PID parameter tuning method for particle swarm optimization based on inertia weight cosine adjustment

Abstract

The invention discloses a PID parameter tuning method for particle swarm optimization based on inertia weight cosine adjustment, which relates to the technical field of particle swarm algorithms. Themethod comprises the following steps: (1) population initialization operation is carried out; (2) a particle swarm individual is decoded to a proportion Kp, an integral Ki and a differential Kd; (3) asimulink model is operated to select an integrated time absolute error (ITAE) criterion as a fitness function of the algorithm; (4) the fitness value of the initial population is calculated accordingto the fitness function, and an individual extreme value pBest and a global extreme value gBest are obtained; (5) the inertia weight is dynamically adjusted according to an inertia weight cosine adjustment formula, the speed and the position are calculated according to an evolutionary iteration formula, and a new-generation particle swarm is thus obtained; (6) the fitness value of a particle swarm individual after updating is calculated by the ITAE criterion; (7) if the termination condition of the maximum iteration number or the minimum precision is reached, an eighth step is carried out, and if the termination condition is not reached, the fifth step is carried out; and (8) the operation is terminated, and a global optimal value is obtained.

Description

technical field [0001] The invention relates to the technical field of particle swarm algorithm, and more particularly to a PID parameter tuning method of inertia weight cosine adjustment particle swarm optimization algorithm. Background technique [0002] In the current industrial control, the most widely used controller is the PID controller (Process Identifier), whose function is to make the error approach the required smaller and smaller direction, so as to achieve the control accuracy required by the industrial control. It has the advantages of simple structure, convenient adjustment and the like. Usually, the controller is composed of proportional (Kp), integral (Ki) and differential (Kd) linear components of the deviation, and controls the controlled object. The proportion Kp is proportional to the deviation signal of the reaction control system. Once the deviation is generated, The controller immediately produces a control effect to reduce the deviation; the integra...

AI Technical Summary

Problems solved by technology

In order to improve this defect, researchers have successively proposed improved methods such as the linear weight loss (LDIW) strategy, the random inertia weight (RIW) strategy, and the inertia weight sinusoidal adjustment strategy starting from the inertia weight. The efficiency has been improved to a certain extent, but there are still defects in how to balance the local and global search capabilities and avoid falling into the local optimal solution in the later stage of the algorithm

Images