Comparative Analysis for System Modeling Using System Identification Techniques

System modelling is an essential task for controller design. System identification is widely used for dynamic systems modelling and provides high performance practical dynamic systems modelling. In this paper, a least square approach and recursive least square based identification techniques are proposed for comparative analysis of dynamic modelling of a single -input single-output tank process. Chirp and multi-sine signals are utilized for system modelling as they provide the required data for capturing the system dynamics. A comparative analysis is carried out with nonlinear recursive least square techniques such as ARX model, ARMAX model, Box-Jenkins model, Output-Error model for evaluating the proposed approach effectiveness. The ARX method provided high performance for single-input single-output tank system identification in the presence of system noise.