Solving dynamic parameter estimation problems using generalized oppositional BLPSO algorithm

Dynamic parameter estimation (DPE) problems are often encountered in chemical engineering systems. Due to the nonlinearity and multimodality of the DPE models, it is quite difficult to estimate the parameter values accurately. In this paper, an improved BLPSO algorithm called generalized oppositional biogeography-based learning particle swarm optimization (GOBLPSO) is proposed to address the DPE problems. In GOBLPSO, the generalized oppositional learning strategy is introduced to improve the quality of learning exemplars, and thus enhances the search performance. The GOBLPSO algorithm is applied to solve three chemical DPE problems, and experimental results indicate that GOBLPSO exhibits superior performance compared with the original BLPSO and five other meta-heuristic algorithms.