Study on an Evolutionary Hybrid Algorithm
Wenming Cheng · Jisuanji fangzhen · 2007
To solve some problems of the simple algorithm used for numerical optimization problems,a hybrid algorithm based on evolutionary algorithm is provided.It improves the crossover operator,and combines the mutation operator with simulated annealing to form the simulated mutation operator.In order to improve the precision and convergence speed,the self-adaptive trait of evolutionary strategy is adopted.The(μ,λ)select operator is applied to algorithm to increase the probability of escaping from the local optima,and the elites saving strategy can guarantee the global convergence of algorithm.The two typical test functions are used to test the algorithm,the results indicate that the algorithm has the ability of escaping from the local optima,and converging in the global optimum with high efficieney and precision.