Computers & Chemical Engineering最新文献

{"title":"Towards an efficient integrated CSP production network: sustainability, storage and power-to-X selection, based on social metrics","authors":"José Antonio Luceño-Sánchez ,&nbsp;Mariano Martín ,&nbsp;Sandro Macchietto","doi":"10.1016/j.compchemeng.2025.109389","DOIUrl":"10.1016/j.compchemeng.2025.109389","url":null,"abstract":"<div><div>The sustainability and energy sovereignty of electricity production systems is one of the greatest concerns for countries today. Several countries are choosing technologies that are not dependent on fossil fuels. Concentrated Solar Power (CSP) plants are renewable-based facilities which exploit solar radiation to produce electricity, but the oversizing of these facilities to meet the demand during winter is not desirable. In this work, a novel formulation for the deployment of a CSP plants network with energy storage solutions is developed, considering region-related parameters (such as social metrics, sun hours, water availability, etc.) and four types of storage: compressed air (CAES), hydrogen, ammonia, and methanol. The problem was formulated as a multi-period mixed integer linear programing (MILP) problem and considering Spain as case study. The model selected CAES as the best energy storage solution, and the results show the use of CAES storage decreases the total energy production by 21%, but increases LCOE from 0.74-0.82 to 98.34 €/kWh. The regions are located in the south of Spain, because the solar radiation and the social metrics strongly influence the decision-making.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109389"},"PeriodicalIF":3.9,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019865","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Side-stream filtration and cooling for hybrid open-closed circulating cooling water system: Retrofitting and stochastic optimization","authors":"Ben Zhang ,&nbsp;Qiping Zhu ,&nbsp;Runrun Song ,&nbsp;Jingzheng Ren ,&nbsp;Chang He ,&nbsp;Haoshui Yu","doi":"10.1016/j.compchemeng.2025.109385","DOIUrl":"10.1016/j.compchemeng.2025.109385","url":null,"abstract":"<div><div>This study presents a retrofit strategy for traditional open circulating cooling water systems by integrating closed-wet cooling towers (CWCTs) into the side-stream pipeline, creating a hybrid system that combines open and closed loop configurations. To accurately represent the performance of the CWCTs within the retrofitted system, an optimal experimental design is conducted to approximate the multivariate probability distributions by generating a finite set of scenarios across various input variables. These distributions are then propagated through multi-scenario CFD simulations to develop a data-driven reduced-order model (ROM) for the CWCTs. This ROM, combined with simplified models for filtration and the existing open cooling towers, is applied within a two-stage stochastic optimization framework to minimize the investment recovery period under uncertainty. Results from a real-world example show a 5.7% reduction in operational load, 13.7% decrease in makeup water consumption, and 5.4% cost savings, demonstrating the strategy’s effectiveness in balancing water conservation, energy efficiency, and economic viability.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109385"},"PeriodicalIF":3.9,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144997293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic scheduling and control of a single column chromatography process for integrated continuous bioprocessing","authors":"Ian A. Gough ,&nbsp;Brandon Corbett ,&nbsp;Jake Raycraft ,&nbsp;Prashant Mhaskar ,&nbsp;Chris McCready ,&nbsp;David R. Latulippe ,&nbsp;Christopher L.E. Swartz","doi":"10.1016/j.compchemeng.2025.109386","DOIUrl":"10.1016/j.compchemeng.2025.109386","url":null,"abstract":"<div><div>Integrated continuous bioprocesses (ICBs) offer significant advantages for biotherapeutic manufacturing, including enhanced efficiency, reduced costs, and improved product accessibility. However, the adoption of ICBs is hindered by challenges in maintaining robust operation of the downstream processes amidst upstream variability. This study presents a mixed-integer nonlinear programming (MINLP) formulation for adaptive scheduling and control of a single-column bind-elute chromatography process that is integrated with a time-varying bioreactor harvest and a surge vessel. The control system leverages dynamic models, a rolling horizon implementation and a feedforward harvest forecast to optimize the chromatography loading flow rate and duration while ensuring compliance with critical process constraints. Case studies demonstrate the controller's ability to adapt the chromatography process and maintain robust operation under static and dynamic upstream harvest conditions. This framework represents a significant step toward the broader adoption of ICBs by utilizing the surge vessel as a degree of freedom to implement a variable loading flow rate strategy.