Computers & Chemical Engineering最新文献_第2页

From text to meaning: Semantic interpretation of non-standardized metadata in piping and instrumentation diagrams 从文本到含义：管道和仪表图中非标准化元数据的语义解释

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-08 DOI: 10.1016/j.compchemeng.2025.109436

Vasil Shteriyanov , Rimma Dzhusupova , Jan Bosch , Helena Holmström Olsson

{"title":"From text to meaning: Semantic interpretation of non-standardized metadata in piping and instrumentation diagrams","authors":"Vasil Shteriyanov , Rimma Dzhusupova , Jan Bosch , Helena Holmström Olsson","doi":"10.1016/j.compchemeng.2025.109436","DOIUrl":"10.1016/j.compchemeng.2025.109436","url":null,"abstract":"<div><div>The extraction of structured metadata from Piping and Instrumentation Diagrams (P&IDs) is a major bottleneck for digitalization in the process industries. Existing methods, based on Optical Character Recognition (OCR), stop at raw text extraction, failing to interpret critical engineering information encoded within variable-format identifiers like pipeline numbers. This paper bridges this semantic gap by introducing a system for the format-aware interpretation of P&ID pipeline metadata. Our hybrid system architecture integrates deep learning for text recognition with domain interpretation rules that allow the system to adapt to new project formats without model retraining. These rules perform validation, error correction, and semantic mapping of raw text to structured data. We validated our system on a challenging dataset of real-world P&IDs from four distinct industrial projects, each with a unique and complex pipeline number format. Our method achieved 91.1% end-to-end accuracy, demonstrating a significant leap in performance over standard OCR tools, which proved insufficient for the task. This work presents a robust solution that unlocks valuable data from non-standardized engineering documents, providing a practical pathway for creating reliable digital twins and supporting plant lifecycle management in the chemical engineering sector.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109436"},"PeriodicalIF":3.9,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145262345","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

Imbalanced fault diagnosis based on sample-weighted counterfactual 基于样本加权反事实的不平衡故障诊断

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-07 DOI: 10.1016/j.compchemeng.2025.109434

Wei Zheng, Chunfei Gu, Hao Pan, Xin Fan, Sixiang Fu, Xiaoheng Ji

{"title":"Imbalanced fault diagnosis based on sample-weighted counterfactual","authors":"Wei Zheng, Chunfei Gu, Hao Pan, Xin Fan, Sixiang Fu, Xiaoheng Ji","doi":"10.1016/j.compchemeng.2025.109434","DOIUrl":"10.1016/j.compchemeng.2025.109434","url":null,"abstract":"<div><div>In fault detection tasks, the scarcity of fault samples often leads models to learn primarily from normal samples, resulting in biased predictions. To address class imbalance, this study introduces a data augmentation framework. It combines sample weighting based on stable learning with counterfactual generation. First, sample weighting is applied to enhance the model’s ability to capture true feature-outcome causal relationships. Then, a clustering algorithm selects high-weight samples with substantial distribution differences. Based on these representative normal samples, a multi-objective counterfactual generation method synthesizes fault samples under physical constraints, while weighted feature importance is used to identify key diagnostic features. The proposed approach effectively alleviates data imbalance in complex industrial fault diagnosis, improving both the accuracy and interpretability of fault detection. Experiments conducted on the Tennessee Eastman Process and the Multi-phase Flow Process show that our method significantly improves fault diagnosis performance under imbalanced data conditions.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109434"},"PeriodicalIF":3.9,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145262346","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

Regional sustainability of food–energy–water nexus considering water stress using multi-objective modeling and optimization 考虑水分胁迫的食物-能量-水关系区域可持续性研究

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-06 DOI: 10.1016/j.compchemeng.2025.109433

