Short-term and long-term thermal prediction of a walking beam furnace using neuro-fuzzy techniques

The walking beam furnace (WBF) is one of the most prominent process plants\n often met in an alloy steel production factory and characterized by high\n non-linearity, strong coupling, time delay, large time-constant and time\n variation in its parameter set and structure. From another viewpoint, the\n WBF is a distributed-parameter process in which the distribution of\n temperature is not uniform. Hence, this process plant has complicated\n non-linear dynamic equations that have not worked out yet. In this paper, we\n propose one-step non-linear predictive model for a real WBF using non-linear\n black-box sub-system identification based on locally linear neuro-fuzzy\n (LLNF) model. Furthermore, a multi-step predictive model with a precise long\n prediction horizon (i.e., ninety seconds ahead), developed with application\n of the sequential one-step predictive models, is also presented for the\n first time. The locally linear model tree (LOLIMOT) which is a progressive\n tree-based algorithm trains these models. Comparing the performance of the\n one-step LLNF predictive models with their associated models obtained\n through least squares error (LSE) solution proves that all operating zones\n of the WBF are of non-linear sub-systems. The recorded data from Iran Alloy\n Steel factory is utilized for identification and evaluation of the proposed\n neuro-fuzzy predictive models of the WBF process.