Parameter Estimations For Brushless DC Motor Based On Adaptive H-inifnity Filter

With the accelerated development of information technology and intelligent technology,people not only have an increasing demand for intelligent household products but also have higher and higher performance requirements for these products.Brushless DC motors are widely used in many intelligent manufacturing fields such as smart homes and smart medical devices due to their small size,simple control,and easy maintenance.The performance of these intelligent products depends entirely on the quality of the motor control strategy,and the first prerequisite for precise motor control is to obtain accurate motor parameters.Therefore,online identification of motor parameters is of great significance to accurately control the work of the motor,and is of far-reaching importance for improving the performance of intelligent products.There are two main research issues in this paper.Firstly,for Gaussian noise with unknown statistical properties,the traditional H_∞filter is usually used.The noise covariance matrix and performance upper limit are set by engineers based on experience.For the accuracy of the H_∞filter,if the upper limit and t covariance matrix do not set appropriately,it will not only lead to a decrease in the estimation accuracy but may even lead to filter divergence.Secondly,the noise of brushless DC motors is usually complex and unknown and does not always obey the Gaussian distribution.Meanwhile,the inductance and resistance of the motor may change during operation,and these changes are difficult to measure with experimental equipment.In order to solve the above problem,this paper studies adaptive filters based on the traditional extended H_∞filter to estimate the motor stator inductance and resistance,and simulates the validity of the adaptive filters in the Matlab/Simulink environment.The specific work is as follows:(1)Based on the current equation of the brushless DC motor in the static coordinate system,the state space equation of the motor is established.At the same time,the inductance and resistance affected by motor running time are used as the augmented vector of the system.Build a simulation model of the motor in the Simulink.(2)In actual situations,parameters such as the performance boundary and noise covariance matrix of the extended H_∞filter are often determined after repeated experiments by engineers.In this paper,the adaptive H_∞filter is used to realize the online estimation of noise covariance.Although the parameter values can be estimated effectively,the stability of the algorithm is difficult to guarantee.Then,the expectation maximization algorithm is used to reduce the difficulty of selecting these parameters,and an adaptive H_∞filter is studied.This algorithm,the extended Kalman filter and the extended H_∞filter are used to estimate the motor parameters simultaneously.The simulation results further verify the effectiveness of the adaptive H_∞filtering algorithm when the noise covariance matrix is inaccurate.(3)The generalized maximum correntropy criterion is selected as the cost function in this paper.A robust H_∞filter is proposed based on the traditional extended H_∞filter and the generalized maximum correntropy.At the same time,according to the principle of bounded random variables,the convergence analysis of the robust H_∞filter is carried out.Finally,the simulation analysis is carried out when the motor is subjected to non-Gaussian noise,outlier interference,and a step load torque.The results show that the algorithm has significant advantages in estimation accuracy and robustness.