基于长短时记忆算法的热应力下半导体器件故障预测模型

doi:10.3969/j.issn.1000-1093.2021.06.017

摘要/Abstract

摘要： 针对半导体器件故障主要由时间和应力导致的问题，从多源数据入手，提出基于长短时记忆（LSTM）算法的热应力下半导体器件故障预测模型，研究半导体器件状态随热应力等级和应力累计时长的变化趋势，预测半导体器件故障时间。该模型利用LSTM算法具备长期记忆能力的优势搭建多源数据堆叠结构，提高模型拟合半导体器件状态曲线的能力；通过加权滑动平均滤波法滤除高低频噪音；基于1阶预测器数据压缩算法处理连续缓变的特征向量；采用实验数据对模型进行测试。结果表明：所提模型可较好地反映热应力作用下半导体器件状态的变化趋势，5次实验的预测误差均在1.7%以内，具有较高的准确性；模型能够提前预测故障时间，验证了模型的可行性和有效性。

关键词: 半导体器件, 热应力, 故障预测, 长短时记忆

Abstract: For the semiconductor device failure caused by time and stress, starting with multi-source data, a fault predicition model of semiconductor device under thermal stress based on long short-term memory(LSTM) algorithm is proposed to study the change of semiconductor device status with thermal stress level and the cumulative duration of stress, and predict the failure time of semiconductor device. The proposed model uses the advantage of long-term memory ability of LSTM algorithm to build multi-source data stack structure and improve the ability of model fitting the state curve of semiconductor device. The high and low frequency noises are filtered by using the weighted moving average filtering method. The first-order predictor data compression algorithm is used to deal with the feature vectors with continuous and slow variation. And the experimental data is used to test and verify the model. The results show that the model can better reflect the variation trend of semiconductor device status under the action of thermal stress, and the prediction errors of five experiments are within 1.7%, which has high accuracy. The model can predict the failure time in advance, which verifies the feasibility and effectiveness of the proposed model.

Key words: semiconductordevice, thermalstress, faultprediction, longshort-termmemory

中图分类号:

TN386.1

张明宇, 王琦, 于洋. 基于长短时记忆算法的热应力下半导体器件故障预测模型[J]. 兵工学报, 2021, 42(6): 1265-1274.

ZHANG Mingyu, WANG Qi, YU Yang. LSTM-based Fault Prediction Model of Semiconductor Device under Thermal Stress[J]. Acta Armamentarii, 2021, 42(6): 1265-1274.

