Computers & Chemical Engineering最新文献

{"title":"Optimal control of scheduling and production for a multienzymatic system in continuous reactors","authors":"Pablo Silva ,&nbsp;Felipe Scott ,&nbsp;Sai Darshan Adloor ,&nbsp;Vassilios S. Vassiliadis ,&nbsp;Andrés Illanes ,&nbsp;Lorena Wilson ,&nbsp;Raúl Conejeros","doi":"10.1016/j.compchemeng.2025.109209","DOIUrl":"10.1016/j.compchemeng.2025.109209","url":null,"abstract":"<div><div>Enzyme inactivation significantly impacts reactor performance by reducing substrate conversion and product quality. This study, with its focus on optimizing the economic benefits of a novel two-step biocatalytic system, aims to control biocatalyst replacement intervals and operational conditions, thereby enhancing the economic viability of biocatalytic processes. The results demonstrate that optimal control strategies can be effectively implemented for Continuous Stirred Tank Reactors (CSTRs) and Packed Bed Reactors (PBRs). Moreover, PBRs show distinct advantages due to their enhanced capacity to meet demand, primarily resulting from differences in mixing patterns and the extended contact time between reactants and the biocatalyst. An essential contribution of this work is the detailed spatial analysis of temperature distribution within the PBR, an innovative approach to studying multienzyme systems. Considering a 16-week time horizon, the application of the proposed methodology resulted in a total of 3 catalyst changeovers for the CSTR configuration, and one for the PBR, achieving 80% of the total seasonal demand. Furthermore, the development of a comprehensive model that integrates two-stage enzyme inactivation, diffusional limitations, and Michaelis–Menten kinetics for both enzymes provides a thorough understanding and valuable insights into determining optimal biocatalyst replacement times. This approach advances the design and operation of biocatalytic processes for improved economic performance.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109209"},"PeriodicalIF":3.9,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144263017","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":"On ReLU neural networks as piecewise linear surrogate models","authors":"Amirhossein Hosseini ,&nbsp;Martin Guay ,&nbsp;Xiang Li","doi":"10.1016/j.compchemeng.2025.109208","DOIUrl":"10.1016/j.compchemeng.2025.109208","url":null,"abstract":"<div><div>Continuous piecewise linear (CPWL) surrogate models are increasingly used in process systems engineering to represent complex, nonlinear relationships. Neural networks with ReLU activation functions (ReLU-NN) have become a common method to represent CPWL models. However, the structure and behavior of the linear partitions formed by rectifier networks have not been fully examined. In this study, we propose exact mathematical expressions for linear functions and linear regions of small rectifier networks. Moreover, we analyze the performance of the rectifier networks from a polyhedral perspective and introduce the three major challenges associated with these models: redundancy, degeneracy, and low efficiency. Furthermore, we assess difference-of-convex continuous piecewise linear (DC-CPWL) function as an alternative representation of CPWL relationships and compare it to ReLU-based shallow and deep Neural Networks across four industrial case studies. Our findings demonstrate that the DC-CPWL representation consistently yields highly efficient models while the ReLU-NN representation generates less efficient ones.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109208"},"PeriodicalIF":3.9,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144263018","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 digital twin simulator of a pastillation process with applications to automatic control based on computer vision","authors":"Leonardo D. González ,&nbsp;Joshua L. Pulsipher ,&nbsp;Shengli Jiang ,&nbsp;Angan Mukherjee ,&nbsp;Tyler Soderstrom ,&nbsp;Victor M. Zavala","doi":"10.1016/j.compchemeng.2025.109205","DOIUrl":"10.1016/j.compchemeng.2025.109205","url":null,"abstract":"<div><div>We present a digital-twin simulator for a pastillation process. The simulation framework produces realistic thermal image data of the process that is used to train computer vision-based soft sensors based on convolutional neural networks (CNNs); the soft sensors produce output signals for temperature and product flow rate that enable real-time monitoring and feedback control. Pastillation technologies are high-throughput devices that are used in a broad range of industries; these processes face operational challenges such as real-time identification of clog locations (faults) in the rotating shell and the automatic, real-time adjustment of conveyor belt speed and operating conditions to stabilize output. The proposed simulator is able to capture this behavior and generates realistic data that can be used to benchmark different algorithms for image processing and different control architectures. We present a case study to illustrate the capabilities; the study explores behavior over a range of equipment sizes, clog locations, and clog duration. A feedback controller (tuned using Bayesian optimization) is used to adjust the conveyor belt speed based on the CNN output signal to achieve the desired process outputs.