A hybrid intelligent technique for model order reduction in the delta domain: a unified approach

Abstract

This paper presents a unified method for reduced-order modelling of higher-order discrete-time systems in the complex delta domain. The approach is unified in the sense that it is fundamentally discrete time, but at a high sampling frequency converges to its continuous-time counterpart. New hybrid algorithms with a blend of grey wolf optimizer (GWO) and chaotic firefly algorithm (CFA) using iterative chaotic map have been proposed for the order reduction of large-scale systems in the delta domain. GWO explores the entire search domain while CFA carries out local search in the proposed hybrid technique with different tuning parameters of FA improvised by one-dimensional iterative map, thus improving the convergence speed and the quality of the solutions. Two examples of single-input single-output systems are taken up to establish the usefulness of the proposed method. The hybrid approach not only yields good matching in the delta domain, but also delivers quality solutions (minimum fitness value) and provides faster convergence speed than the existing techniques. The performances of the reduced systems are also compared with their respective original systems as well as reduced systems reported in the literature in terms of time-domain and frequency-domain parameters. The proposed topology also proved its superiority in terms of some benchmark error indices well established in the literature of systems approach and control.