参考文献

［1］ LALL P,HANDE M, BHAT C, et al. Prognostics and health
management of electronics[C]∥Proceedings of International Symposium on Advanced Packaging Materials:Processes, Properties and Interface. Atlanta，GA,US: IEEE, 2006.
[2］ DORJ E, CHEN C, PECHT M. A Bayesian hidden Markov model-based
approach for anomaly detection in electronic systems[C]∥
Proceedings of Aerospace Conference. Big Sky，UK: IEEE, 2013.
[3］ KUNCHE S, CHEN C, PECHT M. Optimized diagnostic model combination for improving diagnostic accuracy[C]∥ Proceedings of Aerospace Conference. Big Sky，UK: IEEE, 2013.
[4］ SAHA B, CELAYA J R, WYSOCKI P F, et al. Towards prognostics for electronics components[C]∥Proceedings of Aerospace Conference.Big Sky，UK: IEEE, 2009.
[5］于格,康锐,林焱辉,等. 基于确信可靠度的齿轮可靠性建模与分析[J]. 系统工程与电子技术,2019,41(10):2385-2391.
YU G, KANG R, LIN Y H, et al. Modeling and analysis of gear reliability based on certainty reliability [J]. Systems Engineering and Electronics, 2019, 41(10):2385-2391.(in Chinese)
[6］查国清,黄小凯,康锐. 基于多应力加速试验方法的智能电表寿命评估[J]. 北京航空航天大学学报,2015,41(12):2217-2224.
ZHA G Q, HUANG X K, KANG R. Life evaluation of smart meter based on multi stress accelerated test method [J]. Journal of Beijing University of Aeronautics and Astronautics, 2015, 41(12): 2217-2224. (in Chinese)
[7］查国清,井海龙,陈云霞,等. 基于故障行为模型的产品寿命分析方法[J]. 北京航空航天大学学报,2016,42(11):2371-2377.
ZHA G Q, JING H L, CHEN Y X，et al. Product life analysis method based on fault behavior model[J]. Journal of Beijing University of Aeronautics and Astronautics, 2016, 42(11):2371-2377. (in Chinese)
[8］蒋旭,曾志国,康锐,等. 基于朴素贝叶斯分类的电子产品可靠性仿真试验效果评价方[J]. 电子科学技术, 2015,2(1):49-54.
JIANG X, ZENG Z G, KANG R, et al. Effect evaluation method of electronic product reliability simulation test based on naive Bayes classification [J]. Electronic Science and Technology, 2015, 2(1):49-54. (in Chinese)
[9］ LONG B, TIAN S, MIAO Q, et al. Research on features for diagnostics of filtered analog circuits based on LS-SVM[C]∥Procee- dings of Autotestcom. Baltimore, MD, US: IEEE, 2011: 360-366.
[10］刘震，王厚军，龙兵，等.一种基于加权隐马尔可夫的自回归状态预测模型[J].电子学报，2009，37(10)：2113-2118．
LIU Z, WANG H J, LONG B, et al. An autoregressive state prediction model based on weighted hidden Markov [J]. Acta Electronica Sinica, 2009，37(10)：2113-2118.(in Chinese)
[11］许丽佳,王厚军,龙兵．基于贝叶斯网络的复杂系统故障预测[J].系统工程与电子技术，2008，30(4)：780-784．
XU L J, WANG H J, LONG B. Fault prediction of complex system based on Bayesian Network[J].Systems Engineering and Electronics, 2008，30(4)：780-784．(in Chinese)
[12］许丽佳,龙兵,王厚军．基于LSSVM-HMM的发射机故障预测研究[J].仪器仪表学报，2008，29(1)：21-26．
XU L J, LONG B, WANG H J. Research on transmitter fault prediction based on LSSVM-HMM[J].Chinese Journal of Scientific Instrument, 2008，29(1)：21-26．(in Chinese)
[13］周靖宇,田书林,王厚军,等.采用电压特征的模拟电路单元件预测方法研究[J].微电子学与计算机,2015,32(12):44-48，53.
ZHOU J Y, TIAN S L, WANG H J, et al. Research on prediction method of analog circuit unit based on voltage characteristics[J].Microelectronics & Computer,2015, 32(12):44-48，53. (in Chinese)
[14］吕克洪,程先哲,李华康,等. 电子设备故障预测与健康管理技术发展新动态[J]. 航空学报,2019,40(11):18-29.
L K H, CHENG X Z, LI H K, et al. New development trend of electronic equipment fault prediction and health management technology[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(11):18-29. (in Chinese)
[15］吕克洪,邱静,刘冠军.基于动态损伤及优化AR模型的电子器件寿命预测方法研究[J].兵工学报,2009,30(1):91-95.
L K H, QIU J, LIU G J. Research on the life prediction
method of electronic devices based on dynamic damage and optimized
AR model[J].Acta Armamentarii,2009, 30(1):91-95. (in Chinese)
[16］吕克洪,邱静,刘冠军. 机电系统损伤与时间应力关系的动态描述模型[J].仪器仪表学报,2008,29(8):1788-1792.
L K H, QIU J, LIU G J. Dynamic description model of the relationship between mechanical and electrical system damage and time stress[J].Chinese Journal of Scientific Instrument, 2008, 29(8):1788-1792. (in Chinese)
[17］吕克洪,邱静,刘冠军.基于多组件动态损伤的系统故障预测方法研究[J].系统工程与电子技术,2008,30(10):2014-2018.
L K H, QIU J, LIU G J. Research on system fault prediction method based on multi-component dynamic damage[J].Systems Engineering and Electronics, 2008,30(10):2014-2018. (in Chinese)
[18］席霄鹏,陈茂银,周东华.基于多速率传感器测量数据的剩余寿命预测[J].控制工程,2020,27(5):771-775.
XI X P, CHEN M Y, ZHOU D H. Remaining useful life prediction based on multirate sensor measurement data[J]. Control Engineering of China,2020,27(5):771-775.(in Chinese)
[19］张朝龙,何怡刚,杜博伦, 等.基于深度学习的电力变压器智能故障诊断方法[J].电子测量与仪器学报,2020,34(1):81-89.
ZHANG C L, HE Y G, DU B L, et al. Intelligent fault diagnosis method of power transformer using deep learning [J]. Journal of Electronic Measurement and Instrumentation, 2020, 34(1):81-89. (in Chinese)
[20] 樊浩,李兴文,苏海博, 等.基于主成分分析—支持向量机优化模型的断路器故障诊断方法研究[J].高压电器,2020,56(6): 143-151.
FAN H, LI X W, SU H B, et al. Research on circuit breaker fault diagnosis method based on principal component analysis support vector machine optimization model[J]. High Voltage Apparatus, 2020, 56(6):143-151. (in Chinese)
[21］ VICHARE N,PECHT M.Prognostics and health management of electronics[C]∥Proceedings of the 11st International Symposium on Advanced Packaging Materials: Processes, Properties and Interface. Atlanta，GA, US：IEEE, 2006:222-229.
[22］ HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780.
[23］邱锡鹏.神经网络与深度学习[M].北京：机械工业出版社，2020.
QIU X P. Neural networks and deep learning [M]. Beijing: China Machine Press, 2020. (in Chinese)
[24］集明,熊刚,张海燕.幅相二元距离加权的自适应滑动平均滤波[J].计算机工程与应用,2012,48(10):141-145.
JI M, XIONG G, ZHANG H Y. Amplitude phase binary distance weighted adaptive moving average filtering[J]. Computer Engineering and Applications,2012, 48(10):141-145. (in Chinese)
[25］ GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning [M]. Cambridge, MA, US：MIT Press，2016.
[26］ TESTA A, CARO S D, PANARELLO S, et al. Stress analysis and lifetime estimation on power MOSFETs for automotive ABS systems[C]∥Proceedings of Power Electronics Specialists Conference. Rhodes, Greece:IEEE, 2008:1169-1175.

