551a Systematic Retrofit Design of Batch Processes Using an Indicator and Model Based Framework

Levente L. Simon, Ulrich Fischer, and Konrad Hungerbuehler. Institute for Chemical and Bioengineering, Safety and Environmental Technology Group, ETH Zurich, G-138, HCI, ETH Hoenggerberg, Zurich, 8093, Switzerland

The incentives for a retrofit design of a chemical plant are multiple e.g. capacity expansion, waste reduction, operation efficiency increase due to patent expiration. The most common batch process improvement strategy is based on the identification and removal of production bottlenecks. The back draw of such an approach is that only the bottleneck units are analyzed, however low-cost improvements could be attained using a systematic analysis and improvement strategy. The goal of the new method is to systematically identify the process improvement possibilities in a batch plant with the objectives mentioned above. This methodology focuses on different levels of a multi-product or mono-product batch production plant. For a generic batch plant analysis the highest level (called level 1) can comprise a connection of multi-product lines sharing the same equipment or utility resources or it can consist of similar mono-product lines. Overall productivity limitations can already be identified on this level. A more detailed analysis can be carried out when the analysis is shifted to a lower level (called level 2). On this level the multi-product plant is divided into single product lines and analysis can be carried out for each product. In the case of a mono-product batch plant with multiple similar production lines the analysis can be restricted to a single line with the mention that the shared resources should be allocated accordingly, e.g. based on the shared size or time. The next step of the analysis is to focus on the unit operations (called level 3) which are part of level 2. This is the deepest level of analysis on which most of the improvements are implemented. The new method consists of the following steps: A. Plant analysis. The analysis step of the method is based on the component path flow and path flow indicators framework developed previously for continuous processes. This framework has been extended with indicators that are specific for the batch processes. These indicators help to assess time, volume, and utility bottlenecks and to evaluate and quantify different retrofit alternatives. Additionally, scale-up indicators, process operation variability related indicators are introduced in the analysis. B. Systematic generation of alternatives. In the next step of the algorithm retrofit alternatives are generated based on a database of heuristics. This heuristics database is compiled using information from scientific literature, white papers and in-house knowledge of industry experts. The heuristics database contains improvement suggestions for each analysis levels. C. Parametric process optimization. Important process parameters identified with the help of heuristics in step B are investigated using dynamic process model based sensitivity analysis and optimization. D. Evaluation of structural process changes. Changes suggested in step B are investigated based on developed dynamic models with regard to their improvement potential, their technical feasibility, and the investment cost. The method was applied to an industrial batch plant. For this indicators were calculated and the bottlenecks were identified. Based on the heuristics list improvement suggestions were made and these were evaluated using dynamic batch process models. Using this systematic retrofit method the major achievements consist in improved reactor operation with regard to swelling phenomena, faster conversion using a falling-film evaporator, optimized reaction-distillation sequence, and possibility of scale-up predictions. The developed method was able to identify the major process improvement potentials and proves that a systematic indicator and model based framework can be successfully implemented in the batch process industry.