Neural network for on-line setting of parameters of active-disturbance-rejection controller of permanent magnet synchronous motor

Abstract

The invention discloses a neural network for on-line setting of parameters of an active-disturbance-rejection controller of a permanent magnet synchronous motor. The neural network is specifically implemented according to the following steps that 1, a first-order active-disturbance-rejection controller composed of a tracking differentiator, an expansion state observer and a nonlinear state error feedback control rate is designed; 2, a real-time rotating speed signal and a speed signal obtained by a photoelectric encoder are subjected to one-step delay, and output signals of the signals and an active-disturbance-rejection controller are sent to an input layer of a neural network to serve as input of the neural network; 3, combining the radial basis function neural network with the structure of three input layers, three hidden layers and one output layer with an active disturbance rejection controller, and designing a parameter regulator; and step 4, the designed neural network active-disturbance-rejection controller is used as a speed controller to carry out real-time adjustment on the system.

Description

technical field [0001] The invention belongs to the technical field of permanent magnet synchronous motor speed control, and in particular relates to a neural network for on-line tuning of parameters of an active disturbance rejection controller of a permanent magnet synchronous motor. Background technique [0002] In recent years, permanent magnet synchronous motor speed control systems have been increasingly used in many high-performance industries such as robots, automobiles, and aviation. The rapid development of these industries has also put forward higher requirements for the control accuracy of speed control systems. At present, PI control still occupies a dominant position in the actual industrial application motor control system. The PI controller is usually used in the traditional permanent magnet synchronous motor speed control system, and the control structure is simple and easy to understand and realize. However, the PI controller has the following inherent def...

AI Technical Summary

Problems solved by technology

However, the PI controller has the following inherent defects: ① cannot solve the contradiction between rapidity and overshoot at the same time; ② the same set of PI parameters cannot adapt to the working conditions of frequent and large-scale changes in speed, and the PI parameters of each speed segment need to be adjusted in sections; ③ Even if If the working point changes in a small range, it will also experience a long recovery time, and there will inevitably be static errors, etc.

Another example, Sun Bin et al. studied a speed However, the basic idea is still to use the linear system analysis tool (frequency domain method) to solve the parameter tuning problem of the ADRC controller, and only a rough theoretical parameter range can be given. Specifically, The process of setting the best parameters still needs the help of manual experience

Images