Uncertainty analysis of sensitivity of MEMS microphone based on artificial neural network

Abstract

Due to the uncertainties from manufacturing processes and material properties, MEMS (micro-electro-mechanical system) microphone may exhibit significant variations in their performance compared to the nominal design. The published analytical methods have obvious shortcoming, which cannot meet the needs of both accuracy and efficiency. In order to improve efficiency, the MC (Monte Carlo) simulation based on ANN (artificial neural network) is presented to analyze the uncertainty of sensitivity of polysilicon circular clamped diaphragm microphone. Using the PDS (probabilistic design system) of Ansys software and MC simulation to predict the qualified rate of microphone, the simulated results are 91.2% and 91.4% respectively. The qualified rate of manufactured microphones is 91.5%. In addition, the time-consuming of these two simulations are 10 minutes and about 3000 minutes. This paper also analyzes the change of sensitivity probability densities with the varieties of nominal parameters of diaphragm. The results show that the presented MC simulation with accuracy and high efficiency is an alternative to the traditional methods.