Optimization of Identification Structure Parameters Based on Recursive Maximum Likelihood Iteration

Abstract

The probability distribution of the maximum likelihood method requires that the probability distribution of data is known and obeys the Gauss distribution. The structural parameters of the equivalent prediction error have great influence on the modeling accuracy of the system model. In order to improve the modeling accuracy, a recursive maximum likelihood iterative identification method is proposed to optimize the identification structure parameters. The simulation results verify the feasibility and effectiveness of the proposed identification and modeling method.