Anupam Satyakam Sankoju, Yogendra Shastri

{"title":"Regional sustainability of food–energy–water nexus considering water stress using multi-objective modeling and optimization","authors":"Anupam Satyakam Sankoju, Yogendra Shastri","doi":"10.1016/j.compchemeng.2025.109433","DOIUrl":"10.1016/j.compchemeng.2025.109433","url":null,"abstract":"<div><div>In this work, a regional optimization model is developed to achieve energy–water nexus sustainability in the context of lignocellulosic biofuels. The regional water availability is determined as a function of precipitation. Detailed water balance at the district level accounts for groundwater recharge and surface water availability. Water left over from the previous season can be used in the next season. The water withdrawal for crops providing agricultural residue to produce ethanol is limited by a seasonal water stress factor. The key decision variables are land allocation for selected crops and choice of residues for ethanol production, all while ensuring state-level ethanol production targets are met. The primary goals are to maximize farmers’ profit, minimize ethanol cost, and minimize water withdrawals while meeting the ethanol production targets. The resulting multi-objective mixed-integer linear programming problem is applied to a case study of 33 districts in Maharashtra, India. Results compare optimized land practices from the model with current practices, where current land use causes 78.5% water stress. Upon further reducing water consumption by 38.5% compared to the current land use practices and for an ethanol blending rate of 15%, profits reduced by 27.6%, from USD 388.06/ha to USD 288.88/ha. The ethanol water footprint decreased 1.7-fold from 937 to 528 liters per liter of ethanol, and irrigation water use dropped 14.8-fold from 3878 to 262 m<span><math><msup><mrow></mrow><mrow><mn>3</mn></mrow></msup></math></span>/ha.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"205 ","pages":"Article 109433"},"PeriodicalIF":3.9,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145265071","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

Generative transformer-based deep hierarchical VAE model for the automated generation of chemical process topologies 基于生成式变压器的化工过程拓扑自动生成深层分层VAE模型

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-06 DOI: 10.1016/j.compchemeng.2025.109431

Yeong Woo Son , Ji Hun Pak , Chan Kim, Jong Min Lee

{"title":"Generative transformer-based deep hierarchical VAE model for the automated generation of chemical process topologies","authors":"Yeong Woo Son , Ji Hun Pak , Chan Kim, Jong Min Lee","doi":"10.1016/j.compchemeng.2025.109431","DOIUrl":"10.1016/j.compchemeng.2025.109431","url":null,"abstract":"<div><div>Chemical process synthesis involves two key challenges: defining the process topology and specifying the physicochemical details. To address the first challenge, this work presents a data-driven framework for the automated generation of diverse and structurally valid process topologies. Our approach utilizes a transformer-based generative model to learn the underlying grammar of process structures from a large dataset of designs. By learning a flexible latent representation and enabling constraint-aware generation, our framework rapidly produces a wide range of novel candidate topologies for subsequent, engineering analysis. We compile a database of real-world process flow diagrams (PFDs) and augment it with synthetically generated process topologies using a higher-order Markov model. All flowsheets are encoded as structured text sequences using the simplified flowsheet input-line entry system (SFILES), allowing compatibility with transformer architectures. We train a generative model that integrates a modified transformer architecture with a deep hierarchical variational autoencoder (VAE), and apply a constrained beam search algorithm to ensure syntactic validity and design feasibility. Key contributions include: (1) a transformer-based generation method for latent vector-guided flexible process topology generation; (2) data augmentation using a higher-order Markov model; (3) a SFILES structural validator that checks the grammar and logic of process topologies; (4) a novel model architecture integrating a modified transformer decoder with a hierarchical VAE; and (5) a constrained beam search decoding strategy that enforces design requirements during sequence generation. Our results show that the proposed framework is capable of generating diverse, valid, and feasible topologies, offering a scalable approach to early-stage process development.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"205 ","pages":"Article 109431"},"PeriodicalIF":3.9,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145322891","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

Probabilistic assessment and automatic detection of oscillations in industrial control loops 工业控制回路振荡的概率评估与自动检测

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-05 DOI: 10.1016/j.compchemeng.2025.109437

