Distributed parameter identification algorithm for large‐scale interconnected systems

Abstract

This paper deals with parameter estimation problem of large‐scale systems. A recursive distributed parameter estimation algorithm, based on the minimization of the prediction estimation error method, is developed. Specifically, the class of large‐scale systems that are composed of several interconnected sub‐systems is considered. Each interconnected sub‐system is modelled by a linear discrete‐time state space mathematical model with unknown parameters. The convergence analysis is then achieved using the Lyapunov approach. The theoretical analysis and simulation results prove the effectiveness of the proposed algorithm.

Distributed joint parameter and state estimation algorithm for large‐scale interconnected systems

“…To identify the unknown parameters of the state space subsystem (1), we propose to use a recursive parameter state space estimation algorithm which is developed in our work 24 as follow:where G i is a symmetric positive matrix…”

“…In Reference [23], a distributed parameter estimation algorithm for linear discrete time interconnected state space model is presented, based in this work we develop our approach to identifier the parameters of interconnected systems where we consider the interconnection by the states and the inputs. 24 In the third scenario where both state variables and parameters are unknown, several versions of the (LS) identification algorithms are combined with the KF to produce centralized joint estimation algorithms. In Reference [25], a recursive LS with forgetting factor and an UKF, in Reference [26] a KF based recursive extended LS algorithm, while in Reference [27], a KF-based partially coupled recursive generalized extended LS algorithm.…”