Computers & Chemical Engineering最新文献_第2页

Hybrid semi-supervised fault detection framework under limited high-confidence data scenarios 有限高置信度数据场景下的混合半监督故障检测框架

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-14 DOI: 10.1016/j.compchemeng.2025.109186

Selorme Agbleze , Lawrence J. Shadle , Fernando V. Lima

{"title":"Hybrid semi-supervised fault detection framework under limited high-confidence data scenarios","authors":"Selorme Agbleze , Lawrence J. Shadle , Fernando V. Lima","doi":"10.1016/j.compchemeng.2025.109186","DOIUrl":"10.1016/j.compchemeng.2025.109186","url":null,"abstract":"<div><div>In the broad field of fault detection, approaches utilizing process conditions are established for systems with adequate datasets. However, systems including recently commissioned and novel processes have limited datasets available for model-based fault detection. Moreover, these systems have far greater proportions of normal operating data than adequate fault examples due to the time for which they have been operated. In this work, a combined hybrid framework for fault detection is developed that enables augmentation of the limited dataset available with HAZOP data, allowing for the utilization of both human expert knowledge and generated pseudo-process data. Additionally, the generation of artificial data is performed for reducing false positives in adversarial training. A semi-supervised distance variant of center loss is used to improve the consistency of deep feature activations from paired and unpaired data. A comparison between the proposed approach and an approach utilizing only process data in the limited data case is presented. Overall, the proposed approach shows 4.1 % and 8.8 % improvements in average detection rate when compared to the state-of-the-art supervised method for the Tennessee Eastman process and subcritical coal-fired power plant case studies, respectively, enabling the use of unlabeled data to supplement labeled process data for fault detection.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109186"},"PeriodicalIF":3.9,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144115693","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}

引用次数: 0

Encrypted machine learning-based model predictive control architectures for nonlinear systems 基于加密机器学习的非线性系统模型预测控制体系结构

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-13 DOI: 10.1016/j.compchemeng.2025.109166

Arthur Khodaverdian , Guoquan Wu , Zhe Wu , Panagiotis D. Christofides

引用次数: 0

A diffusion-attention-based algorithm for optimal spatio-temporal sensor placement in distributed parameter systems 分布参数系统中基于扩散注意的传感器时空优化配置算法

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-09 DOI: 10.1016/j.compchemeng.2025.109163

Yeong Woo Son, Jong Min Lee

{"title":"A diffusion-attention-based algorithm for optimal spatio-temporal sensor placement in distributed parameter systems","authors":"Yeong Woo Son, Jong Min Lee","doi":"10.1016/j.compchemeng.2025.109163","DOIUrl":"10.1016/j.compchemeng.2025.109163","url":null,"abstract":"<div><div>Sensor placement design (SPD) for distributed parameter systems (DPSs) remains challenging due to the vast number of potential sensor locations and the associated deployment costs. Traditional SPD methods, such as those based on observability and Kalman filters, are limited by assumptions of linearity and low sensor counts, which can be impractical in complex industrial environments. In this work, we propose a diffusion-attention-based approach that is fully data-driven, eliminating the need for explicit numerical models of the system. Our approach integrates a diffusion model—capable of progressively denoising corrupted data—and an attention mechanism that identifies the most informative sensor locations. By prioritizing sensors with higher attention weights, we ensure accurate reconstruction of the unobserved states despite using relatively few measurement points. We validate the proposed method on two benchmark DPSs, the catalytic rod and the Brusselator. Results demonstrate that our algorithm achieves sufficient accuracy in both state reconstruction and fault detection. Furthermore, the approach scales naturally to scenarios where certain states can be easily measured, thus enhancing performance.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109163"},"PeriodicalIF":3.9,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143941600","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}

引用次数: 0

Automated physical model building from literature sources: Combining equations based on four predefined requirements 从文献来源自动建立物理模型：根据四个预定义的要求组合方程

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-09 DOI: 10.1016/j.compchemeng.2025.109147

Shota Kato, Manabu Kano

{"title":"Automated physical model building from literature sources: Combining equations based on four predefined requirements","authors":"Shota Kato, Manabu Kano","doi":"10.1016/j.compchemeng.2025.109147","DOIUrl":"10.1016/j.compchemeng.2025.109147","url":null,"abstract":"<div><div>Constructing physical models from equations extracted from scientific literature databases is a challenging task due to the presence of redundant, intermediate, or inconsistent equations. This study formalizes model building as the combination of equations to form desired models that satisfy criteria such as input–output completeness and consistency. To address this problem, we propose a refined gradual method, an efficient algorithm that iteratively refines candidate equation groups while ensuring perfect recall and avoiding unnecessary computations. Evaluation of the proposed method on eight case studies, including noisy datasets, complex systems, and diverse equation structures, demonstrated that the refined gradual method reduced computational time compared to existing methods and successfully constructed all desired models. The study also identifies limitations of the proposed method and suggests improvements to enhance efficiency and adaptability. By providing a general framework for solving equation combination problems, this study advances automated model-building techniques and offers a robust approach for handling complex and noisy datasets in scientific and engineering disciplines.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109147"},"PeriodicalIF":3.9,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143948075","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}