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109205"},"PeriodicalIF":3.9,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144290859","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":"Sustainable recovery of Cu, Ag and Au from the waste printed circuit boards and process optimisation by machine learning","authors":"Waqar Muhammad Ashraf ,&nbsp;Ramdayal Panda ,&nbsp;Prashant Ram Jadhao ,&nbsp;Kamal Kishore Pant ,&nbsp;Vivek Dua","doi":"10.1016/j.compchemeng.2025.109237","DOIUrl":"10.1016/j.compchemeng.2025.109237","url":null,"abstract":"<div><div>The sustainable supply of metals, especially precious metals, is critical for the manufacturing of the electronic chips used in the printed circuit boards of mobile phones. At the same time, the large volume of waste printed circuit boards (WPCBs) of mobile phones is a serious environmental issue that requires developing sustainable processes for the recovery of metals and to handle the waste in a resourceful manner. To address the two challenges of sustainable material supplies for chip manufacturing and waste management of WPCBs of mobile phones, we present a machine learning (ML) powered process optimization framework for the sustainable recovery of Cu, Ag and Au from the WPCBs. The process employs NH₄Cl and low-temperature roasting for the recovery of metals for designed experimental conditions. The input-output data obtained from the experiments is deployed to make approximations of the metal recovery profiles for Cu, Ag and Au by Gaussian Process (GP) models. The GP models trained for the three metals are embedded in the objective function of an optimisation problem for determining the optimised experimental conditions that maximise the recovery of the metals from the WPCBs. The verification of optimized experimental conditions, obtained after solving the optimization problem, in made in the lab that confirms 99 %, 90 % and 80 % respectively recovery of Cu, Ag and Au from the WPCBs. This demonstrates the effectiveness of the developed ML powered analysis workflow that improves the material utilisation efficiency and supports sustainable AI by considering material requirements for chip manufacturing and waste management.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109237"},"PeriodicalIF":3.9,"publicationDate":"2025-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144290860","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 early decision-making algorithm for accelerating topical drug formulation optimisation","authors":"Yu Zhang,&nbsp;Yongrui Xiao,&nbsp;Dimitrios Tsaoulidis,&nbsp;Tao Chen","doi":"10.1016/j.compchemeng.2025.109224","DOIUrl":"10.1016/j.compchemeng.2025.109224","url":null,"abstract":"<div><div>Formulated topical drugs (and personal care products) contain diverse and varied mixtures. The experiments for formulation design can be time-consuming, especially those for optimising the delivery of active ingredients into the skin, the so-called in vitro permeation test (IVPT). A single IVPT typically takes 24 hrs and consumes significant resources for sample collection and chemical analysis. In this study, an early decision-making algorithm (EDMA) that can terminate unpromising experiments early, thereby prioritising resources on promising ones and potentially accelerating formulation design is proposed. The algorithm relies on a flexible Gaussian process regression (GPR) model for prediction during the experiments, while the prediction uncertainty is accounted for by a statistical measure, the probability of exceedance (PoE), to guide decision-making. This algorithm was applied to maximise ibuprofen permeation from a gel-like formulation through IVPT. The results show that it is feasible to determine whether a certain formulation has the potential to achieve higher permeation before the end of experiment, leading to significant savings on time and resources.