Parameter self-tuning method for fuzzy PID controller

Abstract

The invention discloses a parameter self-tuning method for a fuzzy PID controller, comprising steps of calculating original quantification factor numerical value combination according to deviation of a control system, a deviate changing rate and basic discourse domains and fuzzy disclosure domains of ratio, differential and integral coefficient in the PID control, reasonably choosing interval radius of the interval to obtain value-taking intervals of all quantization factors to be optimized by taking all original quantization factor values to be centers of interval, evenly selecting values in all value-taking intervals to obtain different quantization factor value combinations, testing control performance of the control system under the combination of all quantization factor values; taking a quantization factor value combination and a corresponding control performance value as a sample data, repeating the test to obtain enough sample data used as training data for an extreme learning machine, training the control system model in an off-line state, and perform optimizing on the system off-line model by using a non-dominated sorting genetic algorithm of an elitist strategy in order to obtain the quantization factor value combination enabling the control performance to be multi-object optimization.

Description

technical field [0001] The invention relates to the field of motor control, in particular to a parameter self-tuning method for a fuzzy PID control system of a single-winding magnetic levitation switched reluctance motor. Background technique [0002] The magnetic levitation switched reluctance motor combines the magnetic levitation technology with the switched reluctance motor. On the basis of inheriting the advantages of the general magnetic levitation motor, such as no friction, no wear, high axial space utilization rate, and large rotor critical speed, it gives full play to the advantages of the switched reluctance motor. The high-speed superiority and adaptability to harsh environments, and through the active control of the radial force, the vibration and noise problems caused by the unbalanced magnetic pull of the switched reluctance motor are effectively improved. At present, in most cases, the magnetic levitation switched reluctance motor with double winding structur...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem that the quantization factor optimization target in the fuzzy PID control is mostly a single target, the purpose of the invention is to introduce the multi-objective optimization problem into the field of fuzzy control parameter optimization, and provide a method for the self-tuning of the quantization factor of the fuzzy PID controller, so that It has the optimal quantitative factor combination under multi-objective optimization, and realizes the optimal comprehensive control performance index for actual engineering requirements

Images