Abstract
Micro-electro-mechanical systems (MEMS) have traditionally been optimized manually based on the solutions to dynamic equations and intuition. This paper presents the application of a multi-objective niched Pareto genetic algorithm (GA) to optimize a synthesized design of a MEM electric field sensor. The geometry of the sensor design is evolved in order to meet the objectives of maximal displacement of at least 5μm, minimal stress, minimal temperature and a resonant frequency near 2kHz. The algorithm gradually evolves a set of solutions towards a Pareto frontier in which no solution is better in all objectives than any other solution. When the algorithm has finished the designer may choose one or more solutions from the set that best meets their objectives given available trade-offs. The results show comparable or better performance in simulation than devices optimized manually or by other means.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