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109224"},"PeriodicalIF":3.9,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144241002","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 multi-parametric programming approach for combined flexibility, availability, and resilience in process design","authors":"Natasha J. Chrisandina ,&nbsp;Eleftherios Iakovou ,&nbsp;Efstratios N. Pistikopoulos ,&nbsp;Mahmoud M. El-Halwagi","doi":"10.1016/j.compchemeng.2025.109198","DOIUrl":"10.1016/j.compchemeng.2025.109198","url":null,"abstract":"<div><div>Process systems are perpetually vulnerable to disruptions from within and outside the system, as well as to uncertainties in operating parameters, all of which may adversely affect system performance. Incorporating resilience in the conceptual process design stage allows for the integration of various correlated design goals such as flexibility, availability, and ability to quickly recover during disruptions. In this work, a methodology based on the flexibility analysis is presented that provides a path to quantifying the ability of a proposed design to manage uncertainties and disruptions through a Combined Flexibility-Availability-Resilience Index (CFARI). This metric represents the likelihood that a design is feasible given the desired flexibility, availability, and resilience goals. The proposed method systematically explores the feasible space as described by the process constraints, uncertainties, and relevant disruptions through multi-parametric programming to determine this likelihood. Given a range of possible values for design variables, the CFARI can be correlated with design variables and then applied to a design optimization formulation to represent the resilience objective. Through this method, resilience is considered holistically through integration with flexibility and availability, and trade-offs with other objectives in the design stage can be explored. Case studies involving different process systems are presented to illustrate the applicability of the CFARI as a resilience metric.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109198"},"PeriodicalIF":3.9,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144240998","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":"Optimal design of Carbon Capture, Utilization and Storage (CCUS) networks under uncertain geological storage volumes and CO2 price","authors":"Chinmay M. Aras ,&nbsp;Lewis Ntaimo ,&nbsp;M.M. Faruque Hasan","doi":"10.1016/j.compchemeng.2025.109203","DOIUrl":"10.1016/j.compchemeng.2025.109203","url":null,"abstract":"<div><div>Carbon Capture, Utilization, and Storage (CCUS) is an enabling pathway for reducing CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> emissions from large stationary sources, such as power plants, cement, iron &amp; steel, chemical manufacturing, and other industries. Due to large number of sources and sinks and their potential connectivity, the design of optimal CCUS supply chain networks is non-trivial. Although underground reservoirs are available for CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> sequestration, uncertainties in the exact storage volumes and the future market prices of captured CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> pose significant additional challenges. Deterministic methods would fail to design even feasible, let alone optimal, CCUS networks under uncertain storage volumes and future selling prices of CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span>. In this work, we employ a stochastic optimization approach to design not only feasible but also most economic CCUS networks under a wide range of future scenarios of geological storage volumes and selling prices. We apply the method for the design of potential nationwide and regional CCUS networks in the United States. Our analysis suggests that uncertain geological storage volumes have a notable effect on the topology and the feasibility of large networks, underscoring the need for careful consideration of these factors in CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> storage selection. Interestingly, it is possible to ensure the feasibility of future CCUS networks for a wider range of scenarios with only a marginal increase in overall CCUS cost. Selling price uncertainty, on the other hand, affects the CCUS revenues. The network configuration mostly remains invariant to selling price variability. These results highlight the critical role of uncertainty-aware decision-making to ensure feasible and cost-effective CCUS deployment while de-risking large investments.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109203"},"PeriodicalIF":3.9,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144298037","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":"Multi-scale modeling and optimization of CO₂ hydrogenation to light olefins via deep learning and optimization algorithms","authors":"Yi Zhao,&nbsp;Lei Zhang,&nbsp;Libo Zhang,&nbsp;Mingshi Gong,&nbsp;Haiou Yuan","doi":"10.1016/j.compchemeng.2025.109233","DOIUrl":"10.1016/j.compchemeng.2025.109233","url":null,"abstract":"<div><div>This study presents a novel multi-scale modeling and optimization framework that integrates deep learning and advanced optimization techniques to enhance CO₂ hydrogenation for producing light olefins. A first-principles model, incorporating reaction kinetics (RWGS, FT, FTS) and heat and mass transfer dynamics, was developed for a one-dimensional fixed-bed reactor, generating a dataset of over 400,000 simulation