Biological evolution principle-based improved particle swarm algorithm

Abstract

The invention provides a biological evolution principle-based improved particle swarm algorithm. The algorithm comprises the following steps of: 1, initializing parameters of an improved particle swarm algorithm; 2, initializing a population position and a speed, calculating fitness value of each particle, and initializing global optimum; 3, judging whether to enter a crossover and variation operation or not, if the judging result is positive, entering step 4, and otherwise, skipping to step5; 4, carrying out grouping on the particles according to sorting of the fitness values, taking the first half of particles as an optimal solution group, taking the rest particles as a bad solution group, carrying out a crossbreeding operation on the optimal solution group, carrying out a variation operation on the bad solution group, combining the two groups of particles into a new population, and sorting the new population with an original population according to the fitness values so as to the first half of new excellent particles; 5, updating the particle positions and speeds, and updating individual optimal and global optimal; and 6, judging whether the improved particle swarm algorithm isconverged or achieves a maximum iteration frequency or not, if the judging result is positive, outputting a position of a global optimal solution as a solution of an optimization problem, and otherwise, returning to execute the step 3.

Description

technical field [0001] The invention belongs to the technical field of evolutionary algorithms, and in particular relates to an improved particle swarm algorithm based on the principle of biological evolution. technical background [0002] Particle Swarm Optimization (PSO) is a group cooperative random search algorithm proposed by J.Kennedy and R.C.Eberhart in recent years by simulating the foraging behavior of birds. PSO randomly generates a group of particles, calculates the fitness value to evaluate the quality of the particles, and finds the optimal solution through iteration. It searches for the global optimum by following the optimal value currently searched. During the whole search process, all particles are members of the group, so PSO is an evolutionary algorithm not based on the principle of survival of the fittest. Its concept is simple, easy to implement, fast in convergence, and strong in robustness. It is widely used in function optimization, resource allocati...

AI Technical Summary

Problems solved by technology

However, in complex optimization problems, the particle swarm optimization algorithm has premature convergence, and it is easy to fall into a local optimal solution, and the convergence speed is slow in the late iteration, and there is a problem of low efficiency.

Images