Artificial Neural Network for Fast and Versatile Model Parameter Adjustment utilizin PAT signals of Chromatography Processes for Process Control under Production Conditions

Preparative chromatography is a well-established operation in the chemical and biotechnology manufacturing. Chromatography achieves high separation performances but often has to deal with the yield versus purity trade-off as the optimization criterium regarding through-put. The initial trade-off is often disturbed by the well-known phenomenon of chromatogram shifts over process lifetime and has to be corrected by operators via adjustment of peak fraction cutting. Nevertheless, with regards to autonomous operation and batch to continuous processing modes, any advanced process control strategy is needed to identify and correct shifts from the optimal operation point automatically. Previous studies already presented solutions for batch-to-batch variance and process control options with the aid of rigorous physico-chemical process model-ling. These models can be implemented as distinct digital twins as well as statistical process op-eration data analysers. In order to utilize such models for advanced process control, the model parameters have to be up dated with aid of inline PAT data to describe the actual operational status. Also including any occurring operational change phenomenon and its relation to their physico-chemical root cause. Typical phenomena are fluid dynamic changes due to packing breakage, channelling or compression as well as mass transfer and phase equilibrium related separation performance decrease due to adsorbent ageing or feed and buffer composition changes. In order to track these changes an Artificial Neural Network (ANN) is trained in this work. The ANN training is in this first step based on the simulation results of a distinct and pre-viously experimentally validated process model. The model is implemented in the open source tool CasADi for python. This allows the implementation of interfaces to e.g. process control systems with relatively low effort. Therefore, PAT signals can easily incorporated for the suffi-cient adjustment of the process model for appropriate process control. Further steps would be the implementation of optimization routines based on the PAT and ANN predictions to derive optimal operation points with the model.