Prediction of Storage Life of Aerospace Electromagnetic Relay Based on POA Optimization Support Vector Regression Model

doi:10.16628/j.cnki.2095-8188.2023.08.010

Abstract

Abstract:

To improve the prediction accuracy of the storage life of aerospace electromagnetic relays, a method based on the pelican optimization algorithm (POA) to optimize the support vector regression (SVR) model for the storage life prediction of aerospace electromagnetic relays is proposed. POA is utilized to optimize SVR model and solve the selection difficulty of kernel parameters in SVR model. Then the performance degradation data obtained from an accelerated degradation test of a certain type of relay is used to validate the proposed method. Finally, compared with the SVR method and SMA-SVR method, the average relative errors decreased by 24.82% and 3.69%, which proves that the proposed method can effectively enhance the prediction accuracy of the relays storage life.

Key words: aerospace electromagnetic relay, pelican optimization algorithm(POA), support vector regression(SVR), life prediction

CLC Number:

TM581.3

ZHU Jiamiao, WANG Zhaobin, LI Jiuxin. Prediction of Storage Life of Aerospace Electromagnetic Relay Based on POA Optimization Support Vector Regression Model[J]. LOW VOLTAGE APPARATUS, 2023, 0(8): 61-67.

Figures/Tables 9

References 19

[1]	王召斌. 航天电磁继电器贮存可靠性退化试验与评价方法的研究[D]. 哈尔滨: 哈尔滨工业大学, 2013.
[2]	王召斌, 符赛, 尚尚, 等. 弹用电磁继电器贮存退化试验及其寿命预测方法[J]. 北京航空航天大学学报, 2016, 42(12):2610-2619.
[3]	李文华, 沈培根. 航天继电器多参数贮存寿命加速退化预测方法研究[J]. 电气工程学报, 2017, 12(1):22-27.
[4]	李文华, 马思宁, 沈培根, 等. 振动条件下铁路继电器寿命预测研究[J]. 电气工程学报, 2017, 12(7):8-15.
[5]	FENG J, SUN Q, JIN T. Storage life prediction for a high-performance capacitor using multi-phase Wiener degradation model[J]. Commun. Stat. Simul. Comput. 2012, 41(8):1317-1335. doi: 10.1080/03610918.2011.624241
[6]	朱佳淼, 王召斌, 李久鑫. 电子电器产品小子样情况下可靠性评估方法综述[J]. 电器与能效管理技术, 2022(12):1-7,12.
[7]	许艳雷, 邱明, 李军星, 等. 基于SKF-KF-Bayes的滚动轴承剩余使用寿命预测方法[J]. 振动与冲击, 2021, 40(19):26-31,40.
[8]	KYEONGJUN L. Bayes and maximum likelihood estimation of uncertainty measure of the inverse weibull distribution under generalized adaptive progressive hybrid censoring[J]. Mathematics, 2022, 10(24).
[9]	陈康宁, 王召斌, 李朕, 等. 基于萤火虫算法优化的改进灰色模型的弹用继电器贮存寿命预测方法[J]. 电器与能效管理技术, 2022(4):1-5.
[10]	关欣, 吕治国, 岳宝强, 等. 基于权重优化组合模型的电磁继电器寿命预测研究[J]. 电器与能效管理技术, 2022(6):45-50.
[11]	ZHAO Y H, ZIO E, FU G C. Remaining storage life prediction for an electromagnetic relay by a particle filtering-based method[J]. Microelectronics Reliability, 2017, 79:221-230. doi: 10.1016/j.microrel.2017.03.026
[12]	WANG Y, HUANG Y H, YANG K, et al. Generator fault classification method based on multi-source information fusion naive bayes classification algorithm[J]. Energies, 2022, 15(24):9635. doi: 10.3390/en15249635
[13]	JIN X M, SONG Y D, LIU X, et al. Probabilistic life prediction for FCG degradation process of turbine disc with small sample data[J]. Engineering Failure Analysis, 2020, 120(4):105026. doi: 10.1016/j.engfailanal.2020.105026
[14]	PARK J P, MOHANTY S, CHI B B, et al. Weibull and Bootstrap-based data-analytics framework for fatigue life prognosis of the pressurized water nuclear reactor component under harsh reactor coolant environment[J]. Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems, 2020, 3(1):1-14.
[15]	LIU Q B, SHI W K, CHEN Z Y. Fatigue life prediction for vibration isolation rubber based on parameter-optimized support vector machine model[J]. Fatigue & Fracture of Engineering Materials & Structures, 2019, 42(3):710-718.
[16]	孙晨峰, 吕卫民, 戴洪德, 等. 一种基于TimeGAN和OCSVM的多元退化设备小子样数据增广方法[J]. 电子学报, 2022, 50(11):2678-2687. doi: 10.12263/DZXB.20220079
[17]	孙世岩, 张钢, 梁伟阁, 等. 基于时间序列数据扩增和BLSTM 的滚动轴承剩余寿命预测方法[J]. 系统工程与电子技术, 2022, 44(3):1060-1068. doi: 10.12305/j.issn.1001-506X.2022.03.40
[18]	TROJOVSK P, DEHGHANI M. Pelican Optimization Algorithm:a novel nature-inspired algorithm for engineering applications[J]. Sensors, 2022, 22(3):855. doi: 10.3390/s22030855
[19]	SMOLA A J, SCHÖLKOPF B. A tutorial on support vector regression[J]. Stats and Computing, 2004, 14(3):199-222.

比较类型	E_r/%
比较类型	SVR	SMA-SVR	POA-SVR
最大值	2.65	2.55	2.46
最小值	0.047 4	0.012 7	0.005 2

预测模型	E_MA/%	E_RMS/%
SVR	2.78	9.69
SMA-SVR	2.17	8.57
POA-SVR	2.09	7.97