(134c) Evaluating Waste Minimization Alternatives in a Chemical Plant

Growing environmental concerns have spurred technological changes to accommodate the concept of waste minimization within chemical process plants. Commercial process simulators such as Aspen Plus, HYSYS, Pro II and ChemCAD, which have been extensively employed to solve plant synthesis problems from the grass-root design to process retrofitting, are also gaining currencies for achieving waste minimization objectives. This is because through process simulators, different design alternatives and process variable changes can be evaluated quickly and compared effectively to the base-case process for improving the environmental performance of the plant. However, the use of these simulators is not a straightforward task as it requires considerable know-how, skills and expertise of the user for recognizing the process variables, which hold keys to the overall waste feature of the process plant. And for such task, very little help is actually provided by these simulators.

Although a number of researchers have successfully applied process simulators as their tools (Hopper et al, 1994; Cabezas et al, 1999; Fu et al, 2000), none has tried on automating the analysis procedures. Previously, we have reported on our attempt to automate waste minimization analysis through integration between a G2 (Gensym) based expert system and HYSYS process simulator (Halim and Srinivasan, 2002). The system, called ENVOPExpert, uses expert knowledge to automatically identify the source of wastes in a chemical process, propose qualitative process changes to minimize them and simultaneously calculate the environmental impact and plant profitability. However to achieve all the above mentioned tasks, the user is required to switch over from ENVOPExpert to HYSYS and vice versa for the qualitative and quantitative waste analysis. In this paper, we present an improved integration framework of ENVOPExpert and HYSYS simulator by capitalizing on the CAPE-OPEN capability for XML data transfer between HYSYS and ENVOPExpert.

In our updated framework, ENVOPExpert will operate as a waste minimization decision support tool by accessing the already available physical property, process chemistry and flowsheet information from the HYSYS simulator on an interactive basis. Once the information is available, ENVOPExpert will perform a two-step waste minimization analysis: qualitative waste assessment and quantitative waste balance simulation. A qualitative analysis is first conducted to identify the sources of wastes and to propose alternatives for eliminating or minimizing them. This will be performed through three fundamental elements: process graph (P-graph) for analyzing the material flow within the process, cause-and-effect knowledge for generating heuristic-based solutions, and functional model for connecting different streams and unit operations on the basis of their functionalities. Next, quantitative simulation will be performed through interfacing between ENVOPExpert and HYSYS simulator. In this way, the effects of variable changes (pressure, temperature, heat input and feed composition) to the amount of waste generated in the plant can be investigated. At the same time, interpretation of the simulation results and the direction of the process variable changes (increase or decrease) can also be automatically searched with multiple objectives of reducing the environmental impact and maximizing the profitability of the plant. Any synergies and trade-offs between different waste minimization alternatives will also be highlighted through automatic plotting of the environmental impact against the plant economic.

We will illustrate our automated approach using a propylene glycol production process and compare our results with the waste minimization suggestions identified by the experts.

References: (1)Cabezas, H.; Bare, J.C.; Mallick, S.K. Computers and Chemical Engineering 1999, 23, 623-634. (2)Fu, Y.; Diwekar, U.M.; Young, D.; Cabezas, H. Clean Products and Processes 2000, 2, 92-107. (3)Halim, I.; Srinivasan, R. An Integrated Decision Support System for Waste Minimization Analysis in Chemical Processes. Environmental Science & Technology 2002, 36(7), 1640-1648. (4)Hopper, J.R.; Yaws, C.L.; Vichalak, M.; Ho, T.C. Waste Management 1994, 14(3), 187-202.