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109386"},"PeriodicalIF":3.9,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145105876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Knowledge-based training of learning architectures under input sensitivity constraints for improved explainability","authors":"Hasan Sildir ,&nbsp;Emrullah Erturk ,&nbsp;Deniz Tuna Edizer ,&nbsp;Ozgun Deliismail ,&nbsp;Yusuf Muhammed Durna ,&nbsp;Bahtiyar Hamit","doi":"10.1016/j.compchemeng.2025.109382","DOIUrl":"10.1016/j.compchemeng.2025.109382","url":null,"abstract":"<div><div>The traditional machine learning (ML) training problem is unconstrained and lacks an explicit formulation of the underlying driving phenomena. Such a formulation, based solely on experimental data, does not ensure the delivery of qualitative knowledge among variables due to many theoretical issues in the optimization task. This study further tightens Artificial Neural Networks (ANNs) training by including input sensitivities as additional constraints and applies to regression and classification tasks based on literature data. In theory, such sensitivity represents the change direction of the target variable per change in measurements from indicators. The resulting nonlinear optimization problem is solved th rough a rigorous solver and includes the sensitivity expressions through algorithmic differentiation. Compared to traditional methods, with an acceptable decrease in the prediction capability, the proposed model delivers more intuitive, explainable, and experimentally verifiable predictions under input variable variations, under robustness to overfitting, while serving robust identification tasks. A classification case study includes a patient-oriented clinical decision support system development based on the impact of cancer-indicating variables. A competitive test prediction accuracy is obtained compared to commonly used algorithms despite 10 % decrease in the training. The regression case is built upon the energy load estimation to account for prominent considerations to obtain desired sensitivity patterns and proposed methodology delivers significant accuracy drop compared to some formulations to address knowledge patterns. The approach delivers a compatible pattern with practitioner expertise and is compared to widely used machine learning algorithms, whose performances are evaluated through common statistics in addition to multi-variable response graphs.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109382"},"PeriodicalIF":3.9,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019864","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated generation of mechanistic models for chemical process digital twins using reinforcement learning part II: Compartmentalization and learning-based recalibration","authors":"Jan-Frederic Laub ,&nbsp;Jiyizhe Zhang ,&nbsp;Mathis Heyer ,&nbsp;Alexei A. Lapkin","doi":"10.1016/j.compchemeng.2025.109384","DOIUrl":"10.1016/j.compchemeng.2025.109384","url":null,"abstract":"<div><div>Developing predictive models is central to building digital twins for chemical processes, which have a variety of applications in their development and operation. Mechanistic models are highly interpretable and have a larger domain of validity compared to data-driven models, but require significant time and expert knowledge to construct. In this contribution, a workflow for automated mechanistic model generation is extended to handle systems comprised of interdependent, spatially distributed phenomena. The search for accurate models is performed by hierarchically connected reinforcement learning agents. Different ways to incorporate human expertise in model generation are explored, and an ontology is introduced to manage expert and modeling knowledge. The extended workflow is shown to reliably find accurate models of chemical systems, exemplified on a phase transfer catalysis reaction and a Taylor-Couette reactor. For the latter, its non-ideal flow patterns were predicted within a deviation of 5 %, and automatically generated compartmentalization results were found to have comparable physical interpretations to bespoke models from literature. Additionally, the reinforcement learning agents were able to accurately recalibrate models up to twice as fast when drawing upon pre-training under a different operation condition. By generalizing all parts of the automated modeling procedures, we enable the efficient (re-)use of knowledge previously confined to the human modeler. We envision that in the future, the role of experts can be shifted from actively constructing each model iteration to curating knowledge and working collaboratively with autonomous agents.