A Modified Extended Bayesian Method for Parameter Estimation

摘要

This paper presents an extended Bayesian method for parameter estimations. In this method, the mean value of the a priori estimation is set as the value of the estimated parameters in previous iteration step. In this way, the parameter covariance matrix can be automatically updated during the estimation procedure, which avoids selecting empirical parameters. Because the extended Bayesian method could be regarded as a Tikhonov regularization, this new method is more stable than lease-squares method and maximum likelihood method. The validity of the proposed method is illustrated by an academic example as well as a real engineering example.