基于改进神经网络的TMR磁传感系统温漂校准方法
Temperature Drift Calibration Method of TMR Magnetic Sensing System Based on Improved Neural Network
邱伟成 1华超1
作者信息
- 1. 国防科技大学 智能科学学院,长沙 410073
- 折叠
摘要
由于环境以及自身元件的发热等因素的影响,TMR磁传感系统实际工作温度会发生一定程度的波动.随着温度的升高或降低,TMR磁传感系统的实际输出会产生一定的误差,出现输出的温度漂移现象.为了抑制温度漂移对TMR磁传感系统的影响,采用反向传播(BP)神经网络对系统输出的温度漂移现象进行补偿,并利用遗传算法(GA)对BP神经网络进行优化,通过对补偿前后数据的对比,使TMR磁传感系统的输出温漂大幅下降了2个量级,得到了较为理想的效果,提升了TMR磁传感系统的性能和可靠性.
Abstract
Due to the influence of environment and system heating,the actual operating temperature of TMR magnetic sensing system will fluctuate to a certain extent.As the temperature increases or decreases,the actual output of the TMR magnetic sensing system will produce certain errors,resulting in temperature drift of the output.In order to suppress the impact of temperature drift on the TMR magnetic sensing system,a back propagation(BP)neural network optimized by genetic algorithm(GA)is used to compensate the temperature drift phenomenon.By comparing the data before and after compensation,the average temperature drift of the TMR magnetic sensing system can be sharply re-duced two orders of magnitude.The proposed method has achieved perfect compensation and improved the perfor-mance and reliability of the TMR magnetic sensing system.
关键词
隧道磁电阻/输出漂移/磁传感系统/温度补偿Key words
tunnel magnetoresistance/output drift/magnetic sensing system/temperature compensation引用本文复制引用
出版年
2024
NETL
NSTL
万方数据