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109384"},"PeriodicalIF":3.9,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145044733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PC-Gym: Benchmark environments for process control problems","authors":"Maximilian Bloor ,&nbsp;José Torraca ,&nbsp;Ilya Orson Sandoval ,&nbsp;Akhil Ahmed ,&nbsp;Martha White ,&nbsp;Mehmet Mercangöz ,&nbsp;Calvin Tsay ,&nbsp;Ehecatl Antonio Del Rio Chanona ,&nbsp;Max Mowbray","doi":"10.1016/j.compchemeng.2025.109363","DOIUrl":"10.1016/j.compchemeng.2025.109363","url":null,"abstract":"<div><div><span>PC-Gym</span> is an open-source tool for developing and evaluating reinforcement learning (RL) algorithms in chemical process control. It features models that simulate various chemical processes, incorporating nonlinear dynamics, disturbances, and constraints. The tool includes customizable constraint handling, disturbance generation, reward function design, and enables comparison of RL algorithms against Nonlinear Model Predictive Control (NMPC) across different scenarios. Case studies demonstrate the framework’s effectiveness in evaluating RL approaches for systems like continuously stirred tank reactors, multistage extraction processes, and crystallization reactors. The results reveal performance gaps between RL algorithms and NMPC oracles, highlighting areas for improvement and enabling benchmarking. By providing a standardized platform, <span>PC-Gym</span> aims to accelerate research at the intersection of machine learning, control, and process systems engineering. Connecting RL with practical industrial process control applications, <span>PC-Gym</span> offers researchers a tool for exploring data-driven control solutions.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109363"},"PeriodicalIF":3.9,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cyclic reinforcement learning for generalization enhancement on T/C imbalance scheduling in TFT-LCD cell manufacturing","authors":"Tzu-Yen Hong ,&nbsp;Chia-Yen Lee ,&nbsp;Kuan-Chun Lu ,&nbsp;Chia-Fan Chu","doi":"10.1016/j.compchemeng.2025.109380","DOIUrl":"10.1016/j.compchemeng.2025.109380","url":null,"abstract":"<div><div>The rising product diversity for Thin-Film Transistor Liquid Crystal Display (TFT-LCD) has amplified the need for an efficient manufacturing process. This study formulates the TFT-LCD cell process scheduling as a dynamic flexible job shop scheduling problem, aiming to balance production between TFT array and color filter substrates (i.e. T/C balance) while accounting for new job arrivals and uncertain processing times. To optimize multiple objectives, including makespan, total weighted tardiness, violation of limited queue time, and T/C balance, a cyclic reinforcement learning (CRL) framework with a cyclic training process is proposed to achieve robustness under uncertain scenarios. A numerical study is conducted to validate the proposed framework, with performance compared against benchmark models, including optimization-based approaches and genetic algorithm. The results show that the CRL outperforms benchmark models in both realized objective value and variation while efficiently handling new job arrivals within a short inference time. Sensitivity analysis further confirms the robustness even in highly uncertain manufacturing environments.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109380"},"PeriodicalIF":3.9,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Environmental impacts prediction using graph neural networks on molecular graphs","authors":"Qinghe Gao ,&nbsp;Lukas Schulze Balhorn ,&nbsp;Alessandro Laera ,&nbsp;Raoul Meys ,&nbsp;Jonas Goßen ,&nbsp;Jana M. Weber ,&nbsp;Gregor Wernet ,&nbsp;Artur M. Schweidtmann","doi":"10.1016/j.compchemeng.2025.109362","DOIUrl":"10.1016/j.compchemeng.2025.109362","url":null,"abstract":"<div><div>The chemical industry needs to undergo a significant transformation towards more sustainable and circular production systems. To guide this transformation, estimating the environmental impacts of chemical production at early product screening or development stages is highly desirable. This study leverages the molecular structure of the process products with graph neural networks (GNNs) for early-stage environmental impact approximation of chemical processes. Specifically, we use end-to-end GNN models to predict fifteen environmental impact categories, utilizing a CarbonMinds dataset of 51,905 processes producing 791 molecules produced in 91 countries, augmented with country-specific energy mix data. Our analysis begins with a comparison of Quantitative Structure-Property Relationship (QSPR) and GNN models for the climate change impact category. Specifically, we develop three different GNN models: (i) GNN with only molecular structure, (ii) GNN with molecular structure and additional geographical features, and (iii) GNN with molecular structure and additional energy mix features. The results indicate that the three GNN models show an improvement over the QSPR models. Furthermore, benchmarking our GNN models against the existing literature in the climate change impact category reveals that our models perform comparably. We then extend our approach by developing both single- and multi-task GNN models to predict all fifteen impact categories. The findings indicate that multi-task learning can improve model performance in complex environmental impact predictions compared to single-task GNNs. Therefore, we recommend using a multi-task GNN for predicting multiple impact categories, with single-task models applied to fine-tune performance on underperforming categories. Although our proposed approach shows improvements over previous models, the prediction of environmental impacts solely based on molecular information remains a rough approximation.