Optimal Enantiomer Crystallization Operation using Ternary Diagram Information

Marcellos, Caio Felippe Curitiba; Durand, Helen; Kwon, Joseph; Barreto, Amaro Gomes; Lage, Paulo Laranjeira da Cunha; Souza, Maurício B. de; Secchi, Argimiro Resende; Christofides, Panagiotis D.

doi:10.1016/b978-0-444-64241-7.50078-1

Cited by 3 publications

(2 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such models do not have an exact analytical solution and require the use of numerical solution methods such as finite element methods, volume methods or the method of characteristics, which is susceptible to high computational complexity. Although improved optimization algorithms can be used to reduce the computational time, the model needs to be rebuilt corresponding to different processes. , At the same time, the equations involved in the model are still biased toward the ideal and deviate significantly from the real process, making it less suitable for real-time optimal control. Data-driven models are capable of modeling multidimensional nonlinear systems with only process data, without prior knowledge of the physical meaning of the process, avoiding the necessity of developing comprehensive first-principles models, and have been widely used in research on process modeling, optimization, and control.…”

Section: Crystallization Process Control and Optimizationmentioning

confidence: 99%

Recent Advances in the Application of Machine Learning to Crystal Behavior and Crystallization Process Control

Lu,

Rao,

Yue

et al. 2024

Crystal Growth & Design

View full text Add to dashboard Cite

Crystals are integral to a variety of industrial applications, such as the development of pharmaceuticals and advancements in material science. To anticipate crystal behavior and pinpoint effective crystallization techniques, a thorough investigation of crystal structures, properties, and the associated processes is essential. However, conventional methods like experimental procedures and quantum mechanics calculations, while crucial, can be expensive and time-consuming. In response, machine learning has risen as an effective alternative, complementing the traditional approaches based on quantum mechanics and classical force fields. In the recent years, the deployment of machine learning in the realm of crystallization has yielded notable progress. This review offers a concise overview of the application of machine learning techniques in crystallization, focusing on the past five years. Our analysis of the literature indicates that machine learning has accelerated the prediction of crystal structures by streamlining the generation and evaluation of structures. Additionally, it has facilitated the prediction of key crystal properties such as solubility, melting point, and habit. The review further explores the role of machine learning in refining the control and optimization of crystallization processes, highlighting the restrictions of conventional algorithms and sensing technologies. The advantages of endto-end processing for enhancing the accuracy of predictions and the combination of data-driven with mechanism-based models for robustness are also considered. In summary, this review provides insights into the current state of machine learning in the field of intelligent crystallization and suggests pathways for future research and development.

show abstract

Section: Crystallization Process Control and Optimizationmentioning

confidence: 99%

Recent Advances in the Application of Machine Learning to Crystal Behavior and Crystallization Process Control

Lu,

Rao,

Yue

et al. 2024

Crystal Growth & Design

View full text Add to dashboard Cite

show abstract

“…QbC emphasizes the use of mathematical and knowledge-based modeling for quantitative and predictive product and process understanding and is widely recognized as a valuable tool for improved mechanistic understanding, process optimization, and robust control. − However, modeling of pharmaceutical processes, particularly crystallization, can be of great challenge due to their complex, stochastic, and kinetic-driven nature. , Moreover, as contrary to its ubiquity, crystallization is often not well-understood by practitioners, and this renders optimal control a vision rather than a reality. While previous studies have sought to elucidate the optimization and control of crystallization processes, − a majority have adopted the first-principles modeling approach [e.g., population balance model (PBM)], which is susceptible to high computation complexity and is hence less amenable for real-time optimal control. Although improved optimization algorithms could be adopted to enhance the calculation time, determination and periodic adjustments of the various crystallization kinetics based on new process data obtained can be cumbersome.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Modeling and Predictive Control of the Batch Crystallization Process

Zheng

Wang

2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

This work develops a framework for building machine learning models and machine-learning-based predictive control schemes for batch crystallization processes. We consider a seeded fesoterodine fumarate cooling crystallization and dissolution process in a batch reactor and present the methodology and implementation of simulation, modeling, and controller design. Specifically, to address the experimental data scarcity problem, we first develop a one-dimensional population balance model based on published kinetic parameters that were obtained empirically to describe the formation of crystals via nucleation, growth, and agglomeration. Then, recurrent neural network (RNN) and autoencoder−RNN (AERNN) models are developed using data from extensive open-loop simulations of the semi-empirical population balance model under various operating conditions to capture the process dynamic behavior. Two model predictive control (MPC) schemes using the respective RNN and AERNN models are developed to optimize the crystallization process with respect to product yield, crystal size, number of fines in the final product, and energy consumption, while accounting for the constraints on manipulated inputs. Through open-and closed-loop simulations, it is demonstrated that the RNN and AERNN models capture the process dynamics well, and the RNN-and AERNN-based MPCs achieved the desired product yield and crystal size with significantly improved computational efficiency.

show abstract

Chirality in Organic and Mineral Systems: A Review of Reactivity and Alteration Processes Relevant to Prebiotic Chemistry and Life Detection Missions

et al. 2022

View full text Add to dashboard Cite

Chirality is a central feature in the evolution of biological systems, but the reason for biology’s strong preference for specific chiralities of amino acids, sugars, and other molecules remains a controversial and unanswered question in origins of life research. Biological polymers tend toward homochiral systems, which favor the incorporation of a single enantiomer (molecules with a specific chiral configuration) over the other. There have been numerous investigations into the processes that preferentially enrich one enantiomer to understand the evolution of an early, racemic, prebiotic organic world. Chirality can also be a property of minerals; their interaction with chiral organics is important for assessing how post-depositional alteration processes could affect the stereochemical configuration of simple and complex organic molecules. In this paper, we review the properties of organic compounds and minerals as well as the physical, chemical, and geological processes that affect organic and mineral chirality during the preservation and detection of organic compounds. We provide perspectives and discussions on the reactions and analytical techniques that can be performed in the laboratory, and comment on the state of knowledge of flight-capable technologies in current and future planetary missions, with a focus on organics analysis and life detection.

show abstract

Optimal Enantiomer Crystallization Operation using Ternary Diagram Information

Cited by 3 publications

References 1 publication

Recent Advances in the Application of Machine Learning to Crystal Behavior and Crystallization Process Control

Recent Advances in the Application of Machine Learning to Crystal Behavior and Crystallization Process Control

Machine Learning Modeling and Predictive Control of the Batch Crystallization Process

Chirality in Organic and Mineral Systems: A Review of Reactivity and Alteration Processes Relevant to Prebiotic Chemistry and Life Detection Missions

Contact Info

Product

Resources

About