A Generic Framework for Systematic Design of Process Monitoring and Control System for Crystallization Processes

Publication: ResearchConference abstract for conference – Annual report year: 2012

View graph of relations

Crystallization processes have a wide range of application as a solid-liquid separation technique in the chemical, the pharmaceutical and the food industries, due to the fact that high quality crystalline products can be produced. The main specifications of the crystal product are usually given in terms of crystal size, size distribution, shape and purity. However, the main difficulty in crystallization processes is to obtain a uniform and reproducible crystal size distribution (CSD). To this end, supersaturation control is often applied to drive the process within the metastable zone in order to enhance the control of the CSD. Although this approach has been shown to produce high quality crystals, the set point operating policies for the controller are usually chosen arbitrarily or by trial-and-error. Therefore a systematic procedure to generate operational policies that guarantee that a targeted CSD can be achieved, can be very useful. For such a procedure to be generic i.e. applicable to many case studies of different chemical systems, it needs to be model-based, preferably linked to a modelling framework with a model generation option to allow a wide application range. Furthermore, for control and monitoring purposes, an appropriate Process Analytical Technology (PAT) system ensuring that the critical process variables are measured and maintained within the design limits also needs to be integrated.
The objectives of this work are to develop a generic systematic design framework for monitoring and control systems applicable to a wide range of crystallization processes and operational scenarios. This framework contains a generic multi-dimensional modelling framework and features for design of operational scenarios and for design of PAT systems. The generality of this approach allows the users to generate a wide range of problem-system specific models through the generic multi-dimensional modelling framework [1]. In order to obtain the desired crystal products, an analytical CSD estimator and a response surface method are employed to generate the operational policy needed to match the desired target CSD. The generated operational policies provide the supersaturation set point and by maintaining the operation at this point, the targeted CSD is achieved. The resulting problem-system specific models and the operational policies become ready for use in model-based design and control/analysis of crystallization operations within a model-based process monitoring and control system (PAT system) [2]. The application of the systematic design framework will be highlighted through a potassium dihydrogen phosphate (KDP) crystallization process case study where the objective is to obtain a desired two-dimensional CSD and crystal shape. Also, integrated visualization tools are used together with the generated data for process control and for product (crystal) property monitoring, illustrating, thereby, the power and unique features of this systematic model-based design procedure.
Original languageEnglish
Publication date2012
StatePublished

Conference

Conference22nd European Symposium on Computer Aided Process Engineering
CountryUnited Kingdom
CityLondon
Period17/06/1222/06/12

Bibliographical note

Oral presentation.

Reference:
[1] Samad, N.AF.A., Singh, R., Sin, G., Gernaey, K.V. & Gani, R. (2011). Computers and Chemical Engineering, 35,
828-843.
[2] Singh, R., Gernaey, K.V. & Gani, R. (2009). Computers and Chemical Engineering, 33, 22-42.

Keywords

  • Process monitoring and control, Analytical estimator, Crystal distribution, Crystal shape
Download as:
Download as PDF
Select render style:
APAAuthorCBEHarvardMLAStandardVancouverShortLong
PDF
Download as HTML
Select render style:
APAAuthorCBEHarvardMLAStandardVancouverShortLong
HTML
Download as Word
Select render style:
APAAuthorCBEHarvardMLAStandardVancouverShortLong
Word

ID: 9987688