引用次数: 0

LION Data: A roaring transformation in data visualisation LION数据：数据可视化的巨大转变

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-08 DOI: 10.1016/j.compchemeng.2025.109153

Lucy Todd , Arthur Fordham , Ben Deacon , Marc-Olivier Coppens

引用次数: 0

Meta-study and environmental techno-economic assessments (eTEAs) of blue hydrogen processes 蓝氢工艺的元研究与环境技术经济评价

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-08 DOI: 10.1016/j.compchemeng.2025.109154

Philipp S. Schwab , Thomas A. Adams II

{"title":"Meta-study and environmental techno-economic assessments (eTEAs) of blue hydrogen processes","authors":"Philipp S. Schwab , Thomas A. Adams II","doi":"10.1016/j.compchemeng.2025.109154","DOIUrl":"10.1016/j.compchemeng.2025.109154","url":null,"abstract":"<div><div>Blue hydrogen, produced by retrofitting conventional hydrogen technologies with carbon dioxide (CO<sub>2</sub>) capture (CC) systems, offers a low-carbon alternative to gray hydrogen and a transitional pathway alongside green hydrogen, which is generated through renewable-energy-powered water electrolysis. While studies indicate that cradle-to-plant-exit emissions for blue hydrogen can approach those of green hydrogen, inconsistent assumptions in techno-economic and life-cycle assessments, such as variations in plant location, feedstock type, and production scale, complicate direct comparisons and hinder the development of standardized benchmarks. This work addresses these challenges by conducting a systematic literature review of recent environmental techno-economic assessments for blue hydrogen production. A key contribution is the application of a consistent set of assumptions, system boundaries, and definitions to harmonize and standardize the reported data. By providing a unified framework for evaluating blue hydrogen technologies, this study resolves discrepancies in prior assessments and facilitates more accurate comparisons across studies. Implementing CC technologies on conventional hydrogen production can reduce life cycle CO<sub>2</sub> emissions by around 60% (comparable to that of green hydrogen) with a cost increase of only 30% compared to unabated gray hydrogen. Producing green hydrogen results in costs up to three times higher than gray hydrogen and more than double that of blue hydrogen. These results offer valuable insights for policymakers and industry stakeholders to support the adoption and optimization of low-carbon hydrogen technologies.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109154"},"PeriodicalIF":3.9,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143948076","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}

引用次数: 0

Superstructure hybrid optimization framework: An innovative approach for multi-objective supply chain optimization 上层结构混合优化框架：一种多目标供应链优化的创新方法

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-06 DOI: 10.1016/j.compchemeng.2025.109175

Rogelio Ochoa-Barragán, César Ramírez-Márquez, José María Ponce-Ortega

{"title":"Superstructure hybrid optimization framework: An innovative approach for multi-objective supply chain optimization","authors":"Rogelio Ochoa-Barragán, César Ramírez-Márquez, José María Ponce-Ortega","doi":"10.1016/j.compchemeng.2025.109175","DOIUrl":"10.1016/j.compchemeng.2025.109175","url":null,"abstract":"<div><div>Modeling process systems is highly complex and requires advanced tools to generate optimal solutions effectively. While no perfect mathematical model exists, integrating artificial intelligence with mathematical optimization helps overcome the limitations of traditional models. This paper presents a novel approach to supply chain optimization by incorporating machine learning models directly into optimization frameworks. Unlike traditional methods that require complex reformulations for integrating advanced machine learning algorithms, our strategy allows seamless incorporation into both deterministic and metaheuristic techniques, simplifying hybrid model implementation. To evaluate the proposed strategy, we analyze the installation of bioethanol biorefineries in the Michoacán region, Mexico, a major sugarcane producer. This case study leverages well-established processes to generate high-quality data for training machine learning models that enhance predictive accuracy. Additionally, it enables a comprehensive economic and environmental assessment of bioethanol production, highlighting its role in energy security, greenhouse gas reduction, and rural economic development. Results show that using a linear regression model and a neural network model yields an annual profit of 2.55 MM USD while minimizing costs and environmental impact. Although predictive models do not fully replace process simulators, our approach demonstrates that decision-making can significantly benefit from hybrid models, reducing computational complexity while maintaining accuracy.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109175"},"PeriodicalIF":3.9,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143928417","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}

引用次数: 0

Enhancing industrial scheduling through machine learning: A synergistic approach with predictive modeling and clustering 通过机器学习增强工业调度：预测建模和聚类的协同方法

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-05 DOI: 10.1016/j.compchemeng.2025.109174

Maria E. Samouilidou, Nikolaos Passalis, Georgios P. Georgiadis, Michael C. Georgiadis