An-qi Guan , Fang-na Xiang , Ling-feng Hang , Zhi-yan Li , Zhen-hao Lin , Zhi-jiang Jin , Jin-yuan Qian

{"title":"Probabilistic assessment and automatic detection of oscillations in industrial control loops","authors":"An-qi Guan , Fang-na Xiang , Ling-feng Hang , Zhi-yan Li , Zhen-hao Lin , Zhi-jiang Jin , Jin-yuan Qian","doi":"10.1016/j.compchemeng.2025.109437","DOIUrl":"10.1016/j.compchemeng.2025.109437","url":null,"abstract":"<div><div>Loop oscillation is a prevalent issue in industrial control loops. Affected by changes in production tasks, loop load, and external environment, industrial control systems typically have more complex oscillation patterns. Industrial signals often exhibit multimodal superposition, noise interference, and non-stationarity. Binary judgment of oscillation is prone to false alarms or missed detections in industrial control loops. More fundamentally, the binary classification framework fails to quantify oscillation risks. Therefore, complex oscillations in industrial control loops still need a more flexible assessment framework. In this paper, a probabilistic assessment framework for oscillations is proposed from the perspective of the statistical characteristics of zero-crossings. To enhance the reliability of signal preprocessing, adaptive VMD and significant IMFs identification are combined. By incorporating the statistical characteristics for coefficient of variation into regularity test of oscillation, the conventional binary classification is transformed into the probabilistic assessment. In simulation studies, the effectiveness of adaptive VMD, significant IMFs identification, and probabilistic assessment in complex signals and negative feedback loops is verified through three examples. In industrial scenario studies, the performance of the proposed method is analyzed in 93 benchmark industrial loops. The proposed method is compared with 12 distinct methods. The detection results of benchmark industrial loops show that the performance of this method is superior to most detection methods. The proposed method can not only ensure high accuracy, sensitivity and specificity, but also evaluate and grade the oscillation probability without historical data and model training. This detection method provides reference value for risk classification and decision optimization of process monitoring.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"205 ","pages":"Article 109437"},"PeriodicalIF":3.9,"publicationDate":"2025-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324018","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

Integration of graph-based descriptors with machine learning algorithm for QSPR modeling of fluoroquinolones 基于图的描述符与机器学习算法在氟喹诺酮类药物QSPR建模中的集成

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-04 DOI: 10.1016/j.compchemeng.2025.109430

Muhammad Farhan Hanif

{"title":"Integration of graph-based descriptors with machine learning algorithm for QSPR modeling of fluoroquinolones","authors":"Muhammad Farhan Hanif","doi":"10.1016/j.compchemeng.2025.109430","DOIUrl":"10.1016/j.compchemeng.2025.109430","url":null,"abstract":"<div><div>In this article, QSPR analysis of antibiotic drugs such as Captopril, Norfloxacin, Dorzolamide, Saquinavir, Indinavir, Ritonavir, Oseltamivir, Zanamivir, Imatinib, Zolmitriptan, and Aliskiren has been investigated. The following properties were considered: density (D), refractive index (IR), molar refractivity (MR), polarizability (POL) surface tension (ST), Bioconcentration Factor (BCF) and molar volume (MV). The modified reverse counterparts were applied to model the relationship between molecular structure and physicochemical properties as effective descriptors for the prediction of drug behavior. Predictive models were thereafter developed, focusing on the role these indices might play in capturing structural influences, aided by different types of statistical regression models, including both linear and cubic, joined lately by the Extreme Gradient Boosting (XGBoost) machine learning algorithm, a novel tree-based ensemble model useful for testing in comparison with traditional approaches that rely on regression. On such grounds, it clearly follows from this that robustness in the performance of adjusted topological indices is achieved when approaches are combined with some state-of-the-art regression methods. Among these, XGBoost had the best predictive capability, nonlinearly modeling the relationships between structure and property more effectively than any other regression model. This study focuses on the potential for integrating degree-based topological indices with the machine learning tolerance of QSPR modeling.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109430"},"PeriodicalIF":3.9,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145262343","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

