设备部件延寿的灰色多指标畸变预测模型

doi:10.12305/j.issn.1001-506X.2023.02.33

摘要/Abstract

摘要：

针对设备部件延寿预测问题, 首先, 提出了上下限两种异常畸变指标, 并给出灰色多指标延寿畸变预测模型的架构图和延寿预测示意图。其次，运用灰色系统方法分别构建了两阶段灰色预测模型, 在第一阶段构建了灰色上下限畸变预测模型, 得到下一次发生畸变的日期; 在第二阶段构建了GM(1, 1)模型, 对设备部件延寿以及上下限指标的异常值进行预测。同时，对模型的预测精度进行检验和分析, 并根据模型的预测结果确定了设备部件延寿的时间。最后，以半导体制造业设备靶材延寿为实际应用案例，验证了该模型的有效性和可行性, 为合理制定部件的最优维护更换时间以及对降低企业运维成本具有重要指导意义。

关键词: 设备延寿, 剩余寿命, 灰色畸变预测, 半导体靶材

Abstract:

Aiming at the problem of life extension prediction of equipment components, two abnormal distortion indexes of upper and lower limits are put forward firstly. And framework diagram and life extension prediction diagram of grey multi index life extension distortion prediction model are given. Secondly, the grey system method is used to build a two-stage grey prediction model. In the first stage, the grey upper and lower limit distortion prediction model is constructed to obtain the date of the next distortion; In the second stage, GM (1, 1) model is built to predict the life extension of equipment components and the abnormal values of upper and lower limit indicators. At the same time, the prediction accuracy of the model was tested and analyzed, and the life extension time of equipment components was determined according to the prediction results of the model. Finally, the effectiveness and feasibility of the model are verified by taking the semiconductor manufacturing equipment target life extension as an actual application case, which has important guiding significance for reasonable formulation of the optimal maintenance and replacement time of components and for reducing the operation and maintenance costs of enterprises.

Key words: equipment life extension, residual life, grey distortion prediction, semicoductor target material

中图分类号:

TP23

李强, 刘思峰. 设备部件延寿的灰色多指标畸变预测模型[J]. 系统工程与电子技术, 2023, 45(2): 597-605.

Qiang LI, Sifeng LIU. Grey multi-index distortion prediction model of equipment component life extension[J]. Systems Engineering and Electronics, 2023, 45(2): 597-605.

