Remaining Useful Life Estimation of Solder joints using an ARMA Model Optimized by Genetic Algorithm
Microelectronic package is an advanced technology which meets the needs of system miniaturization and integration,however,high density integration brings more severe thermal effect problems which is the key to affect the reliability.Moreover,due to the particularity of high density packaging devices,quantitative analysis of its application reliability is often difficult to carry out.In this situation,we evaluate the remaining useful life(RUL)using data-driven prognostic method through the analysis of the performance change of solder joints under temperature cycling test.The health state can be indicated by the current state and the future trend is obtained by RUL prediction.The resistivity is extracted as the characteristic parameter expressing by time series data,and ARMA model is established as mathematical description,with genetic algorithm(GA)optimizing the order estimation.The prediction presents well-fitted for the degradation of microelectronic packaging device characterized by the resistivity of solder joints,showing an effective performance of application in microelectronic packaging device reliability evaluation.
remaining useful life prediction solder joints ARMA model genetic algorithm
Yiwen Long Hongwei Luo Yue Zhi Xiaoqiang Wang
China Electronic Product Reliability and Environmental Testing Research Institute Guangzhou,China
国际会议
上海
英文
1108-1111
2018-08-08（万方平台首次上网日期，不代表论文的发表时间）