Research On BLDC Control Method About Improvement And Gain Optimization Of Line Back-EMF Observer

The brushless DC motor has been widely used in various fields since its inception due to its excellent speed regulation performance,concise internal structure and high use efficiency.Research on its control method has also become a research hotspot in the field of modern electric driver.The common control method of the brushless DC motor is to install a mechanical sensor to determine the rotor position and commutate the three-phase stator winding to achieve its rotation.However,position sensors will limit the further promotion and application of brushless DC motors due to system size,circuit complexity,anti-interference,and control costs.Therefore,the sensorless control technology of brushless DC motors has become an important research direction at home and abroad.The key issue of position sensorless control is the collection of motor rotor position information.After reading many related literatures,and studying several mainstream sensorless control methods at home and abroad,the advantages and disadvantages of the existing methods are summarized.Taking the brushless DC motor control system as the research object,by analyzing its overall structure,mathematical model and control principle,etc.After comprehensive consideration,the state observation theory is used to study and design a line back-EMF state observer to estimate the motor speed,rotor position and other related information,based on which the sensorless control of the brushless DC motor is performed.At the same time,in view of the general defects in the above traditional state observer design theory,such as including a large number of high-frequency interference components,switching gain selection needs to be high-frequency and constant,and the estimated state quantity contains a lot of buffeting.A new structure combining linear and nonlinear error feedback is proposed,which greatly reduces the buffeting in the estimated speed and the high-frequency disturbance component in the estimated back EMF,accelerates the convergence of the estimated state quantity and improves the observer Stability.In addition,in order to further determine the optimized observer gain parameters and the optimal values of the adjustable parameters in the internal feedback term function,this paper abandons the traditional method of pole configuration to obtain gain values,and consults a large number of optimization algorithms and chooses a novel Meta-heuristic optimization algorithm——Gray Wolf Optimization algorithm,using the ITAE of motor speed error as the optimization performance index to optimize and calculate the relevant parameters of the observer to ensures the stability and accuracy of the observer in the process of estimating the state quantity,reduces the error between the observer and the actual state quantity,and ensures the optimization of the motor speed output.Finally,MATLAB / Simulink is used as the simulation platform.Compared with the traditional observer and different working conditions,the improved and optimized observer proposed in this project has greatly improved in its convergence speed and reduced the fluctuation of the absolute mean and peak errors,and the optimal control of the brushless DC motor without position sensor is realized.