Computers & Chemical Engineering最新文献

{"title":"A hierarchical scheme for dynamic monitoring of multi-scale multi-mode systems","authors":"Jiaorao Wang ,&nbsp;Lishuai Li ,&nbsp;S. Joe Qin","doi":"10.1016/j.compchemeng.2025.109107","DOIUrl":"10.1016/j.compchemeng.2025.109107","url":null,"abstract":"<div><div>Complex chemical plant operations exhibit multi-scale dynamics and multi-mode characteristics. Existing methods typically address either multi-scale, dynamic, or multi-mode monitoring separately. This paper proposes a general hierarchical scheme for dynamic monitoring of systems with multi-mode dynamic behaviors. The core strength of the proposed method lies in its iterative procedure, which comprises two steps: dynamic pattern modeling and mode segmentation. Firstly, dynamic patterns across different modes are captured using latent vector autoregressive (LaVAR) modeling. In mode segmentation, data representing new dynamic patterns are filtered for the construction of the next LaVAR model, guided by two monitoring indices. The hierarchical structure sequentially extracts dynamic patterns, inherently dealing with unbalanced data common in industrial applications. Experiments are conducted to demonstrate the effectiveness of the proposed scheme for multi-mode dynamic system monitoring.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109107"},"PeriodicalIF":3.9,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768561","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":"A homotopy for the reliable estimation of model parameters in chromatography processes","authors":"Dominik H. Cebulla ,&nbsp;Christian Kirches ,&nbsp;Nico Kümmerer ,&nbsp;Andreas Potschka","doi":"10.1016/j.compchemeng.2025.109100","DOIUrl":"10.1016/j.compchemeng.2025.109100","url":null,"abstract":"<div><div>Mathematical modeling, simulation, and optimization can significantly support the development and characterization of chromatography steps in the biopharmaceutical industry. Particularly mechanistic models become used, as these models, once carefully calibrated, can be employed for a reliable optimization. However, model calibration is a difficult task in this context due to high correlations between parameters, highly nonlinear models, and limited prior knowledge of certain parameters, among others.</div><div>In this work we propose a homotopy-based globalization strategy that can be used in combination with iterative algorithms for the solution of parameter estimation problems. Our method is tailored to measurement data showing a single peak or multiple, possibly overlapping peaks, which often arises from bind-and-elute experiments. With our approach, convergence to a local optimum can be achieved even when initial guesses are far away from a solution. Moreover, we describe and discuss the calibration procedure for a real-world ion exchange chromatography process, here considering a complete chromatography system. This description may serve as a general blueprint for the estimation of model parameters in chromatography processes.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"199 ","pages":"Article 109100"},"PeriodicalIF":3.9,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards zero carbon emissions: Electrification and decarbonization of an ethylene Plant's utility system","authors":"Aleksa Miladinović ,&nbsp;Aleksandar S. Grujić ,&nbsp;Mirjana Kijevčanin ,&nbsp;Vladimir Stijepović ,&nbsp;Sabla Y. Alnouri ,&nbsp;Mirko Stijepović","doi":"10.1016/j.compchemeng.2025.109117","DOIUrl":"10.1016/j.compchemeng.2025.109117","url":null,"abstract":"<div><div>The industrial sector faces significant challenges in decarbonization, within the petrochemical industry being a major contributor to global greenhouse gas (GHG) emissions. Ethylene, a key feedstock for polymers such as polyethylene (PE) and polyethylene terephthalate (PET), is primarily produced through steam cracking, a process heavily reliant on fossil fuels and responsible for substantial CO₂ emissions. This study introduces a systematic approach to retrofit steam utility system for an ethylene production plant, featuring the integration of hydrogen-fired boilers, electric superheaters, electric boilers and supplementary turbines linked to electrical generators. Powered entirely by renewable energy sources, this innovative electrification strategy is designed to enhance operational efficiency while optimizing both capital and operating costs. By adopting renewable energy options within the utility system, the proposed design significantly reduces the plant's carbon footprint, contributing to a more sustainable and environmentally responsible ethylene production process.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109117"},"PeriodicalIF":3.9,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748531","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":"Analytical identification of process design spaces using R-functions","authors":"S. Kucherenko ,&nbsp;N. Shah ,&nbsp;O.V. Klymenko","doi":"10.1016/j.compchemeng.2025.109112","DOIUrl":"10.1016/j.compchemeng.2025.109112","url":null,"abstract":"<div><div>A process design space (DS) is defined as the combination of process design and operational conditions that guarantees the assurance of product quality. This principle ensures that, as long as a process operates within its DS, it consistently yields a product that meets specifications. A novel DS identification method called the R-DS identifier has been developed in this work. It makes no assumptions about the underlying model - the only requirement is that each model constraint (e.g., defining product Critical Quality Attributes or process Key Performance Indicators) should be approximated by a closed-form function, e.g., a multivariate polynomial model. The method utilizes the methodology of V.L. Rvachev's R-functions and allows for explicit analytical representation of the DS with only a limited number of model runs. R-functions provide a framework for representing complex geometric shapes and performing operations on them through implicit functions and inequalities defining the regions. The theory of R-functions enables the solution of geometric problem such as identification of DS through algebraic manipulation. It is more practical than traditional sampling or optimization-based methods. The method is illustrated using a batch reactor model.