Research On Identification Algorithm Of AC Servo System Test Parameters

Changes in the mechanical parameters and electrical parameters of the permanent magnet synchronous motor will directly affect the control performance of the servo system,how to achieve efficient parameter identification,and through the parameter identification to make the test system to simplify the test steps and improve the test efficiency is the current servo test The mainstream research direction in the system.This dissertation mainly studies the parameter identification on the high-performance servo test system.The test system can not only complete the test items of the motor and the servo system,but also can realize the speed filtering and parameter identification functions with high efficiency.This article mainly makes the following researches :In order to optimize the parameter identification test technology of s ervo test system,the servo system is simulated and modeled,the actual hardware device is used to measure the speed,and the EKF algorithm is used to filter out the environmental noise in the encoder output information,providing accurate speed for the parameter identification algorithm.The value is verified by simulation to verify the feasibility and effectiveness of this method.The variable gain Landau algorithm and the EKF algorithm are used to build the simulation model for the two inertia moments.T he simulation results are compared and analyzed.The variable-gain Landau algorithm can solve the contradiction between the convergence speed and the identification accuracy.The algorithm adjusts the value of the adaptive gain based on the identification bias to obtain the identification results with short convergence time,high estimation accuracy,and good dynamic capability.The EKF algorithm takes into account the effects of model errors and system noise,reasonably selects the covariance matrix and only recognizes the inertia parameters,and can solve the shortcomings of poor dynamic performance and large amount of calculation in the past applications.The algorithm can account for changing system parameters and environmental noise.All have good robus tness and can get a good signal-to-noise ratio.For the problem of identifying the stator resistance,the direct-axis inductance and the permanent magnet flux chain of the FFRLS-MRAS algorithm,it is difficult to design the adaptive law.A variable-forgetting factor and variable weighting factor combined with the variable-rate discount least squares method is proposed in this dissertation.The problem of saturation of data,long convergence time,and poor identification accuracy in the permanent magnet synchronous motor parameter identification process is solved,and the dynamic tracking,identification accuracy and speed of electrical parameter identification are improved.According to the actual application of servo test system required performance indicators and test items for analysis,design a specific program,and detailed analysis of the test function.Using the virtual instrument technology and Matlab mixed programming,the software of the upper computer software for speed filtering and parameter identification is set up,and the servo-testing system test platform based on motor drag is established.The experimental results show that the servo test system can not only achieve the required performance.The test can also obtain the parameter identificati on results with good anti-interference,high recognition accuracy and fast convergence speed.