Pub Date : 2026-03-15Epub Date: 2026-02-03DOI: 10.1016/j.psep.2026.108550
Osarieme Uyi Osazuwa , Kim Hoong Ng , Dai-Viet N. Vo , Yoke Wang Cheng
Dry reforming of methane (DRM) converts CH4 and CO2 into valuable syngas, offering environmental benefits and a practical route to exploit carbon capture streams. However, its endothermic nature and coking severely limit scalability. To overcome this, photocatalytic dry reforming of methane (PDRM) harnesses photon energy, boosting activity and stability under milder conditions. In pursuit of effective PDRM catalysts, a wide range of catalyst designs have been explored, such as metal‑on‑mixed‑oxide, metal‑on‑oxide, metal‑on‑mixed‑support, metal‑on‑oxide heterojunctions, encapsulated‑metal/zeolite (core‑shell) structures, engineered support structures, multi‑metallic alloys on 2‑D heterostructures, perovskites, metal‑on‑perovskite, 2D‑2D carbon‑based hybrids, metal‑free porous organic polymers, morphology‑controlled oxide supports, and bimetallic catalysts on carbon‑oxide composites. Building on this design diversity, case studies show impressive results. For instance, a Rh#CeO2 nanocomposite with an intertwined ≈ 5 nm Rh‑CeO2 network and a 2.7 eV bandgap achieved >60 % CH4 conversion, remained stable for >100 h, and showed no side reactions under UV light. Further reviews uncovered a range of pathways such as oxygen-vacancy mediation, charge-separated redox, dual-site charge-separated redox, Z‑scheme charge transfer, Schottky-barrier assistance, dual-site push‑pull charge transfer, hot‑carrier/near-field enhancement, carbonate-mediated bifunctional route, and HCOO-intermediate pathway highlighting how these mechanisms collectively enable efficient, sustainable syngas production under mild conditions and paving the way for further PDRM advances.
甲烷干重整(DRM)将CH4和CO2转化为有价值的合成气,提供了环境效益和开发碳捕获流的实用途径。然而,它的吸热性质和结焦严重限制了可扩展性。为了克服这个问题,光催化甲烷干重整(PDRM)利用光子能量,在较温和的条件下提高活性和稳定性。为了追求有效的PDRM催化剂,已经探索了各种各样的催化剂设计,例如金属- on -混合氧化物、金属- on -氧化物、金属- on -混合载体、金属- on -氧化物异质结、封装金属/沸石(核壳)结构、工程支撑结构、2D异质结构上的多金属合金、钙钛矿、金属- on -钙钛矿、2D - 2D碳基杂化物、无金属多孔有机聚合物、形态控制的氧化物载体、以及碳氧化物复合材料上的双金属催化剂。基于这种设计多样性,案例研究显示了令人印象深刻的结果。例如,具有缠绕≈ 5 nm的Rh - CeO2网络和2.7 eV带隙的Rh#CeO2纳米复合材料实现了>;60 %的CH4转化率,在>;100 h内保持稳定,并且在紫外光下没有副反应。进一步的综述揭示了一系列的途径,如氧空位介导、电荷分离氧化还原、双位点电荷分离氧化还原、Z - scheme电荷转移、schottkey势垒辅助、双位点推拉电荷转移、热载流子/近场增强、碳化物介导的双功能途径和hco -中间体途径,强调了这些机制如何共同实现在温和条件下高效、可持续的合成气生产,并为PDRM的进一步发展铺平了道路。
{"title":"Catalyst innovations and mechanism in photocatalytic dry reforming of methane: Recent advances and perspectives","authors":"Osarieme Uyi Osazuwa , Kim Hoong Ng , Dai-Viet N. Vo , Yoke Wang Cheng","doi":"10.1016/j.psep.2026.108550","DOIUrl":"10.1016/j.psep.2026.108550","url":null,"abstract":"<div><div>Dry reforming of methane (DRM) converts CH<sub>4</sub> and CO<sub>2</sub> into valuable syngas, offering environmental benefits and a practical route to exploit carbon capture streams. However, its endothermic nature and coking severely limit scalability. To overcome this, photocatalytic dry reforming of methane (PDRM) harnesses photon energy, boosting activity and stability under milder conditions. In pursuit of effective PDRM catalysts, a wide range of catalyst designs have been explored, such as metal‑on‑mixed‑oxide, metal‑on‑oxide, metal‑on‑mixed‑support, metal‑on‑oxide heterojunctions, encapsulated‑metal/zeolite (core‑shell) structures, engineered support structures, multi‑metallic alloys on 2‑D heterostructures, perovskites, metal‑on‑perovskite, 2D‑2D carbon‑based hybrids, metal‑free porous organic polymers, morphology‑controlled oxide supports, and bimetallic catalysts on carbon‑oxide composites. Building on this design diversity, case studies show impressive results. For instance, a Rh#CeO<sub>2</sub> nanocomposite with an intertwined ≈ 5 nm Rh‑CeO<sub>2</sub> network and a 2.7 eV bandgap achieved >60 % CH<sub>4</sub> conversion, remained stable for >100 h, and showed no side reactions under UV light. Further reviews uncovered a range of pathways such as oxygen-vacancy mediation, charge-separated redox, dual-site charge-separated redox, Z‑scheme charge transfer, Schottky-barrier assistance, dual-site push‑pull charge transfer, hot‑carrier/near-field enhancement, carbonate-mediated bifunctional route, and HCOO-intermediate pathway highlighting how these mechanisms collectively enable efficient, sustainable syngas production under mild conditions and paving the way for further PDRM advances.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108550"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-03DOI: 10.1016/j.psep.2026.108530
Yan Zhao , Liu Yang , Siyuan Li , Mingxiu Ji , Ruipeng Dong
Global groundwater systems face escalating organic contamination crises, threatening ecological security and human health. Groundwater circulation well (GCW) technology, an innovative in-situ remediation approach, uses hydraulic circulation to mobilize contaminants and synergizes with chemical oxidation for enhanced degradation. However, its practical application faces critical challenges, including a restricted remediation radius of influence, uneven oxidant distribution, and persistent contamination residues caused by preferential flow bypassing low-permeability zones. Theoretical evidence suggests that ultrasound technology could overcome these limitations through cavitation and mechanical vibrations, resulting in the restructuring of pore structure. While proven effective in enhancing the permeability of consolidated rock formations, the synergistic mechanisms between ultrasound and GCW systems, particularly in sandy aquifers, remain systematically unverified. This study systematically elucidates the mechanisms underlying ultrasound-enhanced GCW remediation through an integrated multi-scale experimental framework: (1) Microstructural characterization through NMR-coupled column experiments quantifying porosity and hydraulic conductivity enhancement; (2) Mesoscale transport analysis employing 2D sandbox simulations to investigate the expansion mechanisms of remediation radius under ultrasound-GCW coupled operation; and (3) Synergistic investigation through an innovative ultrasound-GCW-chemical oxidation system evaluating degradation efficiency and interaction mechanisms. This study confirms the permeability enhancement effect of ultrasonic stimulation in unconsolidated aquifers and demonstrates the improved contaminant remediation effectiveness of the GCW-ultrasound coupled system. These findings not only significantly expand the engineering application prospects of GCW-coupled remediation technology but also provide theoretical and technical support for the effective treatment of organic pollution in low-permeability zone aquifers.
