Pub Date : 2025-07-28DOI: 10.1016/j.dche.2025.100251
Bálint Levente Tarcsay, János Abonyi, Sándor Németh
This work presents clustering algorithms for identifying movement patterns in trajectories, with a focus on applications in chemical engineering. The exponential growth of dynamic system data necessitates algorithms that account for both local and global trajectory trends. Existing methods often overlook these aspects. We propose two DBSCAN-based variants that cluster trajectories from dynamic systems using agglomeration criteria reflecting the temporal evolution of object neighborhoods in phase space. The first algorithm groups objects with similar movement patterns over a defined observation period, while the second clusters objects with consistent neighborhood similarity over extended periods. These approaches enable the identification of localized neighborhood preservation and trajectory similarity, alongside global trends. We demonstrate the method by clustering particle trajectories generated via computational fluid dynamics, revealing characteristic flow regions within a tank equipped with static mixers. This highlights the methods’ utility for analyzing and optimizing dynamic processes in chemical engineering.
{"title":"Neighborhood preservation-based trajectory clustering for analyzing temporal behavior of dynamic systems","authors":"Bálint Levente Tarcsay, János Abonyi, Sándor Németh","doi":"10.1016/j.dche.2025.100251","DOIUrl":"10.1016/j.dche.2025.100251","url":null,"abstract":"<div><div>This work presents clustering algorithms for identifying movement patterns in trajectories, with a focus on applications in chemical engineering. The exponential growth of dynamic system data necessitates algorithms that account for both local and global trajectory trends. Existing methods often overlook these aspects. We propose two DBSCAN-based variants that cluster trajectories from dynamic systems using agglomeration criteria reflecting the temporal evolution of object neighborhoods in phase space. The first algorithm groups objects with similar movement patterns over a defined observation period, while the second clusters objects with consistent neighborhood similarity over extended periods. These approaches enable the identification of localized neighborhood preservation and trajectory similarity, alongside global trends. We demonstrate the method by clustering particle trajectories generated via computational fluid dynamics, revealing characteristic flow regions within a tank equipped with static mixers. This highlights the methods’ utility for analyzing and optimizing dynamic processes in chemical engineering.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"16 ","pages":"Article 100251"},"PeriodicalIF":4.1,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144750404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-25DOI: 10.1016/j.dche.2025.100256
Jana Mousa, Stéphane Negny, Rachid Ouaret
The increasing reliance on neural networks (NN) in chemical process modeling highlights their capability for accurate predictions, yet their standalone application often struggles to adhere to fundamental physical laws such as equilibrium constraints and mass balance. Addressing this limitation, hybrid methods that integrate data-driven insights with physical consistency have gained prominence. This study systematically explores the integration of NNs with nonlinear data reconciliation (NDR) across multiple testing dimensions, including a Gibbs reactor, data robustness evaluations, and reactor-distillation system integration. Hybrid methodologies such as NN + NDR, NN + KKT (Karush-Kuhn-Tucker), and KKT + PINN (Physics-Informed Neural Networks with KKT conditions) are comparatively assessed. The proposed NN + NDR framework demonstrates superior performance in minimizing errors and enforcing physical laws, with minimal computational overhead. This work emphasizes the scalability, robustness, and transformative potential of modular hybrid strategies in advancing reliable, physically consistent chemical process modeling.
{"title":"Hybrid neural networks for improved chemical process modeling: Bridging data-driven insights with physical consistency","authors":"Jana Mousa, Stéphane Negny, Rachid Ouaret","doi":"10.1016/j.dche.2025.100256","DOIUrl":"10.1016/j.dche.2025.100256","url":null,"abstract":"<div><div>The increasing reliance on neural networks (NN) in chemical process modeling highlights their capability for accurate predictions, yet their standalone application often struggles to adhere to fundamental physical laws such as equilibrium constraints and mass balance. Addressing this limitation, hybrid methods that integrate data-driven insights with physical consistency have gained prominence. This study systematically explores the integration of NNs with nonlinear data reconciliation (NDR) across multiple testing dimensions, including a Gibbs reactor, data robustness evaluations, and reactor-distillation system integration. Hybrid methodologies such as NN + NDR, NN + KKT (Karush-Kuhn-Tucker), and KKT + PINN (Physics-Informed Neural Networks with KKT conditions) are comparatively assessed. The proposed NN + NDR framework demonstrates superior performance in minimizing errors and enforcing physical laws, with minimal computational overhead. This work emphasizes the scalability, robustness, and transformative potential of modular hybrid strategies in advancing reliable, physically consistent chemical process modeling.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"16 ","pages":"Article 100256"},"PeriodicalIF":4.1,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144721106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-23DOI: 10.1016/j.dche.2025.100249
Yuanyuan Zou, Xu Ma, Yaru Yang, Shaoyuan Li
With the increasing complexity of production requirements and the constant change of operating conditions, the optimization of process control systems (PCSs) has become an important issue in chemical industry production. Motivated by this urgent need, an overview of advanced real-time optimization, model predictive control, and data-driven operation-optimization approaches is presented. In particular, our discussions highlight approaches that focus on typical problems such as dynamic and steady-state economic performance improvement, robust constraint satisfaction, stable and offset-free operation, and multi-mode operation, which should be addressed foremostly under complex operating conditions. The aim of this paper is to provide a better understanding of the methods and their parameter-tuning routines, which can be a reference for the readers to align the suitable techniques with the PCSs, according to the practical operation-optimization requirements in chemical processes.
