Recognition of multi-population genetic algorithm _ pattern Recognition

Multi-group Genetic algorithm is the intelligent algorithm for writing papers, the advanced version of genetic algorithm.

Its flowchart is as follows:

The SGA is a common standard genetic algorithm, it can be seen that the multiple groups of different groups of genetic algorithms in the various control parameters of the SGA to maintain the diversity of the population.

The control parameter is the crossover probability pc and the mutation probability PM, its value determines the algorithm global search and the local search ability equalization, can calculate according to the following type:

In the formula: PCO,PMO are initial crossover probability and mutation probability, g is population number, C,m is crossover, the interval length of mutation operation, Frand is the function that produces random number. It is found that it is usually randomly generated in [0.7,0.9] interval after full study. Usually randomly generated in [0.001, 0.05] intervals.

Immigrant operator and artificial selection operator: Set up immigrant operator, that is, the optimal individual in the source population replaces the worst individual in the target population, and achieves the goal of multi population coevolution. The function of the artificial selection operator is to select the optimal individual in various groups and put them into the elite population to keep the optimal individuals from being destroyed and lost.

MPGA determines the termination of the algorithm according to the elite population.