Fuzzy-Adaptive Matrix-Based PSO with Group Learning

The particle swarm optimization (PSO) algorithm is a random search algorithm that simulates biological activities in nature and swarm intelligence. Traditional PSO algorithms still suffer from their inherent problems such as long computing time and premature convergence when dealing with a large number of decision variables. How to reduce the computing time while ensuring its accuracy is one of the main research issues in the literature. This paper proposes a PSO algorithm with matrix-based fuzzy adaptation and group learning. Firstly, to reduce the computing time, we propose a strategy to balance the global search ability and the local search ability by controlling the number of iterations adaptively. Second, to accelerate the convergence speed of the algorithm, we further propose a group learning mechanism. Along with the above two considerations, the proposed algorithm is able to achieve a promising balance between exploration and exploitation and, consequently, improves the performance. Through extensive comparison experiments on six classic functions, we show that our proposed PSO algorithm can achieve competitive or even better performance than several state-of-the-art methods.