A model of optical micro electro mechanical pressure sensor using artificial neural networks

Multilayer thin-film diaphragm has been designed and modeled. Commercial Finite element software, ANSYS was implemented to model the diaphragm. This multi-layer diaphragm can be used as sensor. A simplified model is introduced to reduce the CPU time using APDL macro. The pressure measurement is actually a complicated nonlinear function with multiple variables of sensor output power and test conditions (e.g., temperature, etc.). The most important factor that effect on multilayer thin-film diaphragm is temperature. In this paper we presented an artificial neural network that compensated the effect of temperature on multilayer thin-film diaphragm.