions of the Institute of Measurement and Control, 2020, 42(13): 2507-2518
[9] 郑甲宏, 赵敬超. 一种基于PCA-BP的直升机起落架着舰载荷评估方法[J]. 中国测试, 2021, 274(5): 156-161
[10] ZHANG H, ZHANG Q, SHAO S, et al. Attention-based LSTM network for rotatory machine remaining useful life prediction[J]. IEEE Access, 2020, 8: 132188-132199
[11] CHEN Z, WU M, ZHAO R, et al. Machine remaining useful life prediction via an attention-based deep learning approach[J]. IEEE Transactions on Industrial Electronics, 2020, 68(3): 2521-2531
[12] SONNENFELD G, GOEBEL K, CELAYA J R. An agile accelerated aging, characterization and scenario simulation system for gate controlled power transistors[C]//2008 IEEE Autotestcon. IEEE, 2008: 208-215.
[13] 刘子英, 朱琛磊. 基于Elman神经网络模型的IGBT寿命预测[J]. 半导体技术, 2019, 44(5): 395-400
[14] 史业照, 郭斌, 郑永军. 基于LSTM网络的IGBT寿命预测研究[J/OL]. 中国测试: 1-6[2022-02-26]. http://kns.cnki.net/kcms/detail/51.1714.TB.20200810.1312.006.html.