Multidimensional Data Generation Method Based on WGAN and Its Application in RUL Prediction

doi:10.3901/JME.2025.16.070

Abstract

Abstract: Considering that the monitoring data of random degraded equipment under the background of big data presents the characteristics of “fragmented, high-dimensional, and incomplete”, a method for the remaining useful life(RUL) prediction of random degraded equipment in the presence of missing data is proposed. First, for non-ideal sensor data, based on Wasserstein generative adversarial network(WGAN) analysis the hide information of data, expand and complete discontinuous time series, thereby improving data quality. In addition, considering the mutual coupling and mutual influence of multiple performance degradation variables, the Hausdorff distance is used to measure the multi-dimensional similarity between the data from the time and space attributes, and the potential associations between the variables are deeply explored. Finally, the multi-source sensor monitoring data is merged to construct the equipment performance degradation model, and the RUL prediction is realized by the time when the degradation feature first reaches the failure threshold.

Key words: big data, data missing, Wasserstein generative adversarial network (WGAN), multi-dimensional similarity, remaining useful life (RUL) prediction

CLC Number:

TG156

ZHANG Shengfei, LI Tianmei, HU Changhua, SI Xiaosheng, ZHANG Bowei. Multidimensional Data Generation Method Based on WGAN and Its Application in RUL Prediction[J]. Journal of Mechanical Engineering, 2025, 61(16): 70-82.