图/表 13

图1

图2

图3

图4

表1

表2

表3

图5

图6

表4

表5

表6

图7

参考文献 30

1	GEBRAEEL N , LAWLEY M , LI R , et al. Residual-life distributions from component degradation signals: a Bayesian approach[J]. ⅡE Transactions, 2005, 37 (2): 543- 557.
2	GEBRAEEL N . Sensory-updated residual life distributions for components with exponential degradation patterns[J]. IEEE Trans.on Auto-mation Science and Engineering, 2006, 3 (4): 382- 393. doi: 10.1109/TASE.2006.876609
3	米良, 程珩, 权龙. 基于泊松随机过程的风力发电机叶片疲劳寿命估算[J]. 机械工程学报, 2016, 52 (18): 134- 139.
	MI L , CHENG H , QUAN L . Estimation of blade fatigue life of wind turbine based on Poisson Stochastic process[J]. Journal of Mechanical Engineering, 2016, 52 (18): 134- 139.
4	王浩伟, 徐廷学, 刘勇. 基于随机参数Gamma过程的剩余寿命预测方法[J]. 浙江大学学报(工学版), 2015, 49 (4): 700- 705.
	WANG H W , XU T X , LIU Y . Remaining useful life prediction method based on Gamma processes with random parameters[J]. Journal of Zhejiang University (Engineering Science), 2015, 49 (4): 700- 705.
5	SI X S , WANG W B , MAO Y C , et al. A degradation path-dependent approach for remaining useful life estimation with an exact and closed-form solution[J]. European Journal of Operational Research, 2013, 226 (1): 53- 66. doi: 10.1016/j.ejor.2012.10.030
6	WANG X L , BALAKRISHNAN N , GUO B . Residual life estimation based on a generalized wiener degradation process[J]. Reliability Engineering System Safety, 2014, 12 (4): 13- 23.
7	TANG S J , YU C Q , WANG X , et al. Remaining useful life prediction of lithiumionbatteries based on the wiener process with measurement error[J]. Energies, 2014, 7 (2): 520- 547. doi: 10.3390/en7020520
8	彭才华, 钱富才, 杜许龙. 新型非线性退化建模与剩余寿命预测[J]. 计算机集成制造系统, 2019, 25 (7): 1647- 1654.
	PENG C H , QIAN F C , DU X L . New nonlinear degradation modeling and residual life prediction[J]. Computer Integrated Manufacturing Systems, 2019, 25 (7): 1647- 1654.
9	KHAROUFEH J P , MIXON D G . On a Markov modulated shock and wear process[J]. Naval Research Logistics, 2009, 56 (6): 563- 576. doi: 10.1002/nav.20366
10	KHAROUFEH J P , COX S M . Stochastic models for degradation-based reliability[J]. ⅡE Transaction, 2005, 37 (6): 533- 542. doi: 10.1080/07408170590929009
11	GIORGIO M , GUIDA M , PULCINI G . An age and state-dependent Markov model for degradation processes[J]. ⅡETransactions, 2011, 43 (9): 621- 632.
12	VLOK P J , WNEK M , ZYGMUNT M . Utilizing statistical residual life estimates of bearings to quantify the influence of preventive maintenance actions[J]. Mechanical Systems and Signal Proce-ssing, 2004, 18 (1): 833- 847.
13	SUN Y , MA L , MATHEW J . Mechanical systems hazard estimation using condition monitoring[J]. Mechanical Systems and Signal Processing, 2006, 20 (5): 1189- 1201. doi: 10.1016/j.ymssp.2004.10.009
14	奚立峰, 黄润青, 李兴林, 等. 基于神经网络的球轴承剩余寿命预测[J]. 机械工程学报, 2007, 43 (10): 138- 143.
	XI L F , HUANG R Q , LI X L , et al. Residual life predictions for ball bearing based on neural networks[J]. Journal of Mechanical Engineering, 2007, 43 (10): 138- 143.
15	LI X , DING Q , SUN J Q . Remaining useful life estimation in prognostics using deep convolution neural networks[J]. Reliability Engineering and System Safety, 2018, 172, 1- 11. doi: 10.1016/j.ress.2017.11.021
16	宋亚, 夏唐斌, 郑宇, 等. 基于Autoencoder-BLSTM的祸扇发动机剩余寿命预测[J]. 计算机集成制造系统, 2019, 25 (7): 1611- 1619.
	SONG Y , XIA T B , ZHENG Y , et al. Remaining useful life prediction of turbofan engine based on Autoencoder-BLSTM[J]. Computer Integrated Manuf-acturing Systems, 2019, 25 (7): 1611- 1619.
17	ZHANG C , LI M P , QIN A K , et al. Multiobjective deep belief networks ensemble for remaining useful life estimation in prognostics[J]. IEEE Trans.on Neural Networks and Learning Systems, 2017, 28 (10): 2306- 2318.
18	刘惠, 刘振宇, 郏维强, 等. 基于深度学习的剩余使用寿命预测技术研究现状与挑战[J]. 计算机集成制造系统, 2020, 12 (8): 2079- 2088.
	LIU H , LIU Z Y , JIA W Q . Current research and challenges of deep learning-based remaining useful life prediction[J]. Computer Integrated Manufacturing Systems, 2020, 12 (8): 2079- 2088.
19	SON J B , ZHANG Y L , SANKAVARAM C , et al. RUL Prediction for individual units based on condition monitoring signals with a change point[J]. IEEE Trans.on Reliability, 2014, 64 (1): 182- 196.
20	CHENG Y J , LU C , LI T Y , et al. Residual lifetime prediction for lithium-ion battery based on functional principal component analysis and bayesian approach[J]. Energy, 2015, 90 (Part 2): 1983- 1993.
21	DENG J L . Control problems of grey systems[J]. Systems & Control Letters, 1982, 1 (5): 288- 294.
22	邓聚龙. 灰色控制系统[M]. 武汉: 华中工学院出版社, 1985.
	DENG J L . Grey control system[M]. Wuhan: Press of Huazhong University of Sciencs and Technology, 1985.
23	LIU S F , YANG Y J , JEFFREY F . Grey data analysis[M]. Singapore: Springer-Verlag, 2017.
24	刘思峰. 灰色系统理论及其应用[M]. 9版北京: 科学出版社, 2021.
	LIU S F . Grey systems and its applications[M]. 9th ed Beijing: Science Press, 2021.
25	刘思峰, 杨英杰. 灰色系统研究进展(2004-2014)[J]. 南京航空航天大学学报, 2015, 47 (1): 1- 18.
	LIU S F , YANG Y J . Advances in gray system research (2004-2014)[J]. Journal of Nanjing University of Aeronautics & Astronautic, 2015, 47 (1): 1- 18.
26	CZESLAW C . Decomposition of symptom observation matrix and grey forecasting in vibration condition monitoringof machines[J]. International Journal of Applied Mathematics and Computer Science, 2008, 18 (4): 568- 578.
27	刘思峰, 曾波, 刘解放, 等. GM(1, 1)模型的几种形式及其适用范围研究[J]. 系统工程与电子技术, 2014, 36 (3): 501- 508.
	LIU S F , ZENG B , LIU J F , et al. Several basic models of GM(1, 1) and their applicable bond[J]. Systems Engineering and Electronics, 2014, 36 (3): 501- 508.
28	王旭亮, 聂宏. 基于灰色系统GM(1.1)模型的疲劳寿命预测方法[J]. 南京航空航天大学学报, 2008,, 40 (6): 845- 848.
	WANG X L , NIE H . On a method offatigue life prediction based on gray system model GM(1, 1)[J]. Journal of Nanjing University of Aeronautics & Astronautic, 2008, 40 (6): 845- 848.
29	SHEU D D , KUO J Y . A model for preventive maintenance operations and forecasting[J]. Journal of Intelligent Manufacturing, 2006, 17 (4): 441- 451.
30	黄魁, 苏春. 基于灰色神经网络组合模型的故障预测[J]. 系统工程与电子技术, 2020, 42 (1): 238- 244.
	HUANG K , SU C . Gray neural network model based on fault prediction[J]. Systems Engineering and Electronics, 2020, 42 (1): 238- 244.

