Method and system for compensating dynamic hysteresis based on recurrent neural network

Abstract

The invention discloses a method and system for compensating dynamic hysteresis based on a recurrent neural network. The method comprises the steps that: a given magnetic field signal is applied to asensor, and an output signal of the sensor is collected; a recurrent neural network is established, and a network structure is determined, the input quantity of the recurrent neural network is the output signal of the sensor, the output quantity of the recurrent neural network is the compensated output signal of the sensor; the network parameters of the recurrent neural network are initialized, the recurrent neural network is trained through training data, and the predicted value of the output quantity of the recurrent neural network is subtracted from a target value, so that a subtracting result can be obtained; the network parameters are adjusted by using a gradient descent method, and the recurrent neural network is trained through the training data according to the preset maximum number of iterations; and the recurrent neural network of the current network parameter is determined as a trained recurrent neural network when the subtracting result reaches a threshold standard or the recurrent neural network is trained to reach the maximum number of iterations.

Description

technical field [0001] The present invention relates to the technical field of measurement, and more specifically, to a method and system for compensating dynamic hysteresis based on a cyclic neural network. Background technique [0002] Giant magnetoresistors, including giant magnetoresistance (GMR) and TMR (Tunnel MagnetoResistance) devices, are widely used in modern industry and electronic products to measure physical parameters such as current, position, and direction by inducing magnetic field strength. However, because GMR and TMR inevitably have different degrees of hysteresis, the measurement accuracy is greatly reduced. Hysteresis is the main cause of the nonlinearity of the magnetic sensor, which seriously affects the measurement accuracy of the magnetic sensor. In fact, most sensors have hysteresis, so it can be extended to "hysteresis". Therefore, the modeling of hysteresis has been started very early, and hysteresis modeling is the basic premise of hysteresis ...

AI Technical Summary

Problems solved by technology

[0010] The technical solution of the present invention provides a method and system for compensating dynamic hysteresis based on long-short-term memory network to solve how to compensate dynamic hysteresis based on long-short-term memory network issue of compensation

Images