{"title":"An overview of chemical process operation-optimization under complex operating conditions","authors":"Yuanyuan Zou, Xu Ma, Yaru Yang, Shaoyuan Li","doi":"10.1016/j.dche.2025.100249","DOIUrl":"10.1016/j.dche.2025.100249","url":null,"abstract":"<div><div>With the increasing complexity of production requirements and the constant change of operating conditions, the optimization of process control systems (PCSs) has become an important issue in chemical industry production. Motivated by this urgent need, an overview of advanced real-time optimization, model predictive control, and data-driven operation-optimization approaches is presented. In particular, our discussions highlight approaches that focus on typical problems such as dynamic and steady-state economic performance improvement, robust constraint satisfaction, stable and offset-free operation, and multi-mode operation, which should be addressed foremostly under complex operating conditions. The aim of this paper is to provide a better understanding of the methods and their parameter-tuning routines, which can be a reference for the readers to align the suitable techniques with the PCSs, according to the practical operation-optimization requirements in chemical processes.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"16 ","pages":"Article 100249"},"PeriodicalIF":3.0,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144702516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-18DOI: 10.1016/j.dche.2025.100250
Kathleen B. Aviso
There are several challenges to decarbonizing the chemical industry as it utilizes significant amounts of fossil fuels as feedstock and as source of energy. As a result, the industry contributes about 5 % to global CO2 emissions. Various strategies and technologies which include the use of alternative feedstock, electrification, and negative emissions technologies are available to aid in the industry’s decarbonization. These strategies can be implemented at different stages of the chemical production life cycle. The adoption of digital technologies has reported improvements in the economic, environmental, and societal performance of manufacturing industries. This review intends to investigate how available digital technologies can be utilized to accelerate the decarbonization of the chemical industry.
{"title":"Decarbonizing the chemical industry through digital technologies","authors":"Kathleen B. Aviso","doi":"10.1016/j.dche.2025.100250","DOIUrl":"10.1016/j.dche.2025.100250","url":null,"abstract":"<div><div>There are several challenges to decarbonizing the chemical industry as it utilizes significant amounts of fossil fuels as feedstock and as source of energy. As a result, the industry contributes about 5 % to global CO<sub>2</sub> emissions. Various strategies and technologies which include the use of alternative feedstock, electrification, and negative emissions technologies are available to aid in the industry’s decarbonization. These strategies can be implemented at different stages of the chemical production life cycle. The adoption of digital technologies has reported improvements in the economic, environmental, and societal performance of manufacturing industries. This review intends to investigate how available digital technologies can be utilized to accelerate the decarbonization of the chemical industry.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"16 ","pages":"Article 100250"},"PeriodicalIF":3.0,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144329715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-11DOI: 10.1016/j.dche.2025.100248
Rei Tamaki , Yusuke Hayashi , Yuki Uno , Masahiro Kino-oka , Hirokazu Sugiyama
This work presents a circular exploration of cryoprotective agents (CPAs) for stem cells using computer-aided molecular design approaches that can comprehensively consider compounds. An exploration cycle was developed that consists of the following five steps: setting conditions, computational evaluation, experimental evaluation, verification experiments, and discussions with experts in biotechnology. It aims to discover promising CPA candidate compounds by incorporating domain knowledge through discussions with the experts. The developed cycle can be applied to fields where the required physical properties have not been clearly known. As a result, 1-methylimidazole and pyridazine were selected as promising CPA candidate compounds, which were both heterocyclic amines. Hence, heterocyclic amines could be a stepping-stone toward the future development of CPAs for stem cells. By repeatedly using the exploration cycle, CPA candidate compounds with better cryoprotective effects could be discovered.
