Research on probability-based hysteresis modeling method of TMR sensor

Hysteresis is one of the main factors affecting the measurement accuracy of TMR (tunnel magneto resistance) sensors. Generally, hysteresis is compensated by hysteresis analysis to reduce its influence. The commonly used hysteresis analyses are usually based on the Presach model, which has the disadvantages of complex modeling and slow calculation speed and cannot fully improve the measurement accuracy of TMR in practical application. Aiming at the modeling problem of TMR hysteresis characteristics, a modeling method based on hysteresis operator probability estimation is proposed in this paper. The probability distribution of the conversion threshold of the hysteresis operator is estimated, and the static hysteresis model of the TMR sensor is constructed. The probability distribution of the conversion speed of the hysteresis operator under different frequency magnetic fields is estimated and combined with the static hysteresis model, the dynamic hysteresis model of TMR sensor is constructed. Two typical TMR sensors are used to verify the proposed hysteresis model. The maximum error is reduced by 13.3%, and the variance is reduced by 52.1%. The results show that the proposed method can effectively improve the measurement accuracy of the TMR sensor, and the calculation is simple and has good applicability.