Comparison between Binary Particles Swarm Optimization (BPSO) and Binary Artificial Bee Colony (BABC) for nonlinear autoregressive model structure selection of chaotic data

Abstract

This paper presents a comparison between the Binary Artificial Bee Colony (BABC) and Binary Particle Swarm Optimization (BPSO) algorithm for structure selection of a Nonlinear Auto-Regressive Model (NAR) of the chaotic Mackey-Glass time-series data. Both stochastic optimization algorithms are swarm-based in nature with BABC mimicking bee colonies and BPSO mimicking the swarming behavior of birds. Recent research has suggested that the ABC algorithm has better solution quality compared to PSO. However, research on whether this advantage applies to the structure selection case in system identification has not been investigated. We conduct extensive tests to determine the convergence performance of both algorithms. It found that the BABC had managed to significantly outperform BPSO in terms of convergence consistency with a slight advantage in terms of solution quality.