A Modified PSO Algorithm for IIR Digital Filter Modeling

• Published in: Journal of circuits, systems, and computers Vol. 27; no. 5; p. 1850073
• Main Author: Chang, Wei-Der
• Format: Journal Article
• Language: English
• Published: Singapore World Scientific Publishing Company 01.05.2018; World Scientific Publishing Co. Pte., Ltd
• Subjects: Algorithms; Computer simulation; Control systems design; Digital filters; Digital systems; IIR filters; Impulse response; Model reduction; Modelling; Particle swarm optimization; multiple subpopulations; IIR digital filter; PSO algorithm
• ISSN: 0218-1266; 1793-6454
• DOI: 10.1142/S0218126618500731

Modeling for an unknown plant is a very important and primary task in the control system engineering. By using this established model, an adequate controller can be further designed to meet certain control specifications. This paper will propose a new modeling method which is based on the infinite impulse response (IIR) digital filter structure. A modified particle swarm optimization (PSO) with multiple subpopulations is utilized to solve for the IIR filter coefficients to model an unknown digital system. The developed PSO algorithm is able to overcome the multimodal error surface problem that often occurs in the system modeling of IIR digital filter. Finally, three different kinds of digital systems are examined to demonstrate the feasibility of the proposed method, including the same system order and reduced order structures. It is concluded from several simulation results that the proposed method has an excellent performance on solving the IIR digital modeling problem.