Font Size: a A A

Research Of Compensation Method Of Temperature Characteristics For MEMS Accelerometer

Posted on:2018-10-09Degree:MasterType:Thesis
Country:ChinaCandidate:T LiuFull Text:PDF
GTID:2322330542967166Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
MEMS(Micro Electronic Mechanical Systems)accelerometers are used to measure the acceleration of moving objects,which is widely used in aerospace,smart phones,biological medical,transportation and other fields.However,the performance of MEMS accelerometers are easily affected by manufacturing process,material,work environment,etc.Temperature has a great influence on the performance of MEMS accelerometer because the devices are made of silicon.Hence,the research of reasonable measures on how to reduce the negative impact caused by temperature variation has prominent significance and practical value for engineering applications.This paper studied a series of compensation methods of pendulous micro machined silicon accelerometer SD6068.Main work is as follows:(1)The structure and working principle of the accelerometer are studied,on this basis,the temperature characteristics of the accelerometer are analyzed.Also,the performance parameters of the accelerometer are given.The temperature compensation model of pendulous micro machined silicon accelerometer are established by polynomial surface fitting,generalized regression neural network,support vector machine.The advantages and disadvantages of these methods are summarized,which lays the foundation for selecting the compensation method.(2)According to the requirements of the compensation method,a real-time compensation system is designed based on the STM32F4RGT6.Including hardware circuit module and software programming module.(3)The compensation methods are implemented by the compensation system,including support vector machine regression and surface fitting.Referring to the relevant inertial device testing standard,the real-time testing of the compensation system is carried on.Experimental results show that after the SVM regression compensation the temperature coefficient of scale factor,maximum full-temperature zero bias and the Non-linearity of the accelerometer are reduced from 264ppm/??71.83mg?2.07% to 105ppm/??10.31mg??0.25%,after the surface fitting compensation the temperature coefficient of scale factor,maximum full-temperature zero bias and the Non-linearity of the accelerometer are reduced to 105ppm/??14.19mg??0.37%,respectively,which shows that major parameters of the accelerometer are improved and the temperature compensation system is feasible and effective,which has a broad application prospect.
Keywords/Search Tags:pendulous micro-machined silicon accelerometer, SVM, PSO, temperature compensation
PDF Full Text Request
Related items