Multi-objective and multi-modal particle swarm optimization method based on Bayesian adaptive resonance

Abstract

The invention discloses a multi-objective and multi-modal particle swarm optimization method based on Bayesian adaptive resonance, using Bayesian adaptive resonance theory to divide all particles into several populations; sort the particles; use the individual optimality of particles and the global optimality of the population to update the particles in the population; connect the non-dominated solution sets of various populations end to end to form a closed ring topology, and use particle swarm optimization based on ring topology The algorithm performs local exploration; repeat the above two update and exploration processes until the termination condition is satisfied, and output all non-dominated solution sets and Pareto frontiers. The present invention is applicable to the optimization of multi-objective and multi-modal problems. It can not only find the distribution of Pareto front in the target space, but also find the corresponding Pareto optimal solution set in the decision variable space, and provide a redundant backup method. , Improve the reliability of engineering practice activities.

Description

technical field [0001] The invention relates to the technical field of optimization algorithms, and more specifically relates to a multi-objective multi-mode particle swarm optimization method based on Bayesian adaptive resonance. Background technique [0002] In real life, there are often problems with multiple optimization objectives that conflict and restrict each other. For example, in the design process of an aircraft, it is necessary to ensure the stability of the aircraft and to pursue the maneuverability of the aircraft. Stability and maneuverability are two At the same time, the solution to the same goal may contain multiple solutions. For example, different aircraft design schemes may obtain the same stability and maneuverability. Such problems are called multi-objective multi-modal problems. The space composed of the objective function is called the target space, and the space composed of the variables used to generate the objective function is called the decision...

AI Technical Summary

Problems solved by technology

For multi-modal problems, niche particle swarm optimization algorithm is a better solution; for multi-objective problems, particle swarm optimization algorithms based on k-means clustering and Euclidean distance clustering have also been proposed, but now Some clustering-based particle swarm optimization algorithms often need to pre-set the number of clusters according to the number of solutions, but in reality it is difficult to know in advance how many solutions there are for multi-objective multi-modal functions, so the past methods generally use different Experiment with the number of clusters, and select a better number of clusters according to the experimental results, which brings great uncertainty and problem dependence to the optimization problem.

Images