References

[1] 李天梅,司小胜,刘翔,等. 大数据下数模联动的随机退化设备剩余寿命预测技术[J]. 自动化学报, 2022, 48(9): 23 LI Tianmei, SI Xiaosheng, LIU Xiang, et al. Data-model interactive remaining useful life prediction technologies for stochastic degrading devices with big data[J]. Acta Automatica Sinica, 2022, 48(9): 23.
[2] SI X S, WANG W B, HU C H, et al. Remaining useful life estimation : A review on the statistical data driven approaches[J]. European Journal of Operational Research, 2011, 213(1): 1-14.
[3] 裴洪,胡昌华,司小胜,等. 基于机器学习的设备剩余寿命预测方法综述[J]. 机械工程学报, 2019, 55(8): 1-13. PEI Hong, HU Changhua, SI Xiaosheng, et al. Review of machine learning based remaining useful life prediction methods for equipment[J]. Journal of Mechanical Engineering, 2019, 55(8): 1-13.
[4] SI X S, LI T M, ZHANG Q, et al. Prognostics for linear stochastic degrading systems with survival measurements[J]. IEEE Transactions on Industrial Electronics, 2020, 67(4): 3202-3215.
[5] KUNDU P, DARPE A K, KULKARNI M S. Weibull accelerated failure time regression model for remaining useful life prediction of bearing working under multiple operating conditions[J]. Mechanical Systems and Signal Processing, 2019, 143: 106302.
[6] 王泽洲,陈云翔,蔡忠义,等. 基于复合非齐次泊松过程的不完美维修设备剩余寿命预测[J]. 机械工程学报, 2020, 56(22): 14-23. WANG Zezhou, CHEN Yunxiang, CAI Zhongyi, et al. Remaining useful lifetime prediction of the equipment subjected to imperfect maintenance based on the compound nonhomogeneous poisson process[J]. Journal of Mechanical Engineering, 2020, 56(22): 14-23.
[7] 杨洋. 基于 ARIMA 和 BP 神经网络组合模型的锂电池寿命预测[D]. 海口:海南大学, 2020. YANG Yang. Battery life prediction based on ARIMA with BPNN[D]. Haikou: Hainan University, 2020.
[8] LEE S, KIM H J, KIM S B. Dynamic dispatching system using a deep denoising autoencoder for semiconductor manufacturing[J]. Applied Soft Computing, 2020, 86: 105904.
[9] YU Y, HU C H, SI X S, et al. Averaged Bi-LSTM networks for RUL prognostics with non-life-cycle labeled dataset[J]. Neurocomputing, 2020, 402: 134-147.
[10] PILLAI S , VADAKKEPAT P. Two stage deep learning for prognostics using multi-loss encoder and convolutional composite features[J]. Expert Systems with Applications, 2021, 171(57): 114569.
[11] KHAN S, YAIRI T. A review on the application of deep learning in system health management[J]. Mechanical Systems & Signal Processing, 2018, 107: 241-265.
[12] 王久健,杨绍普,刘永强,等. 一种基于空间卷积长短时记忆神经网络的轴承剩余寿命预测方法[J]. 机械工程学报, 2021, 57(21): 88-95. WANG Jiujian, YANG Shaopu, LIU Yongqiang, et al. A method of bearing remaining useful life estimation based on convolutional long short-term memory neural network[J]. Journal of Mechanical Engineering, 2021, 57(21): 88-95.
[13] 李乃鹏,蔡潇,雷亚国,等. 一种融合多传感器数据的数模联动机械剩余寿命预测方法[J]. 机械工程学报, 2021, 57(20): 29-37, 46. LI Naipeng , CAI Xiao , LEI Yaguo , et al. A model-data-fusion remaining useful life prediction method with multi-sensor fusion for machinery[J]. Journal of Mechanical Engineering, 2021, 57(20): 29-37, 46.
[14] 雷亚国,贾峰,孔德同,等. 大数据下机械智能故障诊断的机遇与挑战[J]. 机械工程学报, 2018, 54(5): 94-104. LEI Yaguo, JIA Feng, KONG Detong, et al. Opportunities and challenges of machinery intelligent fault diagnosis in big data era[J]. Journal of Mechanical Engineering, 2018, 54(5): 94-104.
[15] KAISER J. Dealing with missing values in data[J]. Journal of Systems Integration, 2014, 5(1): 42-51.
[16] AMIRI M, JENSEN R. Missing data imputation using fuzzy-rough methods[J]. Neurocomputing, 2016, 205: 152-164.
[17] PURWAR A, SINGH S K. Hybrid prediction model with missing value imputation for medical data[J]. Expert Systems with Applications, 2015, 42(13): 5621-5631.
[18] KANTARDZIC M. Data mining: Concepts, models, methods, and algorithms[M]. New York: John Wiley & Sons, 2011.
[19] HASTIE T, MAZUMDER R, LEE J, et al. Matrix completion and low-rank SVD via fast alternating least squares[J]. Journal of Machine Learning Research, 2014, 16(1): 3367-3402.
[20] HLALELE N , NELWAMONDO F , MARWALA T. Imputation of missing data using PCA, neuro-fuzzy and genetic algorithms[C]//International Conference on Neural Information Processing. Springer, Berlin, Heidelberg, 2008: 485-492.
[21] MAO Y , ZHANG J , QI H , et al. DNN-MVL : DNN-Multi-View-Learning-Based recover block missing data in a dam safety monitoring system[J]. Sensors, 2019, 19(13): 2895.
