Application of the Meta-Multiparametric methodology to the control of emissions in the industry under continuous and discrete uncertain parameters

Production scheduling becomes a highly complex task in the process industry, especially when it is necessary to take immediate decisions for unexpected situations, since most industrial systems exhibit nonlinear behavior and many uncertain parameters (UP). This paper presents an industrial case study where immediate reactive production scheduling is required to control unexpected changes in the plant emissions profile associated to the existence of bounded UP. Since only a very time consuming Simulation Based Optimization (SBO) system of the plant is available to find the new optimal schedule which minimizes the emissions' peaks, the proposed solution relies on the use of Meta-Multiparametric (M-MP) techniques, which will off-line use the SBO system to create a model which, given any initially unknown situation, will describe the changes to be introduced in the system to limit these emissions' peaks. The results show that the M-MP techniques used are able to model the optimum input-output relations behind the SBO data and that, in contrast to existing similar procedures, both continuous and discrete UP (like batch sequence information) can be considered.