基于改进深度信念网络的TMR电流传感器温漂与地磁场校正方法

Correction Method for Temperature Drift and Geomagnetic Field of TMR Current Sensor Based on Improved Deep Belief Network

精确电流值的测量是电网精益化运行决策的重要前提,高灵敏度、高精度TMR电流传感器的提出有效提升了电流测量能力.与此同时需要重点考虑温度漂移以及空间地磁场在TMR电流传感器测量过程中的影响.针对该问题,本文提出了基于改进深度信念网络的TMR电流传感器温漂与地磁场校正方法.首先,针对TMR电流传感器由于受到强磁场干扰或故障下的异常输出数据,利用贝叶斯结合信息熵理论识别并剔除;其次,使用改进深度信念网络重构空间地磁场、温度与TMR电流传感器测量输出的映射关系;最后,本文对所研发的TMR电流传感器进行了标定实验和误差分析.实验结果表明,在-40~80℃的温度变化范围内,算法补偿后的温度漂移系数由900×10^(-6)/℃降至32.33×10^(-6)/℃.TMR电流传感器对地磁场的敏感程度明显降低,平均绝对百分比误差由2.1530%降低到0.4109%,均方根误差由0.1048 A降低为0.0200 A.

Accurate current measurement is an essential prerequisite for lean power grid operation.The highsensitivity and high-precision TMR current sensor has effectively enhanced the current measurement capability.Simultaneously,the influence of temperature drift and space geomagnetic field in the measurement process of TMR current sensor needs to be considered.To solve this problem,a correction method for temperature drift and geomagnetic field of TMR current sensor based on improved deep belief network is proposed.First,for the abnormal output data of the TMR current sensor because of strong magnetic field interference or failure,Bayesian combined with information entropy theory is used to identify and eliminate;second,the improved deep belief network is used to reconstruct the mapping relationship between the spatial geomagnetic field,temperature,and the measurement output of the TMR current sensor;finally,the calibration experiment and error analysis of the developed TMR current sensor are conducted.The experimental results show that within the temperature range of-40—80℃,the temperature drift coefficient after algorithm compensation is reduced from 900×10^(-6)/℃to 32.33×10^(-6)/℃.The sensitivity of the TMR current sensor to the geomagnetic field is significantly reduced,the average absolute percentage error is reduced from 2.1530%to 0.4109%,and the root mean square error is reduced from 0.1048 A to 0.0200 A.