Two-Dimensional (2D) Particle Swarms for Structure Selection of Nonlinear Systems
For researchers in nonlinear system identification, this work offers a new optimization-based structure selection method that explicitly handles model cardinality, though it is an incremental improvement over existing swarm-based approaches.
The paper introduces a 2D particle swarm optimization (2D-UPSO) method for structure selection in nonlinear system identification, which incorporates cardinality information into the search process. The method successfully identifies correct structures in simulated and real-world wave-force data, outperforming GA, BPSO, and OFR methods.
The present study proposes a new structure selection approach for non-linear system identification based on Two-Dimensional particle swarms (2D-UPSO). The 2D learning framework essentially extends the learning dimension of the conventional particle swarms and explicitly incorporates the information about the cardinality, i.e., number of terms, into the search process. This property of the 2D-UPSO has been exploited to determine the correct structure of the non-linear systems. The efficacy of the proposed approach is demonstrated by considering several simulated benchmark nonlinear systems in discrete and in continuous domain. In addition, the proposed approach is applied to identify a parsimonious structure from practical non-linear wave-force data. The results of the comparative investigation with Genetic Algorithm (GA), Binary Particle Swarm Optimization (BPSO) and the classical Orthogonal Forward Regression (OFR) methods illustrate that the proposed 2D-UPSO could successfully detect the correct structure of the non-linear systems.