Model parameters identification for excess oxygen by Standard Genetic Algorithm

Abstract

In this paper, a realistic excess oxygen model parameter identification by Standard Genetic Algorithms (SGAs) is proposed and demonstrated. The realistic excess oxygen model is developed by three sub-model; air-fuel ratio conversion model, dynamic continuous transfer function and excess oxygen look-up table to characterise the real excess oxygen plant's numerical data. The predetermined time constant approximation method is applied on 1st, 2nd, 3rd, 4th and 5th model orders for an initial value estimation with SGAs. For an optimal model order assessment and selection, the information criteria are applied. The simulation results assured that the 4th order continuous transfer function as a realistic model well characterises the real excess oxygen plant's response.