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109362"},"PeriodicalIF":3.9,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144933343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimization of direct air capture processes using reactive transport models of adsorption-desorption cycles","authors":"Hector A. Pedrozo ,&nbsp;Mayra G. Gonzalez-Ramirez ,&nbsp;Tiras Y. Lin ,&nbsp;Thomas Moore ,&nbsp;Thomas Roy ,&nbsp;Du T. Nguyen ,&nbsp;Pratanu Roy ,&nbsp;Sarah Baker ,&nbsp;Lorenz T. Biegler ,&nbsp;Grigorios Panagakos","doi":"10.1016/j.compchemeng.2025.109379","DOIUrl":"10.1016/j.compchemeng.2025.109379","url":null,"abstract":"<div><div>In this study, we develop and implement a reactive transport model in COMSOL Multiphysics® to address the challenges of direct air carbon capture. The model is validated against experimental data and used to simulate the cyclic steady state of the adsorption-desorption process. The optimization of this model is achieved through advanced trust-region methods integrated with Gaussian Processes. Key decision variables, including adsorption and desorption times, desorption temperature and pressure, input velocity, bed porosity, column length, and radius were optimized to minimize the capture cost. After optimization, a sensitivity analysis revealed the complex interplay between the decision variables and their effect on the specific energy and cost of removing the CO₂. We optimized the capture cost while taking into account the trade-off between energy consumption and productivity. The resulting minimum capture cost was determined to be 265.2 $/t-CO₂, which aligns with expected values reported in the literature. Numerical results suggest the effectiveness of the optimization strategies applied, and underscore the importance of simultaneous decision variable selection in improving the performance in direct air capture processes.</div><div>We also extend the modeling approach to a 2D axisymmetric model to better visualize CO₂ uptake and temperature profiles, revealing significant radial gradients during the regeneration step. As a main drawback, this enhanced model comes with a computational cost approximately 40 times higher than that of the 1D model.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109379"},"PeriodicalIF":3.9,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic mode decomposition accelerated forecast and optimization of geological CO2 storage in deep saline aquifers","authors":"Dimitrios Voulanas ,&nbsp;Eduardo Gildin","doi":"10.1016/j.compchemeng.2025.109377","DOIUrl":"10.1016/j.compchemeng.2025.109377","url":null,"abstract":"<div><div>DMDc and DMDspc models successfully expedite CO₂ fluid flow forecast and optimization, aiding in the acceleration of risk assessment, overall decision-making, and regulatory approvals for geological CO₂ storage by shortening regulatory-critical modeling cycles and being simple to train, while requiring fewer computational resources than traditional high-fidelity reservoir simulators, machine learning, and reduced-physics proxy models. DMDc and DMDspc models were trained independently with single set weekly, monthly, and yearly commercial simulator pressure and CO₂ saturation fields. DMDc/DMDspc reduced the snapshot reconstruction from several hours to minutes. The DMD models forecast performance was cross compared with independent and individual simulator snapshots sets generated with different well rates. Only DMD models that generalized well across different injection regimes and have errors below 5 % PCE for pressure and 0.01 MAE for saturation were considered acceptable. Regarding optimization, we propose the reconstruction of only monitored-during-optimization cells as it reduces even further optimization time while providing consistent results with the full snapshot reconstruction optimization. Optimized CO₂ injection and water production amounts were consistent across selected DMD models and all time scales. DMDspc delivers order-of-magnitude faster snapshot reconstruction with small accuracy loss and lower memory because only a compact set of dominant modes are reconstructed. While modern machine learning surrogates can match inference speed and accuracy with DMDspc, they are much harder to build, whereas DMDspc is non-intrusive, deterministic, and straightforward to deploy. To the best of our knowledge, this is the first application of DMD, particularly DMDspc, for forecast and optimization of geological CO₂ storage.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109377"},"PeriodicalIF":3.9,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}