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109112"},"PeriodicalIF":3.9,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interpretation of high-dimensional regression coefficients by comparison with linearized compressing features","authors":"Joachim Schaeffer ,&nbsp;Jinwook Rhyu ,&nbsp;Robin Droop ,&nbsp;Rolf Findeisen ,&nbsp;Richard D. Braatz","doi":"10.1016/j.compchemeng.2025.109099","DOIUrl":"10.1016/j.compchemeng.2025.109099","url":null,"abstract":"<div><div>Linear regression is often deemed inherently interpretable; however, challenges arise for high-dimensional data. We focus on further understanding how linear regression approximates nonlinear responses from high-dimensional functional data. We develop a linearization method to derive feature coefficients, which we compare with the closest regression coefficients of the path of regression solutions. We showcase the methods on battery data case studies where a single nonlinear compressing feature, <span><math><mrow><mi>g</mi><mo>:</mo><msup><mrow><mi>R</mi></mrow><mrow><mi>p</mi></mrow></msup><mo>→</mo><mi>R</mi></mrow></math></span>, is used to construct a synthetic response, <span><math><mrow><mi>y</mi><mo>∈</mo><mi>R</mi></mrow></math></span>. This unifying view of linear regression and compressing features for high-dimensional functional data helps to understand (1) how regression coefficients are shaped in the highly regularized domain, (2) how regression coefficients relate to linearized feature coefficients, and (3) how the shape of regression coefficients changes as a function of regularization to approximate nonlinear responses by exploiting local structures.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109099"},"PeriodicalIF":3.9,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705448","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":"A mixed-integer linear programming model proposal to determine material sustainability for new product development processes in production","authors":"Çağrı Kuram,&nbsp;Özge Nalan Bilişik","doi":"10.1016/j.compchemeng.2025.109108","DOIUrl":"10.1016/j.compchemeng.2025.109108","url":null,"abstract":"<div><div>Sustainable product design entails a comprehensive examination of the environmental, economic, and social impacts associated with the materials utilized in product development. The overarching goal is to foster a design approach that minimizes adverse effects on both the environment and human well-being. While existing literature predominantly concentrates on the selection of optimal materials, this study shifts the focus towards optimizing the entire process of transforming raw materials into finished products, with the dual objectives of cost efficiency and reduced use of hazardous materials. In this study, a mixed integer linear programming (MILP) model is proposed to analyze raw material sustainability and new product development processes. Unlike traditional approaches that often prioritize either environmental or economic sustainability, the proposed model endeavors to strike a balance between these two critical dimensions, thereby promoting holistic sustainability. Furthermore, the model addresses the intricate challenge of assigning materials to specific processing cells, ensuring that the sustainability objectives are seamlessly integrated into the operational workflow. By adopting this integrated approach, the study not only contributes to advancing the theoretical understanding of sustainable product design but also offers practical insights for industry practitioners seeking to enhance operational excellence while mitigating environmental impact. While the mathematical model aims to minimize both environmental and economic sustainability, the problem of assignment to cells is also addressed. As a result of the study, this research seeks to pave the way for more informed decision-making in the realm of sustainable manufacturing, ultimately fostering a transition towards a more environmentally conscious and socially responsible industrial ecosystem. Sensitivity analysis, which examines alterations in objective functions resulting from an escalation in sales price, is encompassed as well. Based on the findings, the variation in sales price triggers sensitivity in both the amount manufactured and the objective function. Furthermore, a modification in cell selection was observed, even though it is considered unimportant.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109108"},"PeriodicalIF":3.9,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716198","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":"An integrated approach to the optimal design of sustainably efficient biorefinery supply chains","authors":"Lucas M. Machin Ferrero ,&nbsp;Richard Cabrera Jiménez ,&nbsp;Jonathan Wheeler ,&nbsp;Carlos Pozo ,&nbsp;Fernando D. Mele","doi":"10.1016/j.compchemeng.2025.109104","DOIUrl":"10.1016/j.compchemeng.2025.109104","url":null,"abstract":"<div><div>The agribusiness sector needs to strategically align its activities with the 2030 Sustainable Development Goals. Biorefineries offer a solution for regions rich in natural resources, allowing them to utilize available biomass and progressively reduce dependence on fossil fuels. To ensure the sustainable design and management of biorefinery supply chains, key decisions need to be carefully assessed, taking into account economic, environmental and social impacts. This study proposes a purpose-driven strategy for the multi-criteria design of sustainable biorefinery supply chains, combining a multi-scenario MILP optimization approach with an efficiency (and super-efficiency) ranking using Data Envelopment Analysis (DEA) to address economic, environmental, and social trade-offs. The capabilities of the approach are demonstrated through a real case study in Argentina, using sugarcane and lemon derivatives as feedstocks for bioproducts such as ethanol, citric acid, methanol and lactic acid. The results show the production of bioethanol from sugarcane juice, together with the production of other bioproducts from sugarcane bagasse, as the optimal and sustainably efficient scenario.