Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Hi-index | 0.00 |
An adaptive genetic algorithm based on information entropy is presented in this paper. Unlike traditionally approach, the proposed AGA let the crossover- and mutation- rate optimized by GA itself and user need not confirm the concrete values of the two parameters. Hence, it greatly decreases the workload for iterative debugging the corresponding parameters. As a modified algorithm, this AGA has the following holistic characters: (1) the quasi-exact penalty function is developed to solve nonlinear programming (NLP) problems with equality and inequality constraints, (2) entropy-based searching technique with narrowing down space is taken to speed up the convergence, (3) a specific strategy of reserving the most fitness member with evolutionary historic information is effectively used to approximate the solution of the nonlinear programming problems to the global optimization, (4) A new adaptive strategy is employed to overcome the difficulty in confirming the genetic parameters, (5) a new iteration scheme is used in conjunction with multi-population genetic strategy to terminate the evolution procedure appropriately. Numerical examples and the performance test show that the proposed method has good accuracy and efficiency.