A novel hybrid combination optimization algorithm based on search area segmentation and fast Fourier transform

ABSTRACT

A novel hybrid optimization algorithm combining search area segmentation technique and the fast Fourier transform (HSAS/FFT) is presented to solve the numerical optimization problems. Firstly, the spectrum of each dimension of the objective function can be acquired by the FFT. The search space is segmented by using the spectrum to ensure that each subspace is unimodal. Secondly, the population of subspaces is produced and the optimal individual can be obtained by gradient descent algorithm. Finally, the local optimal solution in the optimal subspace is generated by the binary search algorithm. Make the optimal individual the new search space and repeat the process until meeting the termination condition. The proposed HSAS/FFT was tested on the CEC2017 benchmark, which evaluates the performance of the proposed algorithm on solving global optimization problems. Results obtained show that HSAS/FFT has an excellent performance and better convergence speed in comparison with some of the state-of-the-art algorithms.

KEYWORDS:

• Fast Fourier transform
• search area segmentation
• gradient descent
• binary search
• numerical optimization problems

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was financially supported by the National Natural Science Foundation of China under grant number 61663023. It was also supported by the Key Research Programs of Science and Technology Commission Foundation of Gansu Province [2017GS10817], Wenzhou Public Welfare Science and Technology Project [G20170016], General and Special Program of the Postdoctoral Science Foundation of China, the Science Foundation for Distinguished Youth Scholars of Lanzhou University of Technology under grant numbers 2012M521802, 2013T60889, and J201405, respectively.