基于ARMA模型的工业电容退化状态研究方法  

Research on Degradation Study of Industrial Capacity Based on ARMA Model

作　　者：张峥[1] 陶耀东[1,2] 余骋远 

机构地区：[1]中国科学院大学,北京100049 [2]中国科学院沈阳计算技术研究所有限公司,沈阳110168

出　　处：《计算机系统应用》2017年第7期30-35,共6页Computer Systems & Applications

基　　金：沈阳市2014年科技计划项目(F14-056-7-00);"高档数控机床与基础制造装备"科技重大专项(2013ZX04007031)

摘　　要：随着中国制造2025计划的推广,工业电源正处于高速发展期,大型铝电解电容的需求越来越大.电容的故障会令工业流水线中断,产生极大的损失,因此工业电容的寿命预测(RUL)具有重要意义.本文通过对铝电解电容充电状态EIS频谱分析,建立电容的状态的量化模型.根据电容状态量化数据,建立ARMA电容退化预测模型.最后通过美国航天航空局的等效串联电阻EIS频谱数据集进行验证.结果表明ARMA电容退化模型对铝电解电容的状态预测有很大的准确性.With the promotion of ＂Made in China 2025＂, the supply in industrial power is in a growing demand for large aluminum electrolytic capacitors. Capacitor failure can interrupt the industrial assembly line, causing a great loss. The prediction of RUL（Remaining Useful Life） of industrial capacitors is hence very important. In this paper, based on the analysis of the EIS of the aluminum electrolytic capacitor, a quantitative model of the state of the capacitor is established.Based on the quantitative data, the degradation prediction model of ARMA capacitor degradation is established. Finally,the equivalent series resistance EIS spectrum data set was verified by NASA. The results show that the short term ARMA capacitor degradation model is accurate in the prediction of the state of aluminum electrolytic capacitors.

关 键 词：工业铝电解电容 退化预测模型 剩余使用寿命 ARMA PHM 

分 类 号：TM535.1[电气工程—电器]