{"title":"Improved remediation mechanism of groundwater circulation wells based on ultrasonic permeability enhancement","authors":"Yan Zhao , Liu Yang , Siyuan Li , Mingxiu Ji , Ruipeng Dong","doi":"10.1016/j.psep.2026.108530","DOIUrl":"10.1016/j.psep.2026.108530","url":null,"abstract":"<div><div>Global groundwater systems face escalating organic contamination crises, threatening ecological security and human health. Groundwater circulation well (GCW) technology, an innovative in-situ remediation approach, uses hydraulic circulation to mobilize contaminants and synergizes with chemical oxidation for enhanced degradation. However, its practical application faces critical challenges, including a restricted remediation radius of influence, uneven oxidant distribution, and persistent contamination residues caused by preferential flow bypassing low-permeability zones. Theoretical evidence suggests that ultrasound technology could overcome these limitations through cavitation and mechanical vibrations, resulting in the restructuring of pore structure. While proven effective in enhancing the permeability of consolidated rock formations, the synergistic mechanisms between ultrasound and GCW systems, particularly in sandy aquifers, remain systematically unverified. This study systematically elucidates the mechanisms underlying ultrasound-enhanced GCW remediation through an integrated multi-scale experimental framework: (1) Microstructural characterization through NMR-coupled column experiments quantifying porosity and hydraulic conductivity enhancement; (2) Mesoscale transport analysis employing 2D sandbox simulations to investigate the expansion mechanisms of remediation radius under ultrasound-GCW coupled operation; and (3) Synergistic investigation through an innovative ultrasound-GCW-chemical oxidation system evaluating degradation efficiency and interaction mechanisms. This study confirms the permeability enhancement effect of ultrasonic stimulation in unconsolidated aquifers and demonstrates the improved contaminant remediation effectiveness of the GCW-ultrasound coupled system. These findings not only significantly expand the engineering application prospects of GCW-coupled remediation technology but also provide theoretical and technical support for the effective treatment of organic pollution in low-permeability zone aquifers.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108530"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109769","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-08DOI: 10.1016/j.psep.2026.108556
Van-Canh Nguyen , Ngoc-Linh Pham , The-Anh Cao , Nhat-Tan Nguyen , Nguyen Anh Thang , Nhu-Trang Le , Thuy-Duong Nguyen
This study proposes a hybrid data-driven framework for multi-objective optimization in finish milling of P20 tool steel molds, with focus on both process safety and environmental protection. The proposed framework combines Gaussian Process Regression (GPR) surrogate modeling, resampling-based data augmentation, and Bayesian multi-objective optimization to simultaneously minimize surface roughness (Ra, Rz) and specific cutting energy (SCE). The resampling-based data augmentation expanded the original experimental dataset (n = 16) to about six times, which significantly improve the accuracy and robustness of surrogate models. As a result, the GPR models achieved high prediction performance with R² values of 0.8680 for Ra, 0.9211 for Rz, and 0.9888 for SCE, while the corresponding MAPE values were 3.25 %, 4.45 %, and 14.73 %, respectively. In addition, Random Forest regression combined with SHAP analysis showed that cutting speed (Vc) is the most influential parameter for Ra prediction (43.4 % importance), whereas depth of cut (ap, 30.4 %) and width of cut (ae, 33.6 %) mainly control SCE, which provide useful guidance for parameter selection. Bayesian multi-objective optimization identified Pareto-optimal cutting conditions (Vc = 40 m/min, fz = 0.07–0.13 mm/tooth, ap = 0.5–2.0 mm, ae = 5.1–8.5 mm) that achieved fine surface quality (Ra = 0.55–0.65 µm) while reducing SCE by up to 92.4 % compared to baseline conditions. Experimental validation confirmed good predictive accuracy, with mean absolute errors below 5 % for surface roughness and about 7 % for energy consumption. For a typical P20 mold cavity with 500 cm³ material removal, the optimized parameters can save 0.099 kWh energy and reduce 0.056 kg CO₂ per part, leading to significant annual saving for industrial production. Process safety analysis also indicated that the optimized conditions maintain spindle load below 5 % of rated capacity, increase tool safety factor above 2.0–5.0, and reduce thermal load by 85–92 %, therefore reducing risks of tool failure, machine damage, and fire hazard. Overall, this study provides a practical and data-efficient optimization approach for sustainable and safe mold manufacturing.
