MKRVM Prediction of Capacitive RF-MEMS Switching Life

Abstract

The elastic coefficient K, a factor affecting the life of capacitive RF-MEMS switch, is selected as the research object. First, we use differential evolution based quantum particle swarm optimization (DE-QPSO) to obtain sparse functions of multiple kernel relevance vector machine (MKRVM), and use the MKRVM algorithm to predict the lifetime of such switches. Then the experimental results show that the method can obtain the health index HI of 3.1043e+06 s, the 3.0657e+06 s. closest to the original data in 0.21 s. At the same time, the conclusion of the longest switching life is obtained when the elastic coefficient is in the range of 4-16 N/m.