An improved harmony search algorithm based on teaching-learning strategy

Harmony Search (HS) algorithm inspired by the music improvisation process is a swarm intelligent algorithm. In this paper, an improved global harmony search algorithm which is based on Teaching-Learning-Based-Optimization (TLBO) algorithm is proposed to solve the complex high-dimensional optimization problems. In the proposed algorithm, harmony memory consideration, teaching-learning strategy, local pitch adjusting and random operation method are adopted to improve the search ability of HS. The parameters are adjusted dynamically in terms of the process of iterations. Finally, the proposed algorithm is tested and compared with three other state-of-the-art HS optimization algorithms. The experimental results on 7 high-dimensional benchmark functions show that the HSTL method has strong convergence, and has better balance capacity of space exploration and local exploitation on high dimension complex optimization problems.