Initial situation: The chemical industry consists of complex operations and processes, which are always subject to a certain parameter variation and dynamics. There are therefore many variables and degrees of freedom (for example pressure, temperature, composition, mass flows, etc.). An optimization is very complex, especially with fluctuating raw materials and process parameters. Currently, mathematical models are used in the chemical industry for optimization. Although changes in the processes are predicted with mathematical models, they can be extrapolated through the nature of mathematical models across physical boundaries. Achieving the targets is therefore sub-optimal and occurs at the limit states of the processes with considerable uncertainties. Therefore, mathematical models find their limits. An improvement would be a dynamic optimization based on thermodynamic models. However, these systems are not available yet in the chemical industry and therefore currently they are not state of the art. This would require a real-time recording and use of the system parameters of process states in the operational area and to enter these physical system parameters directly into the model. This possibility of thermodynamic modelling would currently be unprecedented and has great potential for a wide variety of production facilities in the chemical industry (especially in fluctuating operating conditions). The dynamic optimization based on thermodynamic models is therefore the starting point of the project, especially since an efficiency increase in the use of materials by only 1% in the chemical industry would mean an annual cost reduction of approx. EUR 19 million.
Problem definition: A new approach is needed, which the real-time optimization of the process including physical process limits can be determined exactly (before). This is particular complex since pressure, temperature and the composition of the feed streams must be recorded in real time, so that at the same time the exact system parameters for optimization can be recorded. Such an optimization is currently not done for material systems with hundreds of individual chemical components, since thousands of parameters would have to be calculated in real time via mathematical models and the theoretically determined optimization result due to progressive calculation errors does not correspond to reality. Such real-time optimization is therefore possible only through physical (thermodynamic) models.
Aims: (1) A solution for the chemical industry should be developed using the example of a real process, which control parameters for production can be determined (previously) and simultaneously optimized during operation. (2) A system is to be developed by which essential input parameters such as boiling curve, density and composition of the multicomponent material system can be recorded in real time. (3) An application-oriented output solution is to be developed, which simultaneously outputs the exact system parameters for the current operating optimum in order to be able to set the controlled variables of the system to it. (4) The product to be developed should be easily integrated into the existing production of any chemical Industry
Methodology (excl. Project management): AP2 - Survey and analysis of technical solutions; AP3 - Development of a stationary and dynamic real-time simulation model (to capture the multi-component material system and to output the system parameters and optimize the process); AP4 - simulation and validation (comparison with real operation) using the example of a benzene production; AP5 - Development of a strategy for integrating the solution into production; AP6 - Final evaluation and derivation of recommendations for action / conclusions;
Desired Outcomes: (1) Solution for real-time optimization of processes using thermodynamic process models in the chemical industry. (2) Exact thermodynamic model (static and dynamic) and application results. (3) Sensitization of plant personnel (plant personnel are prepared for this demanding form of production). (4) Verified simulation results based on an application (optimization of benzene production). (5) Standardization of the system for applicability and multiplicability.