[1]	张永梅，赖裕平，马健喆，冯超，束颉. 基于视频的装甲车和飞机检测跟踪及轨迹预测算法[J]. 兵工学报, 2021, 42(3): 545-554.
[2]	赵建忠，叶文，张磊. 基于数据融合和改进新陈代谢不等间距GM(11)模型的导弹装备故障预测[J]. 兵工学报, 2014, 35(10): 1689-1695.
[3]	邓森₁，景博₁，周宏亮₁，朱海鹏₁，刘小平₂. 一种改进的自适应增强-支持向量回归机的故障预测方法[J]. 兵工学报, 2012, 33(8): 991-996.
[4]	黄大荣，宋军，赵刚. 基于错分代价的故障预测技术安全性能评估方法[J]. 兵工学报, 2011, 32(10): 1292-1297.
[5]	王永娟₁, 赵军₂, 唐竹秀₁. 压力载荷与温度场作用下弹壳弹膛系统的动态响应[J]. 兵工学报, 2010, 31(6): 669-673.
[6]	李洪儒，许葆华. 某型导弹发射装置液压泵故障预测研究[J]. 兵工学报, 2009, 30(7): 900-906.
[7]	胡海峰，安茂春，泰国军，胡茑庆. 基于隐半Markov模型的故障诊断和故障预测方法研究[J]. 兵工学报, 2009, 30(1): 69-75.
[8]	付宇明，郑丽娟,任兴利. 含半埋藏井线双裂纹的Crl2合金钢试件电磁热止裂分析[J]. 兵工学报, 2008, 29(8): 944-949.
[9]	马彦恒，韩九强，李刚. 电子装备加速实验中的故障预测方法[J]. 兵工学报, 2008, 29(7): 834-838.
[10]	赵文清. 高速列车“中华之星’’制动盘温度场及热应カ[J]. 兵工学报, 2006, 27(1): 132-136.
[11]	1:赵文清;2:王春生. 湿式多盘制动器研究内容的综述[J]. 兵工学报, 2003, 24(1): 111-114.
[12]	黄景德，王兴贵，王祖光. 自行火炮故障预测模糊专家系统研究[J]. 兵工学报, 2002, 23(3): 299-303.
[13]	黄景德，王兴贵，王祖光. 基于模糊评判的装备故障预测模型研究[J]. 兵工学报, 2001, 22(4): 512-515.
[14]	王富耻，王鲁，吕广庶. 金属陶瓷功能梯度热障涂层瞬态热负荷的破坏分析[J]. 兵工学报, 2000, 21(1): 42-45.
[15]	王鲁、吕广庶、王富耻、范天佑. 功能梯度热障涂层热负荷下的有限元分析[J]. 兵工学报, 1999, 20(1): 51-54.