In most chemical process designs, the designer begins by analyzing simple structures using approximate models. Gradually as the design evolves, more detailed models are used to explain more complex interactions which can make the difference in the success of a design. These complex phenomena are often the reason that a new design has the potential to be more effective; for example, more energy or more cost efficient. However, models of more complex interactions usually have a richer solution space; that is, they exhibit more solutions for a given set of specifications. The number and stability of the solutions varies with the specifications and with the parameters of the model. In process design, problems can arise when algorithms converge on a solution that is not physically correct. In the design stage, experimental data often are not available and the designer may continue to work based on a wrong solution. The Principal Investigator proposes to create a prototype software environment to permit experimentation with the design of control structures for complex systems. The PI will develop new techniques for solving both steady state and dynamic simulation problems to demonstrate the utility of these techniques in the design of chemical processes. Emphasis will be placed on the analytical and algorithmic methods that rationalize the design process.
