This paper presents a systematic approach to the advanced control of continuous crystallization processes using model predictive control.We provide a tutorial introduction to controlling complex particle size distributions by integrating population balance equations with detailed models of various continuous crystallizers.Since these high-fidelity models are often too complex for online optimization, we propose the use of data-driven surrogate models that enable efficient optimization-based control.Through two case studies, one with a low-complexity system allowing direct comparison with traditional methods and another involving a spatially distributed crystallizer, we demonstrate how our approach enables real-time model predictive control while maintaining accuracy.The presented methodology facilitates the use of complex models in a model-based control framework, allowing precise control of key particle size distribution characteristics, such as the median particle size 50 and the width 90 − 10 .This addresses a critical challenge in pharmaceutical and fine chemical manufacturing, where product quality depends on tight control of particle characteristics.
This paper provides a tutorial overview of model predictive control (MPC) methods for continuous crystallization processes, emphasizing the integration of population balance equations with surrogate models to facilitate real-time optimization. The authors demonstrate the application of MPC through two case studies—one representing a straightforward system and the other a more complex spatially distributed crystallizer. Key particle size distribution characteristics, specifically median particle size (d̅43) and width (d90-d10), are targeted for precise control, which is crucial in pharmaceutical manufacturing. The methodology leverages advancements in machine learning, particularly surrogate modeling, to address the computational challenges typically associated with detailed first-principle models. The overall aim is to bridge the gap between advanced control strategies and practical implementation in crystallization processes, advocating for stronger collaboration between control systems and crystallization communities.
This paper employs the following methods:
- Model Predictive Control (MPC)
- Population Balance Equations
- Surrogate Models
The following datasets were used in this research:
- Enables real-time model predictive control while maintaining accuracy
- Allows precise control of key particle size distribution characteristics
- Demonstrates successful application of complex models for continuous crystallizers in MPC
The authors identified the following limitations:
- Lack of comprehensive dynamic modeling techniques for both particle size distribution and spatially distributed crystallizers
- Current implementations primarily focus on simplified models lacking spatial variations
- Number of GPUs: None specified
- GPU Type: None specified
- Compute Requirements: None specified