(41h) A Toolbox for Integrated Process Design, Control and Analysis

Authors: 
Gani, R., Technical University of Denmark
Tula, A., Zhejiang University
Wang, J., Zhejiang University
Chen, X., Zhejiang University
Eden, M. R., Auburn University
Constant developments of computer technologies have also led to the development of useful computer-aided tools in process engineering. While the use of steady state process simulators in industry and in academia is very common these days, the use of dynamic process simulators is less common and the use of computer-aided tools for controllability and operability analyses is even less common. At the same time, software tools for controller design and its verification through model-based systems have been developed and are routinely used in education, training and industrial applications. Also, use of software tools providing real-time optimization options is increasing rapidly. Yet, computer-aided tools for integrated process design and control, which has been attracting much attention since 1990s, have been reported mainly for problem-specific applications but not as general-purpose software tools. The main issue with the development of a tool that can offer options for integrated design-control is how to integrate the different individual tools for process synthesis-design, process simulation-optimization, process control, and process analysis tools into a single integrated tool? The challenge is how to manage the complexity of integration of the work-flow (algorithmic steps) with the associated data-flow and the computational tools needed at each step of the work-flow?

The objective of this paper is to present the architecture of a software prototype, ProCACD, that offers options for integrated process design, control and analysis. The starting point for an integrated design-control-analysis problem solution with the current prototype is an already existing process, whose design will be analysed and improved with respect to control. According to the work-flow implemented in this prototype, first steady state as well as dynamic models are developed through a modelling toolbox (MoT) to analyse the operability and controllability issues with respect to process design-control parameters. Next, the controller structure is designed in terms of variables to control (specified for process) and variables to manipulate (designed for process) through an analysis-based algorithm for design and control. Next, specific controllers (P, PI, PID, MPC, etc) are selected and the corresponding controller parameters are determined through established controller design methods (through in-house implemented methods as well as links to tools available on the internet, for example, the Taji-PID tuning system). Next, the performance of the implemented controllers is evaluated through analyses of closed-loop simulations through the modelling toolbox offering a selection of analysis options (controllability, operability, etc.) through CO-analyzer. Although the work-flow is sequential, decisions at every step are made taking into account design-control objectives such that return to an earlier step is minimized. That is, a decomposition-based solution scheme is employed to solve the complex mathematical problem, rather than a simultaneous solution of the involved model equations, which can be potentially very difficult to solve.

The prototype is useful for education and training as it guides the user not only through the specific problem solution steps, but also explains the need for data, the methods available for the specific step and how to select the most appropriate, and the solution strategy employed. The presentation will highlight some of the educational and training options available through the prototype as well as applications to problems taken form published literature.

ProCACD is a c# based software tool developed in Visual Studio. MoT (Model Toolbox) is a mathematical modeling tool that helps the user to quickly import and external process model and/or develop a new one. MoT also includes solvers for steady state and dynamic simulation as well as optimization. CO-analyzer contains calculators for a collection of performance (operability, controllability, cost) indicators. Other optional tools available to the user are ECON (process economics), LCSoft (Life Cycle Assessment factors), ProSAFE (a collection of safety indicators) and ProMonit (a tool for monitoring the process operation in terms of process measurements and process set-points).