{"title":"Data-driven multi-objective optimization for process-safe and sustainable finish milling of P20 tool steel","authors":"Van-Canh Nguyen , Ngoc-Linh Pham , The-Anh Cao , Nhat-Tan Nguyen , Nguyen Anh Thang , Nhu-Trang Le , Thuy-Duong Nguyen","doi":"10.1016/j.psep.2026.108556","DOIUrl":"10.1016/j.psep.2026.108556","url":null,"abstract":"<div><div>This study proposes a hybrid data-driven framework for multi-objective optimization in finish milling of P20 tool steel molds, with focus on both process safety and environmental protection. The proposed framework combines Gaussian Process Regression (GPR) surrogate modeling, resampling-based data augmentation, and Bayesian multi-objective optimization to simultaneously minimize surface roughness (R<sub>a</sub>, R<sub>z</sub>) and specific cutting energy (SCE). The resampling-based data augmentation expanded the original experimental dataset (n = 16) to about six times, which significantly improve the accuracy and robustness of surrogate models. As a result, the GPR models achieved high prediction performance with R² values of 0.8680 for R<sub>a</sub>, 0.9211 for R<sub>z</sub>, and 0.9888 for SCE, while the corresponding MAPE values were 3.25 %, 4.45 %, and 14.73 %, respectively. In addition, Random Forest regression combined with SHAP analysis showed that cutting speed (V<sub>c</sub>) is the most influential parameter for R<sub>a</sub> prediction (43.4 % importance), whereas depth of cut (a<sub>p</sub>, 30.4 %) and width of cut (a<sub>e</sub>, 33.6 %) mainly control SCE, which provide useful guidance for parameter selection. Bayesian multi-objective optimization identified Pareto-optimal cutting conditions (V<sub>c</sub> = 40 m/min, f<sub>z</sub> = 0.07–0.13 mm/tooth, a<sub>p</sub> = 0.5–2.0 mm, a<sub>e</sub> = 5.1–8.5 mm) that achieved fine surface quality (R<sub>a</sub> = 0.55–0.65 µm) while reducing SCE by up to 92.4 % compared to baseline conditions. Experimental validation confirmed good predictive accuracy, with mean absolute errors below 5 % for surface roughness and about 7 % for energy consumption. For a typical P20 mold cavity with 500 cm³ material removal, the optimized parameters can save 0.099 kWh energy and reduce 0.056 kg CO₂ per part, leading to significant annual saving for industrial production. Process safety analysis also indicated that the optimized conditions maintain spindle load below 5 % of rated capacity, increase tool safety factor above 2.0–5.0, and reduce thermal load by 85–92 %, therefore reducing risks of tool failure, machine damage, and fire hazard. Overall, this study provides a practical and data-efficient optimization approach for sustainable and safe mold manufacturing.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108556"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146138309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-12DOI: 10.1016/j.psep.2026.108603
Liang Ma, Yifei Peng, Kaixiang Peng
Propagation path identification is an important part of fault diagnosis. It is often used to identify the propagation paths and locate the root causes of faults, which provides information supports for safety assurance and operating maintenance of industrial processes. Nonlinear causalities are common in dynamic industrial processes due to the strong couplings among subsystems and physical properties. When a fault occurs suddenly, its impact is often propagated with delay along causalities, resulting in lags of abnormal responses for related subsystems or control loops. Meanwhile, in high-dimensional industrial processes, traditional methods are prone to the problems of poor efficiency and accuracy, thus compromising process safety. Inspired by those problems, in this paper, a new cloud–edge–device collaboration based propagation path identification framework is proposed for faults in nonlinear dynamic industrial processes. Firstly, the multi-order lag encoder based graph convolutional network is proposed to extract the lag causality features of variables, and thus realizing nonlinear causality analysis by the spatial–temporal information. Secondly, the time-varying dynamic Bayesian network is constructed to identify the propagation paths and predict the future propagation directions of faults by combining the above algorithm and Bayesian estimation. Then, static Bayesian networks of edges and time-varying dynamic Bayesian network of cloud are constructed by the cloud–edge–device collaborative framework for causality analysis of high-dimensional time series and efficiency improvement of propagation path identification. Finally, three datasets from hot rolling process and Tennessee Eastman process are used to verify the accuracy and efficiency of the proposed framework.
