Reliability-Based MEMS System Modeling and Optimization

A methodology for reliability-based system modeling, analysis and optimization design of micro-electromechanical systems (MEMS) is presented that accounts for stochastic variations in device geometry parameters and operating conditions. The optimization objective function considers minimization of several uncertainty factors related to the overall system performance while satisfying target requirements specified. A probabilistic sufficiency factor approach is proposed in the form of constraints on micro-fabrication processes and materials system that combine safety factor and probability of failure. The design problem is decomposed into two analysis systems; uncertainty effects analysis and performance sensitivity analysis. Each analysis system can be partitioned into several subsystems according to the different functions they perform. The entire problem has been treated as a multi-disciplinary design optimization (MDO) for maximum robustness and performance achievement. The probabilistic sufficiency factor approach represents a factor of safety relative to a target probability of failure. It is known that the use of probabilistic sufficiency factor as a design constraint boundary is much more accurate and reasonable, and this accelerates the convergence of reliability-based design optimization. In this study, the analysis results are provided as optimized device geometry parameters governing the resonant frequency and the trans-conductance values for the example of a selected micro-resonator device