[22] CHE Z, PURUSHOTHAM S, CHO K, et al. Recurrent neural networks for multivariate time series with missing values[J]. Scientific Reports, 2018, 8(1): 1-12.
[23] FEDUS W, GOODFELLOW I, DAI A M. Maskgan: Better text generation via filling in the-[C/CD]// 6th International Conference on Learning Representations. Vancouver: ICLR, 2018.
[24] 郝雨微. 基于深度学习的医疗时序数据补值模型研究[D]. 长春:吉林大学, 2019. HAO Yuwei. Research on deep learning based imputation model for clinical time series[D]. Changchun : Jilin University, 2019.
[25] YOON J, JORDON J, SCHAAR M. Gain: Missing data imputation using generative adversarial nets[C]// International Conference on Machine Learning. PMLR, 2018: 5689-5698.
[26] SHANG C, PALMER A, SUN J, et al. VIGAN: Missing view imputation with generative adversarial networks[C]//2017 IEEE International Conference on Big Data (Big Data). New York: IEEE, 2017: 766-775.
[27] LUO Y, CAI X, ZHANG Y, et al. Multivariate time series imputation with generative adversarial networks[C]// Proceedings of the 32nd International Conference on Neural Information Processing Systems. 2018: 1603-1614.
[28] 张晟斐,李天梅,胡昌华,等. 基于深度卷积生成对抗网络的缺失数据生成方法及其在剩余寿命预测中的应用[J]. 航空学报, 2022, 43(8): 441-455. ZHANG Shengfei, LI Tianmei, HU Changhua, et al. Deep convolutional generative adversarial network based missing data generation method and its application in re-maining useful life prediction[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(8): 441-455.
[29] GOODFELLOW I J, POUGET A J, MIRZA M, et al. Generative adversarial networks[C/CD]//International Conference on Neural Information Processing System. MIT Press, 2014.
[30] LEDIG C, THEIS L, HUSZÁR F, et al. Photo-realistic single image super-resolution using a generative adversarial network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 4681-4690.
[31] BAI Y, ZHANG Y, DING M, et al. Sod-mtgan: Small object detection via multi-task generative adversarial network[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 206-221.
[32] YU L, ZHANG W, WANG J, et al. Seqgan: Sequence generative adversarial nets with policy gradient[C/CD]// Proceedings of the AAAI Conference on Artificial Intelligence, 2017.
[33] ARJOVSKY M, CHINTALA S, BOTTOU L. Wasserstein generative adversarial networks[C]//International Conference on Machine Learning. PMLR, 2017: 214-223.
[34] ARJOVSKY M , BOTTOU L. Towards principled methods for training generative adversarial networks[J]. arXiv Preprint arXiv: 1701.04862, 2017.
[35] HUSZÁR F. How (not) to train your generative model: Scheduled sampling, likelihood, adversary?[J]. arXiv: 1511.05101, 2015.
[36] 胡铭菲,刘建伟,左信. 深度生成模型综述[J]. 自动化学报, 2022, 48(1): 40-74. HU Mingfei, LIU Jianwei, ZUO Xin. Survey on deep generative model[J]. Acta Automatica Sinica, 2022, 48(1): 40-74.
[37] LI Y, ZHANG T. A hybrid Hausdorff distance track correlation algorithm based on time sliding window[C]//MATEC Web of Conferences. EDP Sciences, 2021, 336: 07015.
[38] 牟含笑,郑建飞,胡昌华,等. 基于 CDBN 与 BiLSTM 的多元退化设备剩余寿命预测[J]. 航空学报, 2022, 43(7): 308-319. MU Hanxiao, ZHENG Jianfei, HU Changhua, et al. Remaining useful life prediction of multivariate degradation equipment based on CDBN and Bi LSTM[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(7): 308-319.
[39] HINTON G, OSINDERO S, THE Y. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7): 1527-1554.
[40] HINTON G. A practical guide to training restricted Boltzmann machines[J]. Momentum, 2010, 9(1): 926-947.
[41] BIN Y, YANG Y, SHEN F, et al. Bidirectional long-short term memory for video description[C]// Proceedings of the 24th ACM International Conference on Multimedia. New York: ACM, 2016: 436-440.
[42] SAXENA A, GOEBEL. C-MAPSS data set[EB/OL]. [2021-09-06]. https://www.dssz.com/2148128.html.
[43] KINGMA D P, BA J. Adam: A method for stochastic optimization[J]. arXiv Preprint arXiv: 1412.6980, 2014.
[44] BABU G S, ZHAO P, LI X L. Deep convolutional neural network based regression approach for estimation of remaining useful life[C]//International Conference on Database Systems for Advanced Applications. Berlin: Springer, Cham, 2016: 214-228.