Identification of interconnected multiple-input multiple-output systems: application to a multiple-input multiple-output thermal process

A new algorithm for estimation of the parameters of multiple-input multiple-output (MIMO) systems with strong interconnections between inputs and outputs. The proposed algorithm uses a new transformation of the input-output difference equation. A modified Kalman filter is employed to estimate recursively and sequentially parameters of the MIMO system and the unmeasurable partial outputs. The convergence analysis will be studied. To illustrate the performance of the proposed technique, we shall propose two examples. The first is a simulation example and the second is an application to a MIMO thermal process.