System identification and machine setup of color xerographic machines using multivariate modeling and multiobjective optimization

This paper describes an approach for performing system identification of the xerographic process and developing a model of the overall system. The experiment design is performed using orthogonal arrays. A nonlinear system model defining a vector mapping between the actuators and performance metrics is constructed using multivariate adaptive regression splines (MARS). These models provide a basis for providing automated and optimal machine setup on color machines. Multiobjective optimization techniques based on an underlying adaptive simulated annealing approach are discussed which are then used for determining the Pareto-optimal setpoints. The overall approach is applied to perform the automated setups of a high-end color machine, namely the Docucolor40. Experimental results show that high quality machine setups can be performed in an efficient and automated manner. The use of these setpoints as the basis of linearization about a nominal setpoint for use in linear control algorithms are also discussed.