Modeling methane production prediction for energy optimization via improved long short-term memory network 基于改进长短期记忆网络的能源优化甲烷产量预测建模

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-03 DOI: 10.1016/j.compchemeng.2025.109426

Yongming Han , Liyuan Feng , Mengzhi Wang , Yue Wang , Min Liu , Xingxing Zhang , Zhiqiang Geng

{"title":"Modeling methane production prediction for energy optimization via improved long short-term memory network","authors":"Yongming Han , Liyuan Feng , Mengzhi Wang , Yue Wang , Min Liu , Xingxing Zhang , Zhiqiang Geng","doi":"10.1016/j.compchemeng.2025.109426","DOIUrl":"10.1016/j.compchemeng.2025.109426","url":null,"abstract":"<div><div>Methane, a highly essential industrial raw material, plays a pivotal role in safeguarding national energy security and advancing sustainable development. Due to the expansion of industrial scale and increased integration in modern methane production, the production data exhibits complex multiscale variability over time, which poses great challenges for accurate methane production prediction. Therefore, a novel production prediction model is proposed by employing an improved Long Short-Term Memory Network (LSTM) combining with the multiscale feature fusion method (MSFF) (MSFF-LSTM). The MSFF decomposes the raw industrial process data into multiple two-dimensional tensors based on periods, which can ravel out the complex temporal fluctuations into multiple intraperiod- and interperiod-variations. Then, the methane prediction model is constructed utilizing multiple LSTM models to extract interactive features at various scales. Finally, using a feature fusion module to fuse the prediction results at different scales can fully aggregate local and global features for complementary prediction. Experimental results demonstrate that, compared with other prediction models, the MSFF-LSTM achieves the state-of-the-art results with the mean absolute error (MAE), the mean square error (MSE), coefficient of determination (R<sup>2</sup>) and the root mean square error (RMSE) of 0.1056, 0.0300, 0.9199 and 0.1733, respectively, which offers the optimization direction for the anaerobic digestion process of straw for methane production.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109426"},"PeriodicalIF":3.9,"publicationDate":"2025-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145262871","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 practical method for maintaining the validity of hybrid AI process models without retraining 一种无需再训练即可维持混合人工智能过程模型有效性的实用方法

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-02 DOI: 10.1016/j.compchemeng.2025.109432

Hsiao-Te Liu , Ming-Chun Fang , Hao-Yeh Lee , Jeffrey D. Ward , Cheng-Ting Hsieh , Tzu-Chieh Hua , Shih-Chieh Lin , Chih-Lung Lee , Tzu-Hsien Huang , Wei-Ti Chou

{"title":"A practical method for maintaining the validity of hybrid AI process models without retraining","authors":"Hsiao-Te Liu , Ming-Chun Fang , Hao-Yeh Lee , Jeffrey D. Ward , Cheng-Ting Hsieh , Tzu-Chieh Hua , Shih-Chieh Lin , Chih-Lung Lee , Tzu-Hsien Huang , Wei-Ti Chou","doi":"10.1016/j.compchemeng.2025.109432","DOIUrl":"10.1016/j.compchemeng.2025.109432","url":null,"abstract":"<div><div>A challenge with AI models is that model validity usually deteriorates over time as properties of the process gradually change. For example, catalyst activity in reactors may decrease due to poisoning or thermal degradation, the heat transfer coefficient in heat exchangers may decrease due to fouling, and the tray efficiency in distillation columns may decrease due to plugging. Usually retraining is required to restore the model prediction performance, which is costly and time-consuming.</div><div>To address this problem, a practical hybrid AI framework is proposed that incorporates equipment parameters such as tray efficiency as input variables. This allows the model accuracy to be restored by adjusting these parameters, which significantly reduces the need for the time-consuming and expensive process of retraining the model. The industrial applicability of this method is demonstrated using an industrial process for coke oven gas (COG) scrubbing and light oil recovery.</div><div>The results show that the error of using the AI model developed in 2024 to predict the data for 2021 is up to 37.55%. By adjusting the equipment parameters, the error can be reduced to 3.77%. This method can effectively solve the problem of model degradation over time and can be applied to most chemical processes.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109432"},"PeriodicalIF":3.9,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145262874","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

