Genetic Algorithm - Biodiversity

Genetic algorithm is an adaptive probabilistic search and global optimization algo-rithm which is formed by simulating the natural environment of biological genetic and evolutionary process. Compared with traditional optimization algorithms, it has the advantages of very strong robustness, self-organization, adaptability, global optimiza-tion and parallelism. It can be widely applied to function optimization, combinatorial optimization, automatic control, image processing and pattern recognition, etc.This article describes the basics of genetic algorithm and the current progress of the study, and provides the specific process of the binary-encoding and real-encoding genetic algorithm. It also introduces the template theory, which lays a theoretical foun-dation in mathematics for genetic algorithm. Because of losing diversity progressively in the reproductive process, genetic algorithm sometimes gives poor results. This article is focused on this issue. Considering the diversity of biological evolution, this article in-troduces an improved algorithm, which gets good optimization results. Commonly used test function Shaffer'F6 function is calculated by using the improved algorithm. It can get better results and avoid the premature convergence problem causing by inbreeding.