{"title":"A circular exploration of cryoprotective agents for stem cells using computer-aided molecular design approaches","authors":"Rei Tamaki , Yusuke Hayashi , Yuki Uno , Masahiro Kino-oka , Hirokazu Sugiyama","doi":"10.1016/j.dche.2025.100248","DOIUrl":"10.1016/j.dche.2025.100248","url":null,"abstract":"<div><div>This work presents a circular exploration of cryoprotective agents (CPAs) for stem cells using computer-aided molecular design approaches that can comprehensively consider compounds. An exploration cycle was developed that consists of the following five steps: setting conditions, computational evaluation, experimental evaluation, verification experiments, and discussions with experts in biotechnology. It aims to discover promising CPA candidate compounds by incorporating domain knowledge through discussions with the experts. The developed cycle can be applied to fields where the required physical properties have not been clearly known. As a result, 1-methylimidazole and pyridazine were selected as promising CPA candidate compounds, which were both heterocyclic amines. Hence, heterocyclic amines could be a stepping-stone toward the future development of CPAs for stem cells. By repeatedly using the exploration cycle, CPA candidate compounds with better cryoprotective effects could be discovered.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"16 ","pages":"Article 100248"},"PeriodicalIF":3.0,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.dche.2025.100244
Arun K. Sharma, Owen McMillan, Selsela Arsala, Supreet Gandhok, Rylend Young
Asphaltenes are complex polycyclic organic molecules in crude oil that readily aggregate and precipitate under varying thermodynamic conditions. Their structural heterogeneity influences key physicochemical properties, including solubility, stability, and reactivity. Molecular polarizability, a crucial property governing intermolecular interactions and electronic behavior, remains challenging to predict due to this structural diversity. This study employs machine learning models to predict isotropic polarizability using two sets of molecular descriptors: WHIM and GETAWAY. A dataset of 255 asphaltene structures was analyzed using stratified sampling, generating 10 independent training (80 %) and testing (20 %) splits. The Wolfram Language’s Predict function evaluated multiple machine learning algorithms—including Random Forest, Decision Tree, Gradient Boosted Trees, Nearest Neighbors, Linear Regression, Gaussian Process, and Neural Network—through an automated model selection process, serving as an AutoML framework. Linear regression was the best-performing model in 9 out of 10 splits for GETAWAY descriptors. GETAWAY-based models achieved an average mean absolute deviation of 0.0920 ± 0.0030 and standard deviation of 0.113 ± 0.004, significantly outperforming WHIM-based models (MAD = 0.173 ± 0.007, STD = 0.224 ± 0.008) with paired t-tests confirming statistical significance (p < 0.001). While R² values were reported, their interpretability was limited by heterogeneity and narrow property ranges in some test sets. These findings demonstrate the effectiveness of AutoML-guided approaches for predicting molecular properties and identify GETAWAY descriptors as a robust, efficient basis for polarizability prediction. Accurate prediction of polarizability is essential for modeling intermolecular forces and improving force field design in petroleum and materials chemistry, issues that are central to industrial and chemical applications.
