High-order Least Squares Method for Parameter Identification of Nonlinear Input and Output Systems
The traditional least squares method usually uses the first-order linear solution of the nonlinear model when identifying the parameters in the nonlinear model,which will introduce errors in the process,resulting in the divergence of the model parameters of the system to be solved.When the nonlinearity of the system model is stronger,the divergence of the model parameters to be identified will be more serious.Thus,the estimation accuracy of the model parameters to be identified is inaccurate.In this study,the high-order expansion of the objective function is adopted to reduce the introduction error and improve the accuracy of the parameters to be identified in the system model.Finally,numerical simulations confirm the effectiveness and feasibility of the proposed method.
least squaresTaylor unfoldshigher-order termsparameter identification
万方数据
维普
