a modified differential evolution algorithm with a balanced performance for exploration and exploitation phases

Recently optimization algorithms are proposed to find the best solution for complex engineering problems. these algorithms can search unknown and multidimensional spaces and find the optimal solution in the shortest possible time. in this paper, we present a new modified differential evolution algorithm. optimization algorithms typically have two stages of exploration and exploitation. exploration refers to global search, and exploitation refers to local search. we used the same differential evolution (de) algorithm. this algorithm uses a random selection of several other search agents to update the new search agent position, which makes the search agents continually have random moves in the search space, which refers to the exploration phase. still, there is no mechanism considered explicitly for the exploitation phase in the de algorithm. in this paper, we have added a new formula for the exploitation phase to this algorithm and named it the balanced differential evolution (bde) algorithm. we tested the performance of the proposed algorithm on standard test functions, cec2005 complex, and combined tests functions. we also apply the proposed algorithm to solve some real problems to demonstrate its ability to solve constraint problems. the results showed that the proposed algorithm has better performance and competitive performance than the new and novel optimization algorithms.