{"title":"Machine learning for asphaltene polarizability: Evaluating molecular descriptors","authors":"Arun K. Sharma, Owen McMillan, Selsela Arsala, Supreet Gandhok, Rylend Young","doi":"10.1016/j.dche.2025.100244","DOIUrl":"10.1016/j.dche.2025.100244","url":null,"abstract":"<div><div>Asphaltenes are complex polycyclic organic molecules in crude oil that readily aggregate and precipitate under varying thermodynamic conditions. Their structural heterogeneity influences key physicochemical properties, including solubility, stability, and reactivity. Molecular polarizability, a crucial property governing intermolecular interactions and electronic behavior, remains challenging to predict due to this structural diversity. This study employs machine learning models to predict isotropic polarizability using two sets of molecular descriptors: WHIM and GETAWAY. A dataset of 255 asphaltene structures was analyzed using stratified sampling, generating 10 independent training (80 %) and testing (20 %) splits. The Wolfram Language’s Predict function evaluated multiple machine learning algorithms—including Random Forest, Decision Tree, Gradient Boosted Trees, Nearest Neighbors, Linear Regression, Gaussian Process, and Neural Network—through an automated model selection process, serving as an AutoML framework. Linear regression was the best-performing model in 9 out of 10 splits for GETAWAY descriptors. GETAWAY-based models achieved an average mean absolute deviation of 0.0920 ± 0.0030 and standard deviation of 0.113 ± 0.004, significantly outperforming WHIM-based models (MAD = 0.173 ± 0.007, STD = 0.224 ± 0.008) with paired <em>t</em>-tests confirming statistical significance (<em>p</em> < 0.001). While R² values were reported, their interpretability was limited by heterogeneity and narrow property ranges in some test sets. These findings demonstrate the effectiveness of AutoML-guided approaches for predicting molecular properties and identify GETAWAY descriptors as a robust, efficient basis for polarizability prediction. Accurate prediction of polarizability is essential for modeling intermolecular forces and improving force field design in petroleum and materials chemistry, issues that are central to industrial and chemical applications.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"15 ","pages":"Article 100244"},"PeriodicalIF":3.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144184928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.dche.2025.100239
Marcin Pietrasik , Anna Wilbik , Yannick Damoiseaux , Tessa Derks , Emery Karambiri , Shirley de Koster , Daniel van der Velde , Kim Ragaert , Sin Yong Teng
Plastic mechanical recycling is the conventional technological step towards circularity. In such aspects, complex mixtures of polyolefin blends are often fed into mechanical recycling systems, resulting in moulded products with uncertain quality. To add to the difficulty of heterogeneous feedstocks, the testing of mechanical properties for plastic products often results in stochastic measurements, making connections from material prediction to systems understanding challenging. This research is aimed at providing a framework capable of generalizing stochastic plastic recycling knowledge via interval-based machine learning for the prediction of properties formulation for unrecycled plastics. The framework is made up of two components: a regressor for point estimation and an interval predictor for generating prediction intervals. We compare several competing methods for each of these components through empirical evaluation on a real-world dataset. The results demonstrate the usefulness of interval-based machine learning in the application of stochastic engineering problems such as plastic mechanical recycling, highlighting such approaches towards better model interpretation and (un)certainty prediction regions.
{"title":"Capturing variability in material property predictions for plastics recycling via machine learning","authors":"Marcin Pietrasik , Anna Wilbik , Yannick Damoiseaux , Tessa Derks , Emery Karambiri , Shirley de Koster , Daniel van der Velde , Kim Ragaert , Sin Yong Teng","doi":"10.1016/j.dche.2025.100239","DOIUrl":"10.1016/j.dche.2025.100239","url":null,"abstract":"<div><div>Plastic mechanical recycling is the conventional technological step towards circularity. In such aspects, complex mixtures of polyolefin blends are often fed into mechanical recycling systems, resulting in moulded products with uncertain quality. To add to the difficulty of heterogeneous feedstocks, the testing of mechanical properties for plastic products often results in stochastic measurements, making connections from material prediction to systems understanding challenging. This research is aimed at providing a framework capable of generalizing stochastic plastic recycling knowledge via interval-based machine learning for the prediction of properties formulation for unrecycled plastics. The framework is made up of two components: a regressor for point estimation and an interval predictor for generating prediction intervals. We compare several competing methods for each of these components through empirical evaluation on a real-world dataset. The results demonstrate the usefulness of interval-based machine learning in the application of stochastic engineering problems such as plastic mechanical recycling, highlighting such approaches towards better model interpretation and (un)certainty prediction regions.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"15 ","pages":"Article 100239"},"PeriodicalIF":3.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144203735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.dche.2025.100245
Bol Ram, Z Ahmad, N Md Nor
Crude oil remains a vital non-renewable resource that supports numerous industries in the current era of industrial advancement. Consequently, petroleum refineries face increasing challenges, including stringent environmental regulations, fluctuating feedstock quality, rising demand, safety requirements, and the need for cost optimization. These challenges, coupled with the inherent complexity of the Crude Distillation Unit (CDU), demand advanced control strategies to ensure stable and efficient operation. This study investigates the application of Dynamic Matrix Control (DMC), a subset of Model Predictive Control (MPC), using Aspen DMC3 for CDU process control—a novel implementation not previously explored. The methodology involves three main stages: validation of a CDU simulation based on real data from the Basrah refinery, generation of dynamic response data through MATLAB integrated with Aspen Dynamics, and the development of a DMC controller using Aspen DMC3. The performance of the DMC controller is compared against conventional Proportional-Integral-Derivative (PID) controllers implemented in Aspen Dynamics using key indicators such as settling time, offset error, maximum deviation, and response smoothness. Results demonstrate that the DMC controller provides superior control performance, with faster settling times, zero offset, minimal deviations, and smoother responses. Additionally, Aspen DMC3′s AI-assisted capabilities enable streamlined controller configuration and real-time optimization through server connectivity, highlighting its potential for robust and efficient CDU operation.
