Research On Temperature Control System Of Mems Resonant Accelerometer

MEMS resonant accelerometer not only has the advantages of general MEMS devices,but also has the advantages of high sensitivity,wide dynamic range,strong anti-interference ability,good stability and convenient signal processing.It is a micro accelerometer with high precision potential.It is widely used in various fields and has become one of the research hotspots of MEMS inertial sensors.However,the output temperature error of MEMS accelerometer caused by the temperature change of working environment is a main error source affecting the performance of accelerometer.Taking the silicon micro resonant accelerometer as the test object and aiming at improving the temperature control accuracy,this paper designs the accelerometer temperature control system to improve the temperature error of the accelerometer,and then improve the performance of the accelerometer.Firstly,this paper studies the working principle of the silicon micro resonant accelerometer,analyzes the temperature error mechanism of the accelerometer,compares various methods to suppress the temperature error,and uses the heating temperature control system to improve the temperature performance of the silicon micro resonant accelerometer.Secondly,the overall scheme of the temperature control system at normal and full temperature is designed in this paper.The multi-point heating subsection control is the basic control mode to improve the output accuracy of the accelerometer.The insulation box is designed and added as the insulation layer structure to realize the system temperature buffer in the full temperature environment.Through the thermodynamic theory analysis,the heating uniformity of the heating plate and the rationality of the design of the normal and full temperature environment control system are verified;The reliability of the temperature control system is verified by modeling and simulating the accelerometer temperature control system;Through the reasonable selection of system components,the temperature control system circuit is built to realize the temperature acquisition,processing and control of the system.Thirdly,an improved fuzzy proportional integral derivative(PID)control algorithm based on fruit fly algorithm(FOA)is designed to reduce the fuzzy PID temperature control error in the system.By comparing position PID with incremental PID,incremental PID is used to reduce the random error of PID control;Through anti windup method and integral separation to optimize PID control,the influence of integral error is reduced;The training process of coding and optimization step length of the Drosophila algorithm is improved.Compared with the traditional Drosophila algorithm,the convergence speed and accuracy of the Drosophila algorithm are improved,and the risk of falling into local optimization is effectively reduced;Based on the optimized Drosophila algorithm,the whole flow of the improved fuzzy PID control algorithm is designed,and the initial parameters of the fuzzy PID are optimized.Finally,an accelerometer temperature control system experimental platform is built and designed to test the performance of the temperature control system and the output of the accelerometer.Through the stability test of the temperature control system,it is verified that the improved fuzzy PID control algorithm can effectively improve the anti disturbance ability of the temperature control system;The temperature control effects of traditional PID,fuzzy PID control and improved fuzzy PID control algorithm are compared through room temperature experiments.The test results show that the improved fuzzy PID temperature control system has obvious advantages in overshoot,convergence time and steady-state error,and improves the temperature control performance;By comparing the outputs of accelerometers with and without temperature control system at room temperature and full temperature,it is verified that the temperature control system can effectively reduce the temperature error of accelerometers and improve the output accuracy of accelerometers.