序号	实际数据	模拟数据	残差	相对模拟误差/%
1	3	3.000	0.000	0.000
2	7	7.835	-0.835	11.925
3	11	10.552	0.448	4.069
4	15	14.213	0.787	5.248
5	19	19.143	-0.143	0.751
6	26	25.783	0.217	0.836

序号	实际数据	模拟数据	残差	相对模拟误差/%
1	2	2.000	0.000	0.000
2	6	7.123	-1.123	18.723
3	10	9.504	0.496	4.965
4	13	12.679	0.321	2.469
5	17	16.915	0.085	0.497
6	23	22.567	0.433	1.880
7	30	30.108	-0.108	0.360

序号	实际数据	模拟数据	残差	相对模拟误差/%
1	10 112	10 112	0	0.000
2	10 235	10 214	21	0.209
3	10 306	10392	-86	0.834
4	10 510	10573	-63	0.604
5	10 684	10785	-74	0.693
6	10 870	10946	-76	0.699
7	10 951	11137	-187	1.709
8	11 420	11332	88	0.774
9	11 653	11529	106	0.907
10	12 060	11731	329	2.730
11	12 150	11936	214	1.764
12	12 272	12144	126	1.026
13	12 390	12356	34	0.273
14	12 431	12 572	142	1.142
15	12 500	12 791	-291	2.332

序号	实际数据	模拟数据	残差	相对模拟误差/%
1	2.190	2.190	0.000	0.000
2	2.250	2.245	0.005	0.236
3	2.110	2.265	-0.155	7.356
4	2.490	2.286	0.204	8.195
5	2.350	2.307	0.043	1.836
6	2.220	2.328	-0.108	4.863
7	2.360	2.349	0.011	0.455

序号	实际数据	模拟数据	残差	相对模拟误差/%
1	89.640	89.640	0.000	0.000
2	88.810	88.315	0.495	0.558
3	87.080	88.107	-1.027	1.179
4	88.140	87.899	0.241	0.273
5	88.310	87.693	0.617	0.699
6	87.160	87.486	-0.326	0.374