Finite Element Method Parameter Estimation of Atomic Sensor Based on Thermal Network Method

Abstract

The heat distribution is one of the key factors which affects the performance of the atomic sensor. Analysis of component temperature is indispensable when determining the need for heat preservation or cooling. Finite element method (FEM) is a commonly thermal analysis method that is widely used in many fields and finite element analysis software is also an essential tool in various thermal analysis cases. Both need accurate parameters to solve the thermal problems. However, some parameters are difficult to be measured in the atomic sensor. To obtain more accurate parameters, a heat transfer model based on the thermal network method is available for parameter estimation that only needs simple conditions to solve problems, which can be measured by experiment. Based on the classical heat transfer theory and thermal network method, the heat transfer model is developed to analyze the heat distribution in the atomic sensor and achieve the parameter estimation of FEM. The model accuracy has been validated by the finite element analysis software and experimental results. The contour maps of the temperature are obtained by finite element analysis software. It is shown that the maximum error of parameter estimation is within 7.8% and the optimal results can achieve 2.1%. The heat transfer model based on the thermal network method can obtain the parameters that can't be measured by experiment under limited boundary conditions, effectively achieve parameter estimation, and make the simulation results closer to the experiment. With the parameters from heat transfer model, the results of FEM become more accurate.