Fruit fly-leapfrog-based linear synchronous motor control method employing fuzzy neural network PID

Abstract

The invention discloses a fruit fly-leapfrog-based linear synchronous motor control method employing a fuzzy neural network PID. A control system comprises an outer speed loop and an inner current loop, the current loop is designed into a traditional PID adjuster and the speed loop is designed into a fuzzy neural network PID speed controller; a fruit fly-leapfrog hybrid algorithm is formed through fusing a fruit fly optimization algorithm and a leapfrog algorithm; structure parameters of a fuzzy neural network are adjusted and optimized in real time, optimal parameters kp, ki and kd suitable for the PID controller are output and self-adaptation and intellectualization of speed control on a permanent-magnet linear synchronous motor are achieved. Simulation analysis and experiment results show that the control accuracy and the disturbance resistance of a permanent-magnet linear synchronous motor control system can be improved by adopting a fuzzy neural network PID speed control strategy optimized on the basis of the fruit fly-leapfrog hybrid algorithm, and an excellent control effect is achieved.

Description

technical field [0001] The invention relates to the control field of a permanent magnet linear synchronous motor, in particular to a speed control method of a permanent magnet linear synchronous motor based on a fruit fly-leapfrog hybrid algorithm to optimize a fuzzy neural network PID. Background technique [0002] Compared with the rotary motor, the linear servo system composed of permanent magnet linear synchronous motor (referred to as PMLSM) can directly convert electrical energy into linear motion mechanical energy because it omits all the mechanical transmission links from the rotary motor to the workbench. It does not require any intermediate conversion mechanism. It has the advantages of simple structure, high speed, high precision, high durability, and direct drive. It has been widely used in industrial control fields such as robots, high-precision CNC machine tools, and semiconductor manufacturing. [0003] At present, in the permanent magnet linear synchronous mo...

AI Technical Summary

Problems solved by technology

[0016] From the analysis of existing research literature, the learning algorithms used to optimize the fuzzy neural network PID controller mainly adopt BP algorithm, genetic algorithm, particle swarm optimization algorithm, etc. These algorithms have long learning process, easy to mature, and easy to fall into local problems in learning and training. Problems such as optimality and poor robustness, learning speed, learning accuracy and optimization efficiency are not high, which directly affect the control performance of the permanent magnet linear synchronous motor fuzzy neural network PID controller

Images