A New Partially-coupled Recursive Least Squares Algorithm for Multivariate Equation-error Systems

This paper focuses on the parameter estimation problems for multivariate pseudo-linear systems. Based on the parameters coupling characteristic of the system model, a new partially-coupled least squares algorithm is proposed. For convenience of comparison, the traditional least squares algorithm for multivariate systems is given. The proposed algorithm has better performances than the traditional algorithm. The calculation amounts of the two algorithms are analyzed, it is shown that the proposed algorithm has better computational efficiency. Two numerical simulation examples are given, and the results indicate that the proposed algorithm has better parameters identification accuracy.