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109104"},"PeriodicalIF":3.9,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143685171","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":"Advanced data-driven fault detection in gas-to-liquid plants","authors":"Nour Basha ,&nbsp;Radhia Fezai ,&nbsp;Byanne Malluhi ,&nbsp;Khaled Dhibi ,&nbsp;Gasim Ibrahim ,&nbsp;Hanif A. Choudhury ,&nbsp;Mohamed S. Challiwala ,&nbsp;Hazem Nounou ,&nbsp;Nimir Elbashir ,&nbsp;Mohamed Nounou","doi":"10.1016/j.compchemeng.2025.109098","DOIUrl":"10.1016/j.compchemeng.2025.109098","url":null,"abstract":"<div><div>Fault detection is a critical part of process monitoring, where the objective is to flag unexpected operating behavior quickly and accurately. In this paper, a novel extension of the Generalized Likelihood Ratio charts is proposed, denoted as the Maximum Multivariate GLR charts. Linear and nonlinear data-driven models, namely principal component analysis and its kernel extension and neural networks, are combined with different statistical charts towards the detection of multiple fault types in three distinct case studies: synthetic, Tennessee Eastman process, and Gas-to-Liquid process. The results show that the MMGLR charts have a better detection accuracy than conventional charts, and that neural networks are more robust modeling techniques than PCA and KPCA for the sake of fault detection.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109098"},"PeriodicalIF":3.9,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143643747","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":"Anomaly detection for drug product manufacturing considering data limitations and shifts: A case study on industrial freeze-dryers","authors":"Gianluca Lombardini ,&nbsp;Sara Badr ,&nbsp;Christian Schmid ,&nbsp;Stephanie Knüppel ,&nbsp;Hirokazu Sugiyama","doi":"10.1016/j.compchemeng.2025.109106","DOIUrl":"10.1016/j.compchemeng.2025.109106","url":null,"abstract":"<div><div>Ensuring the quality of biopharmaceutical products requires robust manufacturing processes and reliable monitoring systems. In industrial applications with real data, traditional data-driven anomaly detection methods often face challenges due to data scarcity and data shifts. To address these challenges, we propose the application of Statistic Alignment (SA) as a domain adaptation technique within the broader framework of transfer learning. A methodology is presented incorporating SA as an effective precursor for One-Class Support Vector Machine (OCSVM) based anomaly detection. Using industrial data from two parallel freeze-dryers, we investigate two cases: (1) transferring a model within the same machine to handle data shifts caused by maintenance, and (2) transferring a model between machines to assess cross-system transferability. We evaluate three SA methods—Mean Alignment, Standard Alignment, and Correlation Alignment—while also exploring data requirements for effective alignment. Moreover, we propose a heuristic-based hyperparameter tuning method for OCSVM to further enhance model performance. Our results demonstrate that SA allows model transfer between domains with F1 scores around 0.9, offering a promising solution for enhancing model robustness in dynamic biopharmaceutical production environments.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109106"},"PeriodicalIF":3.9,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143738209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comparison of strategies to embed physics-informed neural networks in nonlinear model predictive control formulations solved via direct transcription","authors":"Carlos Andrés Elorza Casas,&nbsp;Luis A. Ricardez-Sandoval,&nbsp;Joshua L. Pulsipher","doi":"10.1016/j.compchemeng.2025.109105","DOIUrl":"10.1016/j.compchemeng.2025.109105","url":null,"abstract":"<div><div>This study aims to benchmark candidate strategies for embedding neural network (NN) surrogates in nonlinear model predictive control (NMPC) formulations that are subject to systems described with partial differential equations and that are solved via direct transcription (i.e., simultaneous methods). This study focuses on the use of physics-informed NNs and physics-informed convolutional NNs as the internal (surrogate) models within the NMPC formulation. One strategy embeds NN models as explicit algebraic constraints, leveraging the automatic differentiation (AD) of an algebraic modelling language (AML) to evaluate the derivatives. Alternatively, the solver can be provided with derivatives computed external to the AML via the AD routines of the machine learning environment the NN is trained in. The three numerical experiments considered in this work reveal that replacing mechanistic models with NN surrogates may not always offer computational advantages when smooth activation functions are used in conjunction with a local nonlinear solver (e.g., Ipopt), even with highly nonlinear systems. Moreover, in this context, the external function evaluation of the NN surrogates often outperforms the embedding strategies that rely on explicit algebraic constraints, likely due to the difficulty in initializing the auxiliary variables and constraints introduced by explicit algebraic reformulations.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"198 ","pages":"Article 109105"},"PeriodicalIF":3.9,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143738208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}