Linear Model Estimation of Nonlinear Systems Using Least-Squares Algorithm

This paper presents utilizes Least-Squares Algorithm to obtain more accurate linear models of nonlinear systems using parameter estimation. This approach generates an optimal linear model which is valid over a wide range of trajectories and converges to the desired steady-state value with no errors unlike the existing techniques. The proposed technique is very efficient and does not require storing the data. Therefore, it can easily be used and implemented with limited resources for undergraduate curriculum especially in underdeveloped countries. Most available techniques for linearization of nonlinear system are only valid about the operating point; furthermore, the knowledge of the operating point is required. The advantage of proposed technique is that the linearized model is not sensitive to the operating point; the estimation only requires the order of the system not the operating point. A physical example will be giving to illustrate the linear model of jet engines nonlinear system.