Multi-channel Spatio-Temporal Modeling for Hammerstein Distributed Parameter Systems

A multi-channel spatio-temporal Hammerstein modeling approach is presented in this chapter. As a special case of the model described in Chapter 4, a spatio-temporal Hammerstein model is constructed with a static nonlinearity followed by a linear spatio-temporal kernel. When the model structure is matched with the system, a basic single-channel identification algorithm with the algorithm used in the Chapter 4 can work well. When there is unmodeled dynamics, a multi-channel modeling framework can provide a better performance, because more channels used can attract more information from the process. The modeling convergence can be guaranteed under noisy measurements. The simulation example and the experiment on snap curing oven are presented to show the effectiveness of this modeling method.