rows. Among the evaluated deep learning architectures, the recurrent neural network (RNN) demonstrated superior predictive accuracy and robustness against 2 % Gaussian noise, establishing its efficacy as a surrogate for mechanistic models. Two optimization strategies were employed: (1) The interior-point method, leveraging gradient-based optimization, achieved propylene yields of 37.66 % by tuning inlet temperature and pressure (608.1 K, 1.6406 MPa) and 38.05 % by optimizing catalyst packing density (423 kg/m³), yielding 5.33 % and 5.72 % improvements over the baseline (32.33 %), respectively; and (2) Reinforcement learning (RL) with algorithms including DDPG, PPO, and TD3, where TD3 achieved the highest reward (34.02 %) in an RNN-based environment, demonstrating adaptive control under dynamic conditions. Comparative analysis reveals that the interior-point method excels in static, high-precision optimization, while RL offers robustness in dynamic, uncertain environments. This dual-optimization approach, augmented by domain randomization and mechanistic model augmentation to address plant-simulation mismatches, provides a robust foundation for intelligent carbon capture and utilization (CCU) systems, advancing CO₂ conversion and selective olefin synthesis.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109233"},"PeriodicalIF":3.9,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144272618","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":"Retrofitting Ordinary Portland cement production for reduced greenhouse gas emissions","authors":"Ariana Y. Ojeda-Paredes ,&nbsp;Alexander Mitsos ,&nbsp;Manuel Dahmen","doi":"10.1016/j.compchemeng.2025.109200","DOIUrl":"10.1016/j.compchemeng.2025.109200","url":null,"abstract":"<div><div>Cement production is an energy-intensive process and a major greenhouse gas (GHG) emitter. Carbon capture, utilization and storage (CCUS) technologies and fossil fuel substitution have been studied as carbon mitigation measures in the cement industry. However, their optimal combination for retrofitting the production of Ordinary Portland cement (OPC) has yet to be assessed. We formulate and optimize a superstructure to retrofit the OPC production by optimally combining CCUS technologies and fuel switching. Our analysis shows that the emerging Pareto-optimal designs heavily depend on the local conditions, notably the availability of biomass and carbon storage, electricity prices, and emission factor of the used electricity mix. Economically, carbon capture and storage (CCS) is more cost-effective than carbon capture and utilization (CCU) via power-to-methane at current costs in Germany. Only if renewable electricity can be accessed at very low cost, CCU becomes an attractive option.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109200"},"PeriodicalIF":3.9,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144254866","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":"From clutter to clarity: Emergent neural operators via questionnaire metrics","authors":"Anastasia S. Georgiou ,&nbsp;Arjun Manoj ,&nbsp;Pei-Chun Su ,&nbsp;Ronald R. Coifman ,&nbsp;Ioannis G. Kevrekidis ,&nbsp;Somdatta Goswami","doi":"10.1016/j.compchemeng.2025.109201","DOIUrl":"10.1016/j.compchemeng.2025.109201","url":null,"abstract":"<div><div>Real-world datasets in chemical engineering and bioengineering processes—such as those from catalytic reactors, multiphase flows, polymerization reactors, bioreactors, and clinical trials—can often be unlabeled or disorganized, rendering the training of existing supervised learning models ineffective at learning the underlying dynamics. To salvage these datasets for decision-making, we first seek to obtain clarity from the cluttered data. Here, we present a framework for developing “structural” generative models, discovering emergent equations, and constructing efficient emulators from scrambled datasets by integrating unsupervised organizational learning techniques (Questionnaires) with advanced deep learning architectures (Deep Hidden Physics Models and Deep Operator Networks). Our approach is demonstrated on two illustrative model systems: (a) a 1D advection–diffusion partial differential equation representing a winding underground pipe and (b) an ensemble of Stuart–Landau oscillators, an agent-based system of coupled ordinary differential equations. In both cases, we successfully reconstruct meaningful spatial, temporal, and parameter embeddings from scrambled data, enabling good predictions of system dynamics. We highlight the framework’s potential for broader applications, enabling data-driven system identification in fields with inherently disorganized or hidden parameter spaces.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"201 ","pages":"Article 109201"},"PeriodicalIF":3.9,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144254869","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}