{"title":"Enhancing industrial scheduling through machine learning: A synergistic approach with predictive modeling and clustering","authors":"Maria E. Samouilidou, Nikolaos Passalis, Georgios P. Georgiadis, Michael C. Georgiadis","doi":"10.1016/j.compchemeng.2025.109174","DOIUrl":"10.1016/j.compchemeng.2025.109174","url":null,"abstract":"<div><div>In this study, a novel solution framework is developed integrating Machine Learning (ML) techniques with Mixed-Integer Linear Programming (MILP) to address the optimization of production scheduling in manufacturing industries characterized by multiple products, shared resources and parallel production lines. The dynamic nature of these industries often requires rapid schedule adjustments within minutes to address unexpected events. At the same time, manufacturing facilities face frequent changeover operations to prevent cross-contamination. Inefficient product allocation to packing lines often leads to significant downtime and material waste, especially when introducing new products with unrecorded changeover times. To overcome these challenges, the proposed framework first compiles a representation space in which distances correspond to changeover times. This enables the employment of constrained clustering to group production orders according to the available packing lines, minimizing changeover times within each cluster. Then, the derived allocation is used to restrict the solution space of an MILP-based scheduling model to reduce its computational complexity. Furthermore, to tackle the issue of unavailable changeover data, a predictive ML model is trained to predict unknown changeover times for new or existing products. An evaluation study based on a construction materials plant is conducted to test the applicability of the framework. It is concluded that the proposed approach achieves accurate solutions rapidly, reduces downtime and facilitates the smooth integration of new products into the production process. In addition, it is applicable to a wide range of industries, and it enables the extension to an online scheduling framework due to its computational speed.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109174"},"PeriodicalIF":3.9,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143931698","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}

引用次数: 0

A graph-theoretic framework for analyzing and designing chemical engineering curricula 化工课程分析与设计的图论框架

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-04 DOI: 10.1016/j.compchemeng.2025.109150

Blake Lopez, Yue Shao, Victor M. Zavala

{"title":"A graph-theoretic framework for analyzing and designing chemical engineering curricula","authors":"Blake Lopez, Yue Shao, Victor M. Zavala","doi":"10.1016/j.compchemeng.2025.109150","DOIUrl":"10.1016/j.compchemeng.2025.109150","url":null,"abstract":"<div><div>Topics and courses that compose chemical engineering curricula are interconnected in a complex manner. The organization/structure of chemical engineering curricula closely matches the practice of breaking down chemical processes into fundamental phenomena (e.g., thermo, balances, and transport) and unit operations (e.g., reactors, separators, and heat exchangers). Emergence of modern topics (e.g., sustainability and molecular engineering) and advances in pedagogy call for the analysis and potential re-organization of curricula (e.g., use of case studies to foster integration of courses and include new topics/courses in a synergistic manner). In this work, we propose a graph-theoretic abstraction to represent, analyze, and reorganize the structure of curricula. In this abstraction, nodes represent topics/concepts, edges represent connectivity/dependencies between topics, and courses can be interpreted as collections of topics that are tightly interconnected (also known as clusters or modules). The abstraction enables the use of algorithms and software tools of graph theory and optimization to formalize the visualization and evaluation of curricula (e.g., identify key topics) and to identify re-organization strategies (e.g., defining strategic modules/courses that maximize topic cohesiveness/connectivity). Additionally, the abstraction can help formalize and facilitate discussions between instructors that might have different priorities/perspectives on curriculum content and organization. We provide case studies that analyze real curricula at the University of Wisconsin–Madison to highlight the benefits of the proposed framework.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"200 ","pages":"Article 109150"},"PeriodicalIF":3.9,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143929084","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}

引用次数: 0

ChemDT: A stochastic decision transformer for chemical process control 化学过程控制的随机决策转换器

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-05-03 DOI: 10.1016/j.compchemeng.2025.109155

Junseop Shin, Joonsoo Park, Jaehyun Shim, Jong Min Lee

{"title":"ChemDT: A stochastic decision transformer for chemical process control","authors":"Junseop Shin, Joonsoo Park, Jaehyun Shim, Jong Min Lee","doi":"10.1016/j.compchemeng.2025.109155","DOIUrl":"10.1016/j.compchemeng.2025.109155","url":null,"abstract":"<div><div>The rapid advancement of industries has complicated process modeling, as conventional model-based control methods struggle with models that inadequately capture system complexities and impose significant computational burdens on their use. Reinforcement learning (RL), which leverages practical operational data instead of explicit models, often adapts better to these complexities. However, RL’s need for extensive online exploration poses potential risks in sensitive environments like chemical processes. To address this, we propose an offline RL approach based on the Decision Transformer (DT) architecture, named ChemDT. ChemDT incorporates stochastic policies with maximum entropy regularization, broadening policy coverage under limited offline data. To mitigate DT’s vulnerability to stochastic environments, we introduce a monitoring variable, <span><math><mi>λ</mi></math></span>, enabling selective responses to significant stochastic events amidst pervasive disturbances. Validated on a Continuous Stirred Tank Reactor (CSTR) and an industrial-scale fed-batch reactor, our approach demonstrates superior control performance compared to other offline RL algorithms.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"199 ","pages":"Article 109155"},"PeriodicalIF":3.9,"publicationDate":"2025-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143916842","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}

引用次数: 0