{"title":"Utilization of aspen DMC3 in process control of crude distillation unit (CDU)","authors":"Bol Ram, Z Ahmad, N Md Nor","doi":"10.1016/j.dche.2025.100245","DOIUrl":"10.1016/j.dche.2025.100245","url":null,"abstract":"<div><div>Crude oil remains a vital non-renewable resource that supports numerous industries in the current era of industrial advancement. Consequently, petroleum refineries face increasing challenges, including stringent environmental regulations, fluctuating feedstock quality, rising demand, safety requirements, and the need for cost optimization. These challenges, coupled with the inherent complexity of the Crude Distillation Unit (CDU), demand advanced control strategies to ensure stable and efficient operation. This study investigates the application of Dynamic Matrix Control (DMC), a subset of Model Predictive Control (MPC), using Aspen DMC3 for CDU process control—a novel implementation not previously explored. The methodology involves three main stages: validation of a CDU simulation based on real data from the Basrah refinery, generation of dynamic response data through MATLAB integrated with Aspen Dynamics, and the development of a DMC controller using Aspen DMC3. The performance of the DMC controller is compared against conventional Proportional-Integral-Derivative (PID) controllers implemented in Aspen Dynamics using key indicators such as settling time, offset error, maximum deviation, and response smoothness. Results demonstrate that the DMC controller provides superior control performance, with faster settling times, zero offset, minimal deviations, and smoother responses. Additionally, Aspen DMC3′s AI-assisted capabilities enable streamlined controller configuration and real-time optimization through server connectivity, highlighting its potential for robust and efficient CDU operation.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"15 ","pages":"Article 100245"},"PeriodicalIF":3.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144223561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.dche.2025.100243
Qilin Qu , Linhui Wang , I.-Yen Wu , David Shan-Hill Wong , Ying Zheng , Yuan Yao
Predicting the Remaining Useful Life (RUL) of equipments has recently become a crucial technology for assessing operational safety and assisting maintenance decision-making. Numerous studies have demonstrated that a low-dimensional Health Indicator (HI) can be constructed from multidimensional sensor readings related to degradation, and the prediction of RUL can be based on similarities of HI. However, existing approaches for HI construction ignore neither the slow and monotonic nature of a degradation feature nor correlations between HI and RUL. To address this issue, this paper proposes a degradation-related slow feature analysis (DRSFA) method for extracting HIs and applying them in RUL prediction. Specifically, an objective function and its corresponding closed-form solution are proposed, aiming at extracting a health indicator from multidimensional degradation parameters to represent the slow degradation trend of an equipment and is correlated with its RUL. In DRSFA, HIs of each segment of lifecycle data is extracted separately rather than by a unified model, thereby enhancing its scalability as new data become available. As an HI extractor, DRSFA can serve as a plug-and-play module for RUL prediction based on similarity matching. Finally, experiments conducted on the CMAPSS dataset for aero-engine RUL assessment from NASA validate that the proposed method effectively balances RUL prediction accuracy, interpretability, and scalability.
