The Role of data prefiltering for integrated identification and model predictive control

Abstract The role of data prefiltering in integrated process identification and model predictive control is presented in this paper. A multistep ahead prediction error method is derived. Using the multistep ahead prediction error method, process models that are more suitable for multistep ahead prediction can be identified. It is also found that the new identification method performed under closed-loop condition gives better models for the design of predictive control.