New fruit fly optimization algorithm with joint search strategies for function optimization problems

Abstract

Listen

Translate article

An enhanced fruit fly optimization algorithm (FOA) with joint search strategies named JS-FOA is proposed to optimize continuous function problems. First, a collaborative group search, which includes a new parameter, is conducted to obtain the critical value. Second, a new search strategy similar to biological memory, namely, memory move direction, is proposed to improve solution accuracy. Third, a gradient descent search is used in the collaborative group search to ensure that it does not fall into a local optimum. Finally, a new function, which is similar to the excitation function in a neural network, is proposed to combine the three search strategies. To test the robustness and convergence of the proposed JS-FOA, we used 29 complex continuous benchmark functions. Results show that the proposed JS-FOA outperforms other heuristic algorithms for most functions. The performance of JS-FOA is also evaluated for different parameter values and the results show that parameter values affect convergence speed within a certain range, but do not change the convergence accuracy for the continuous benchmark functions. The proposed JS-FOA may potentially solve high-dimensional optimization problems.