{"title":"A degradation-related slow feature analysis for equipment health indicator extraction and remaining useful life prediction","authors":"Qilin Qu , Linhui Wang , I.-Yen Wu , David Shan-Hill Wong , Ying Zheng , Yuan Yao","doi":"10.1016/j.dche.2025.100243","DOIUrl":"10.1016/j.dche.2025.100243","url":null,"abstract":"<div><div>Predicting the Remaining Useful Life (RUL) of equipments has recently become a crucial technology for assessing operational safety and assisting maintenance decision-making. Numerous studies have demonstrated that a low-dimensional Health Indicator (HI) can be constructed from multidimensional sensor readings related to degradation, and the prediction of RUL can be based on similarities of HI. However, existing approaches for HI construction ignore neither the slow and monotonic nature of a degradation feature nor correlations between HI and RUL. To address this issue, this paper proposes a degradation-related slow feature analysis (DRSFA) method for extracting HIs and applying them in RUL prediction. Specifically, an objective function and its corresponding closed-form solution are proposed, aiming at extracting a health indicator from multidimensional degradation parameters to represent the slow degradation trend of an equipment and is correlated with its RUL. In DRSFA, HIs of each segment of lifecycle data is extracted separately rather than by a unified model, thereby enhancing its scalability as new data become available. As an HI extractor, DRSFA can serve as a plug-and-play module for RUL prediction based on similarity matching. Finally, experiments conducted on the CMAPSS dataset for aero-engine RUL assessment from NASA validate that the proposed method effectively balances RUL prediction accuracy, interpretability, and scalability.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"15 ","pages":"Article 100243"},"PeriodicalIF":3.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144211867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-30DOI: 10.1016/j.dche.2025.100246
Cheick Dosso , Hector A. Pedrozo , Thien Tran , Lingxiang Zhu , Victor Kusuma , David Hopkinson , Lorenz T. Biegler , Grigorios Panagakos
In this work, we study the application of membrane-based separation systems for carbon capture, considering plate-and-frame membrane modules. The successful deployment of membrane CO2 capture systems relies on high-performing membranes, as well as on effective membrane modules that can fully exploit the developed membranes. A plate-and-frame membrane module is especially attractive for CO2 capture from industrial flue gas, due to its reduced pressure drop compared to its counterparts such as spiral wound modules and hollow fiber modules. To design better plate-and-frame modules, we investigate their basic unit - a single membrane stack - through a combination of computational modeling and experimental investigations. The modeling approach is based on computational fluid dynamics (CFD) to represent the multiphysics problem, including the fluid flow and diffusion processes within the membrane stack. We use experimental data collected under different operating conditions to validate the CFD model. Numerical results suggest good agreement between experiments and model outputs for CO2 recovery, CO2 mole fractions in the retentate and permeate, and stage-cut. The CFD model is able to predict accurately the flow behavior, providing valuable insights into the effects of fluid dynamics on the mass transfer of CO2. CFD models achieve high accuracy by capturing complex permeate-side flow patterns exhibiting a 4.5 % maximum relative error compared to experiments. Results suggest that deviations of 1D models, assuming ideal flow patterns, from the CFD increase as separation properties improve with material advancements, and can be as high as 21 % for some cases. We also carry out a sensitivity analysis to identify the effect of key parameters on the CO2 recovery and the CO2 purity of the outlet streams.
{"title":"A computational investigation of high-flux, plate-and-frame membrane modules for industrial carbon capture","authors":"Cheick Dosso , Hector A. Pedrozo , Thien Tran , Lingxiang Zhu , Victor Kusuma , David Hopkinson , Lorenz T. Biegler , Grigorios Panagakos","doi":"10.1016/j.dche.2025.100246","DOIUrl":"10.1016/j.dche.2025.100246","url":null,"abstract":"<div><div>In this work, we study the application of membrane-based separation systems for carbon capture, considering plate-and-frame membrane modules. The successful deployment of membrane CO<sub>2</sub> capture systems relies on high-performing membranes, as well as on effective membrane modules that can fully exploit the developed membranes. A plate-and-frame membrane module is especially attractive for CO<sub>2</sub> capture from industrial flue gas, due to its reduced pressure drop compared to its counterparts such as spiral wound modules and hollow fiber modules. To design better plate-and-frame modules, we investigate their basic unit - a single membrane stack - through a combination of computational modeling and experimental investigations. The modeling approach is based on computational fluid dynamics (CFD) to represent the multiphysics problem, including the fluid flow and diffusion processes within the membrane stack. We use experimental data collected under different operating conditions to validate the CFD model. Numerical results suggest good agreement between experiments and model outputs for CO<sub>2</sub> recovery, CO<sub>2</sub> mole fractions in the retentate and permeate, and stage-cut. The CFD model is able to predict accurately the flow behavior, providing valuable insights into the effects of fluid dynamics on the mass transfer of CO<sub>2</sub>. CFD models achieve high accuracy by capturing complex permeate-side flow patterns exhibiting a 4.5 % maximum relative error compared to experiments. Results suggest that deviations of 1D models, assuming ideal flow patterns, from the CFD increase as separation properties improve with material advancements, and can be as high as 21 % for some cases. We also carry out a sensitivity analysis to identify the effect of key parameters on the CO<sub>2</sub> recovery and the CO<sub>2</sub> purity of the outlet streams.</div></div>","PeriodicalId":72815,"journal":{"name":"Digital Chemical Engineering","volume":"16 ","pages":"Article 100246"},"PeriodicalIF":3.0,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号