Leaking source localization approach in multi-obstacle scenarios based on CNN and attention mechanism 基于CNN和注意机制的多障碍场景泄漏源定位方法

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-10-02 DOI: 10.1016/j.compchemeng.2025.109435

Ye Jiang , Yu Wang , Hanxiao Qian , Yue Quan , Zhuang Jiang , Yili Chu , Di Wu

{"title":"Leaking source localization approach in multi-obstacle scenarios based on CNN and attention mechanism","authors":"Ye Jiang , Yu Wang , Hanxiao Qian , Yue Quan , Zhuang Jiang , Yili Chu , Di Wu","doi":"10.1016/j.compchemeng.2025.109435","DOIUrl":"10.1016/j.compchemeng.2025.109435","url":null,"abstract":"<div><div>Hazardous gas leaks seriously threaten human life, property, and the ecological environment. A timely and accurate approach to locating the leaking source can prevent further expansion of the leakage and facilitate subsequent rescue and repair work. Therefore, the localization of leaking source is of great significance. However, gas diffusion in multi-obstacle scenes is highly random and complex. Most of the traditional localization approaches do not consider the multiple factors that affect gas diffusion and lead to low accuracy. In this paper, FLUENT simulation software is used to build a three-dimensional scene with complex obstacles and simulate several SO<sub>2</sub> leaking scenes based on the real chemical industry park firstly. Then multiple feature maps are constructed using concentration data collected from several monitoring points, serving as input data for the neural network. And a CNN model with attention mechanism is designed to identify the leakage scenes. The final experimental results verify the effectiveness of the localization approach proposed.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109435"},"PeriodicalIF":3.9,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145262872","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

Causal discovery in industrial systems via physics-guided variational attention and probabilistic interventions 通过物理引导的变分注意和概率干预在工业系统中的因果发现

IF 3.9 2区工程技术

Computers & Chemical Engineering Pub Date : 2025-09-30 DOI: 10.1016/j.compchemeng.2025.109420

Mohammadhossein Modirrousta , Alireza Memarian , Biao Huang

{"title":"Causal discovery in industrial systems via physics-guided variational attention and probabilistic interventions","authors":"Mohammadhossein Modirrousta , Alireza Memarian , Biao Huang","doi":"10.1016/j.compchemeng.2025.109420","DOIUrl":"10.1016/j.compchemeng.2025.109420","url":null,"abstract":"<div><div>Industry 4.0 technologies demand robust fault detection and diagnosis systems distinguishing genuine causal relationships from spurious correlations in complex industrial processes. Traditional correlation-based approaches exhibit significant limitations with nonlinear dynamics, temporal dependencies, and uncertain operational conditions. This paper presents a physics-guided variational attention framework for causal discovery, integrating log-normal variational attention mechanisms with probabilistic interventions and domain expertise. The dual-attention architecture utilizes multivariate log-normal distributions to model asymmetric, positive-valued causal strengths, addressing limitations of symmetric Gaussian parameterizations. Physics-informed priors from operator knowledge are incorporated through Gaussian Mixture Models and transformed via moment-matching. Uncertainty quantification employs Monte Carlo sampling and conformal filtering for statistically rigorous causal validation. Evaluation across synthetic time-series data, Australian Refinery Process oscillation diagnosis, and Tennessee Eastman Process demonstrates superior performance versus baseline approaches. Log-normal variational attention consistently outperforms Gaussian alternatives, with physics-informed priors providing improvements under high-uncertainty conditions, establishing a robust foundation for industrial causal discovery applications.</div></div>","PeriodicalId":286,"journal":{"name":"Computers & Chemical Engineering","volume":"204 ","pages":"Article 109420"},"PeriodicalIF":3.9,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145216573","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