北京航空航天大学学报 ›› 2021, Vol. 47 ›› Issue (10): 1959-1968.doi: 10.13700/j.bh.1001-5965.2020.0349

• 论文 • 上一篇    下一篇

基于IFOA的MEMS加速度计无转台标定

戴洪德, 郑伟伟, 郑百东, 戴邵武, 王瑞   

  1. 海军航空大学, 烟台 264001
  • 收稿日期:2020-07-21 发布日期:2021-11-08
  • 通讯作者: 戴洪德 E-mail:dihod@126.com
  • 基金资助:
    山东省自然科学基金(ZR2017MF036);国防科技项目基金(F062102009);山东省高等学校青年创新团队发展计划(2020KJN003)

Calibration of MEMS accelerometer without turntable based on IFOA

DAI Hongde, ZHENG Weiwei, ZHENG Baidong, DAI Shaowu, WANG Rui   

  1. Naval Aviation University, Yantai 264001, China
  • Received:2020-07-21 Published:2021-11-08

摘要: 为提高微机电系统(MEMS)加速度计的标定效率并降低对高精度转台的依赖,提出一种基于改进果蝇优化算法(IFOA)的MEMS加速度计无转台标定方法。首先,根据模观测标定法原理将加速度计标定问题转化为非线性函数优化问题。然后,针对经典果蝇优化算法存在的只能搜索正参数及搜索步长固定的不足,对味道浓度判定值及搜索步长进行改进,使改进后的算法具有全局参数搜索及可变步长2种性能,并利用Rosenbrock函数进行测试,结果表明,IFOA相比于经典果蝇优化算法具有全局参数寻优范围及更高的寻优精度。最后,将IFOA应用于求解加速度计待标定参数的非线性函数优化问题,并将结果与牛顿迭代法和粒子群优化(PSO)算法进行对比。仿真结果表明:IFOA在求解精度方面比牛顿迭代法提高了1~3个数量级;在运行稳定性方面比牛顿迭代法和PSO算法分别提高了30%和34%,在运行时间方面分别减小了15.2%和43.6%;在加速度计无转台标定方面具有良好的应用价值。

关键词: 微机电系统(MEMS), 加速度计, 标定, 模观测法, 果蝇优化算法(FOA)

Abstract: In order to improve the calibration efficiency of Micro-Electro-Mechanical System (MEMS) accelerometers and reduce the dependence on high-precision turntables, a MEMS accelerometer calibration method based on Improved Fruit Fly Optimization Algorithm (IFOA) without turntables is proposed. The method first converts the accelerometer calibration problem into a nonlinear function optimization problem according to the principle of norm-observation. Afterwards, in view of the shortcomings of the classic FOA that can only search for positive parameters and search step size is fixed, the smell concentration judgment value and search step size were improved to make IFOA have global parameter search and variable step size. The two improved performances were tested using the Rosenbrock function. The results show that the IFOA has a global parameter optimization range and higher optimization accuracy than the classic FOA. Finally, the IFOA was applied to solve the nonlinear function optimization problem of accelerometer calibration parameters. The results are compared with those of Newton iteration method and Particle Swarm Optimization (PSO) algorithm. The simulation results show that the IFOA is 1-3 orders of magnitude higher than Newton iteration method in terms of solution accuracy. Compared with Newton iteration method and PSO algorithm, the IFOA improves the running stability by 30% and 34% respectively, and reduces the running time by 15.2% and 43.6% respectively. The IFOA has a good application value in the calibration of accelerometer without turntable

Key words: Micro-Electro-Mechanical System (MEMS), accelerometer, calibration, norm-observation, Fruit Fly Optimization Algorithm (FOA)

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发