Multivariable System Identification of Cement Mill Process Using NARX

This paper deals with the identification of MIMO cement millprocess using Non-linear Autoregressive with Exogenous Inputs(NARX) models with wavelet network. NARX identification,based on a sequence of input/output samples, collected from a realcement mill process is used for black-box modeling of non-linearcement mill process. The NARX model is considered for twoinputs and two outputs of seven hours of data with sample time offive seconds. In order to assess the suitability of identified model,Model validation tests are performed by means of auto-correlationfunction and cross-correlation function. The fitness of NARXidentified model is compared with ARX model. The identifiedNARX model is converted to discrete transfer function and thedynamic characteristic of the identified model are evaluated andresults are given.

Keywords:

MIMO , System identification , Cement mill , NARX