An improved evolutionary programming algorithm

In this paper, an improved evolutionary programming algorithm for global optimization problem is proposed. In the algorithm, the direction of total force, which is based on electromagnetism theory, is received as its random search direction with the probability p, and the random search direction generated according to normal distribution is received with the probability 1-p. The improved algorithm not only overcomes the blindness of the random search for traditionally evolutionary programming algorithm, but also retains the quality of its global search. The algorithm is applied to several typical test examples. The numerical results show the improved evolutionary programming algorithm is feasible and efficient.