Piecewise Modeling Based on Particle Swarm Optimization Algorithm Using Numerical Data

Abstract

We propose a piecewise modeling method using numerical data. The shape of the piecewise model is a rectangular form into a state-space. The vertex values of the rectangular regions are determined using particle swarm optimization because the optimal dividing solution is a nonlinear programming problem. Using the particle swarm optimization as a multi-point search algorithm, the proposed method can determine optimal vertex values of the piecewise regions with minimal modeling errors. This paper considers some examples to demonstrate the effectiveness of the proposed method using numerical simulations.