Performance-optimized applied identification of separable distributed-parameter processes

This note studies practical algorithms for parametric identification of cro ss-directional processes from input/output data. Instead of working directl y with the original two-dimensional array of the high-resolution profile sc ans, the proposed algorithms use separation properties of the problem. It i s demonstrated that by estimating and identifying in turn cross-directional and time responses of the process, it is possible to obtain unbiased least -square error estimates of the model parameters. At each step, a single dat a sequence is used for identification which ensures high computational perf ormance of the proposed algorithm. A theoretical proof of algorithm converg ence is presented. The discussed algorithms are implemented in an industria l identification tool and this note includes a real-life example using pape r machine data.