(51d) Component Based Development of Application Specific Computer-Aided Tools

Kumar Tula, A., Auburn University
Eden, M., Auburn University
Gani, R., Technical University of Denmark
Chen, X., Zhejiang University
Chemical engineering is in constant evolution and access to computational resources are changing the way chemical engineering problems are addressed. Process simulation, which emerged in the 1960s, has become one of the great success stories in the use of computing in the chemical industry. For instance, steady-state simulation has largely reduced experimentation and pilot plant testing in process development for commodity chemicals. Parallel with the development of new process-product development technologies, associated computer aided tools are needed for faster dissemination. However, the current unit-operation based process simulators, although having great success and wide applications, are not able to satisfy the demands for new technologies (multiscale, multipurpose, new concepts leading to new algorithms, visualization, etc.). Therefore, the demand for a new class of computer-aided software has increased, especially for new areas of chemical engineering in which there are no standard software tools. For example, computer tools are developed to target and accelerate the half life of product innovation (time to market) due to competitive global economy and increasing consumer demands.

This work focuses on a generic component-based architecture for development of tailor made computer-aided tools for specific application objectives. The architecture is based on “components”, which can be model-objects that are stored in a library (developed through tools like MoT [1]); collection of property models; databases; numerical solvers; design-algorithms (work-flows) etc. The main idea of these components is to incorporate a plug and play framework, where for a specific purpose, after the work-flow and associated data-flow are established, a specific computer-aided tool can be configured without too much additional programming work. That is, after implementing through an interface, the work-flow and data-flow associated with the new tool, retrieve and connect the necessary components for each step of the work-flow. This architecture has been used to develop various computer-aided tools like ProCAPD (computer aided product design [2]), ProCAFD (computer aided flowsheet design [3]), ProCACD (computer aided controller design), ProCAPE (computer aided property estimation), ProCAMT (computer aided modelling toolbox).

The developed product simulator (ProCAPD) is analogous to a process simulator, that is, it helps make the product design and development easier and faster and provides for a unified and consistent product documentation. However, unlike process simulators, ProCAPD can also be used directly for design of chemicals-based products, such as, single molecule products, formulations, blends, emulsions and devices. To design this chemical product simulator (ProCAPD), the components that must be selected are databases (chemicals, solvents, active ingredients, aroma, color-agents etc); model objects (properties, product performance, etc.); solvers (product attributes, blend compositions, environmental impact, etc.); design algorithms (single molecules, blends, formulations, emulsions, devices). Similarly based on this generic component architecture a computer-aided flowsheet design tool (ProCAFD), capable of enumerating the entire feasible search space for processing routes, analyzing and determining the most sustainable process, is developed. The components required for ProCAFD are databases (chemicals, solvents, process-groups as opposed to unit operations, reactions); model objects (pure and mixture properties, flowsheet property, phase equilibrium, driving force etc.); calculation tools (solvent attributes, mixture compositions, environmental impact, etc.); design algorithms (operational design, solvent design, separation synthesis etc.). Tools like ProCAPE and ProCAMT are more problem specific and require smaller number of components, for example, ProCAPE requires database of chemicals and properties, models for pure component and mixture properties, property calculation options, graphics for phase diagrams and property analysis, and, model parameter estimation toolbox as the main components. The application range of this tool depends on the property models available in the model library. In ProCAMT, the main components are a model generator, a model analyzer, a model solver (collection of numerical solvers) and a model reuser. This tool serves to create model objects for other tools.

This presentation will focus on the architecture, components and the generic main user interface through which different components are connected based on the identified data and work flows. Tailor-made tool development based on this component-based architecture will be illustrated to emphasize the power of true "plug and play".


[1] M. Fedorova, G. Tolksdorf, S. Fillinger, G. Wozny, M. Sales-Cruz, G. Sin, R. Gani, 2015. Development of Computer Aided Modelling Templates for Model Re-use in Chemical and Biochemical Process and Product Design: Import and export of models. Computer Aided Chemical Engineering, 37, 953-958

[2] A.K. Tula, J. Bottlaender, M.R. Eden, R. Gani, 2017, A computer-aided software-tool for sustainable process synthesis-intensification, Computers Chem Eng, 105, 74-95.

[3] S. Kalakul, M.R. Eden, R. Gani, 2017, The Chemical Product Simulator – ProCAPD, Computer Aided Chemical Engineering, 40, 979-984.