Autoregressive model fitting for control
The use of a multidimensional extension of the minimum final prediction error (FPE) criterion which was originally developed for the decision of the order of one-dimensional autoregressive process  is discussed from the standpoint of controller design. It is shown by numerical examples that the criterion will also be useful for the decision of inclusion or exclusion of a variable into the model. Practical utility of the procedure was verified in the real controller design process of cement rotary kilns.
KeywordsController Design Autoregressive Model Manipulate Variable Prediction Error Variance liSr
Unable to display preview. Download preview PDF.
- Akaike, H. (1970). On a semi-automatic power spectrum estimation procedure,Proc. 3rd Hawaii International Conference on System Sciences, 974–977.Google Scholar
- Akaike, H. (1970). On a decision procedure for system identification,Preprints, IFAC Kyoto Symposium on System Engineering Approach to Computer Control, 485–490.Google Scholar
- Masani, P. (1966). Recent trends in multivariate prediction theory,Multivariate Analysis, (P. R. Krishnaiah ed.), Academic Press, New York, 351–382.Google Scholar
- Otomo, T., Nakagawa, T. and Akaike, H. (1969). Implementation of computer control of a cement rotary kiln through data analysis,Preprints, IFAC 4th Congress, Warszawa.Google Scholar
- Wong, K. Y., Wiig, K. M. and Allbritton, E. J. (1968). Computer control of the Clarksville cement plant by state space design method,IEEE Cement Industry Technical Conference, St. Louis, U.S.A., May 20–24.Google Scholar