9528675 Venkatasubramanian Purdue Univ. The chemical processing industry (CPI) constitutes one of the major competitive strengths of the US manufacturing sector, but it is also the source of the of the majority of the total US industrial output of hazardous wastes. In order to maintain and enhance the global competitiveness of the CPI while ensuring a safe, clean, and sustainable environment, waste minimization and processing measures need to be fully integrated within the overall design and operation of future chemical manufacturing facilities. The goal of this work is to develop and demonstrate computing based modeling, optimization and intelligent control methodologies which will facilitate such integrated designs and operations. The work is aimed at the key issues which are characteristic of the batch thrust, the PI's plan to investigate model and optimization based process design methodology which takes into account a variety of relevant issues. Under the continuous process thrust, they will develop tools and approaches which facilitate joint consideration of design and operational decisions and the monitoring and diagnosis of control systems. The technical work includes development of efficient methods for the formulation and solution of highly structured mixed integer linear programming (MILP) systems, control relevant metrics suitable for inclusion into steady state design methodology, mathematical programming based approaches to large scale model predictive control (MPC), and the formulation of control systems performance monitoring problems in terms of a hierarchical, qualitative shape analysis formalism. The key outcomes of this project will be a set of prototype tools demonstrated on a set of industrial benchmark problems which, it is hoped, will energize the industry to pursue environmentally sustainable designs and plant operations.
