Subspace approach to identification of linear repetitive processes

The aim of this paper is to propose a new approach to the identification of the LRPs based on subspace algorithms. The order of a LRP and the unknown process matrices are determined based on the input and output sequences of the actual pass and the output sequence of the previous pass. The proposed approach can be very useful not only for time invariant LRPs but also for processes with slowly evolving dynamics or processes which dynamics changes rapidly from pass to pass. Simulation examples are provided to illustrate the effectiveness of the proposed approach.