Method for estimating status of brushless direct current motor based on extended kalman filter

Abstract

The invention discloses a method for estimating the status of a brushless direct current motor based on an extended kalman filter. By means of the method, debugging difficulty is reduced and observation accuracy is improved under the conditions that system noises and measurement noises are non-zero mean white noises and errors exist in a system model. A measuring error covariance matrix R is regulated according to fuzzy rules, therefore, system debugging difficulty is reduced and accuracy is improved. Furthermore, an attenuation factor is added into a status updating equation, correction weight of the new observed quantity and the existing observed quantity on an estimated value is regulated, and tracking performance is improved. In addition, by means of a method of performing grouping regulation on a value of p (0), Q, serious distortion is controlled, and estimation accuracy of the extended kalman filter is improved.

Description

technical field [0001] The invention relates to the technical field of position sensorless control of a brushless direct current motor, in particular to a position sensorless brushless direct current motor state estimation method based on an extended Kalman filter. Background technique [0002] Permanent magnet brushless DC motor (Brushless DC Motor, BLDCM) combines the advantages of simple control mode of DC motor, good torque characteristics, good speed regulation performance, and simple manufacture of AC motor, no excitation loss, high power density, etc., so it is used in various fields Both have applications, and have attracted extensive attention and research from scholars at home and abroad. However, the operation of the brushless DC motor requires a rotor position signal, and the traditional method is to use a position sensor. With the research and demand for motors with smaller rated power, the volume of the position sensor accounts for an increasing percentage of ...

AI Technical Summary

Problems solved by technology

[0006] The main problem with the traditional method is that the rotor speed and position information can be better estimated in a certain speed range, but when the motor is running at low or high speed, the observation effect drops significantly or even cannot be used.

First of all, since the parameters of the Kalman filter are set at a certain speed (such as the rated speed), when the speed changes, the observation accuracy will decrease

Secondly, in the design process of the extended Kalman filter, it is assumed that the system noise and measurement noise are precisely known zero-mean white noise, and when the system noise and measurement noise do not meet this condition, it may cause the error covariance matrix becomes larger, which reduces the estimation accuracy. If the parameters of the filter are manually adjusted in real time, it is very difficult to debug

Again, the speed tracking performance of the extended Kalman filter is not very good, which is mainly reflected in the fact that there is an obvious phase delay between the estimated speed and the actual speed.

Images