Differential Evolution(DE)is a simple intelligent algorithm with high-performance.It is widely used in applications because of its relatively few parameters,high accuracy and robustness of calculation re-sults.However,DE tends to fall into local optimal solutions,and population variation speed and experi-mental results depend on mutation and crossover of parameter settings.Therefore,we propose an adaptive improved differential evolution algorithm,which can effectively improve the algorithm's optimization-see-king ability by controlling the initialized populations and adaptive parameters.The algorithm is tested on 11 test functions and compared with four representative algorithms(CLPSO,DE,jDE,SDE),and the results show that the algorithm has advantages in terms of accuracy and shows excellent robustness.
维普
万方数据