Research On Parameter Identification Algorithm Of Built-in Permanent Magnet Synchronous Motor

IPMSM is widely used in the power system of modern equipment because of its advantages such as high power density,small size and high efficiency.During the operation of the motor,the working temperature of the motor and electromagnetic effects will cause changes in the motor parameters.It is difficult to accurately model the IPMSM.In engineering,approximate calculations are often used to limit the efficiency of the motor.In order to establish an accurate mathematical model The online parameter identification algorithm is widely used in the IPMSM control system,and the RLS algorithm is more accepted by people.When the traditional RLS algorithm performs IPMSM parameter identification,only the d-axis stator voltage discrete equation is used as the identification equation.The sampling data utilization rate is low and the dependence on a single equation is strong.The addition of the forgetting factor in the recursive equation limits the old The influence of data on the parameter recognition results increases the convergence speed of the RLS algorithm while also increasing the sensitivity of the algorithm to noise,and the forgetting factor processes the different parameters of the motor in a unified manner when processing the RLS algorithm covariance matrix,which causes the algorithm to respond The processing power of different changes of various parameters becomes weak,which reduces the overall performance of the algorithm.In order to improve the application performance of the RLS algorithm for online parameter identification of IPMSM,in view of the above-mentioned problems with the traditional RLS algorithm,this paper has conducted the following studies:(1)On the basis of the original RLS algorithm,theqshaft stator voltage equation is added,and the discreted-q shaft equation is organized into a multi-input multi-output system equation with the same parameters and different input and output variables.(2)At the same sampling time,the IPMSM parameters are identified at the same time according to the discrete equation of thed-q shaft stator voltage,so that the two identification equations are related to the parameter matrix by their respective covariance matrix,forming aninner loop recursive algorithm with the same sampling time The outer loop recursive algorithm at the sampling time weakens the dependence of the identification results on a single equation.(3)Analyze the influence characteristics of the forgetting factor value on the convergence speed of parameter identification and the effect of anti-noise ability to meet the different needs of the forgetting factor value at different stages of the RLS convergence process,and design adaptively adjusted dynamic forgetting Factor to improve the overall performance of the algorithm.(4)Based on the covariance matrix equation in the recursive formula of the RLS algorithm,analyze the influence of forgetting factors on the estimated values of IPMSM parameters,design a dynamic forgetting factor matrix that can adjust different parameter changes separately,and improve the algorithm to cope with changes in different parameters Resilience.(5)Build a vector control model of permanent magnet synchronous motor in Matlab /simulink and embed the designed RLS improved algorithm module.By setting the changes of different parameters of IPMSM,the simulation results verify the effectiveness of the improved algorithm.In this paper,a double-loop RLS algorithm suitable for IPMSM is designed and improved,and a dynamic forgetting factor matrix is designed based on the traditional forgetting factor.When the change rate of each parameter is different,the size of the forgetting factor is automatically adjusted and the covariance matrix data is processed to achieve the difference.The purpose of adjusting the data of each saturation parameter is to effectively alleviate the contradiction between the convergence speed of parameters and the ability of anti-noise interference at the same time,which has certain reference significance for the research field of IPMSM parameter recognition algorithm.