{"title":"A cloud–edge–device collaboration based propagation path identification framework for faults in nonlinear dynamic industrial processes","authors":"Liang Ma, Yifei Peng, Kaixiang Peng","doi":"10.1016/j.psep.2026.108603","DOIUrl":"10.1016/j.psep.2026.108603","url":null,"abstract":"<div><div>Propagation path identification is an important part of fault diagnosis. It is often used to identify the propagation paths and locate the root causes of faults, which provides information supports for safety assurance and operating maintenance of industrial processes. Nonlinear causalities are common in dynamic industrial processes due to the strong couplings among subsystems and physical properties. When a fault occurs suddenly, its impact is often propagated with delay along causalities, resulting in lags of abnormal responses for related subsystems or control loops. Meanwhile, in high-dimensional industrial processes, traditional methods are prone to the problems of poor efficiency and accuracy, thus compromising process safety. Inspired by those problems, in this paper, a new cloud–edge–device collaboration based propagation path identification framework is proposed for faults in nonlinear dynamic industrial processes. Firstly, the multi-order lag encoder based graph convolutional network is proposed to extract the lag causality features of variables, and thus realizing nonlinear causality analysis by the spatial–temporal information. Secondly, the time-varying dynamic Bayesian network is constructed to identify the propagation paths and predict the future propagation directions of faults by combining the above algorithm and Bayesian estimation. Then, static Bayesian networks of edges and time-varying dynamic Bayesian network of cloud are constructed by the cloud–edge–device collaborative framework for causality analysis of high-dimensional time series and efficiency improvement of propagation path identification. Finally, three datasets from hot rolling process and Tennessee Eastman process are used to verify the accuracy and efficiency of the proposed framework.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108603"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-17DOI: 10.1016/j.psep.2026.108628
Hao Wang , Chaoya Guo , Hao Wang , Kun Liu , Tianjiao Bi , Chunjie Sui , Bin Zhang , Feng Zhu , Jieyu Jiang
Cryo-compressed hydrogen (CcH2) storage technology is a promising method for storing hydrogen on a large scale, but once accidental leak occurs, it will cause significant risks such as fire, explosion and secondary disasters. This study investigates the safety risks of CcH2 leakage in hydrogen refueling stations, emphasizing the unique safety implications of cryogenic hydrogen including cryogenic injuries, fire and explosion risks, and other thermal hazards. Numerical simulations are carried out considering different storage temperatures from 70 K to 300 K and pressures from 40 MPa to 80 MPa. Results show that a larger flammable cloud is formed in the case of CcH2 leakage, and the cryogenic injury area (CIA) below 233 K near blast wall is prolonged, exacerbating frostbite risks. This is attributed to the decreased thermal kinetic energy and increased density of CcH2. In the event of explosion, the duration time of high temperature near the blast wall increases with decreasing hydrogen storage temperature. This indicates dual risks of frostbite and high temperature by combustion in CcH2 systems. Additionally, a lower explosion overpressure peak (0.4475 bar at 70 K) is generated and multiple pressure peaks are observed near blast wall. It is worth noting that the pressure peak of a cryogenic hydrogen explosion is only 0.02 times that at normal temperature, which is below the lethal pressure, significantly reducing lethal injury risks. Based on the consequences and risk analysis, the corresponding protective measures are recommended, such as installing exhaust fans on blast wall to accelerate cryogenic hydrogen dispersion and reduce explosion risks.
{"title":"A safety analysis of the cryo-compressed effect on hydrogen leakage behavior in hydrogen refueling station","authors":"Hao Wang , Chaoya Guo , Hao Wang , Kun Liu , Tianjiao Bi , Chunjie Sui , Bin Zhang , Feng Zhu , Jieyu Jiang","doi":"10.1016/j.psep.2026.108628","DOIUrl":"10.1016/j.psep.2026.108628","url":null,"abstract":"<div><div>Cryo-compressed hydrogen (CcH<sub>2</sub>) storage technology is a promising method for storing hydrogen on a large scale, but once accidental leak occurs, it will cause significant risks such as fire, explosion and secondary disasters. This study investigates the safety risks of CcH<sub>2</sub> leakage in hydrogen refueling stations, emphasizing the unique safety implications of cryogenic hydrogen including cryogenic injuries, fire and explosion risks, and other thermal hazards. Numerical simulations are carried out considering different storage temperatures from 70 K to 300 K and pressures from 40 MPa to 80 MPa. Results show that a larger flammable cloud is formed in the case of CcH<sub>2</sub> leakage, and the cryogenic injury area (CIA) below 233 K near blast wall is prolonged, exacerbating frostbite risks. This is attributed to the decreased thermal kinetic energy and increased density of CcH<sub>2</sub>. In the event of explosion, the duration time of high temperature near the blast wall increases with decreasing hydrogen storage temperature. This indicates dual risks of frostbite and high temperature by combustion in CcH<sub>2</sub> systems. Additionally, a lower explosion overpressure peak (0.4475 bar at 70 K) is generated and multiple pressure peaks are observed near blast wall. It is worth noting that the pressure peak of a cryogenic hydrogen explosion is only 0.02 times that at normal temperature, which is below the lethal pressure, significantly reducing lethal injury risks. Based on the consequences and risk analysis, the corresponding protective measures are recommended, such as installing exhaust fans on blast wall to accelerate cryogenic hydrogen dispersion and reduce explosion risks.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108628"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-18DOI: 10.1016/j.psep.2026.108617
Wai Cheong Tam , MD. Ismail Siddiqi Emon , Jian Chen , Hongqiang Fang , Jun Deng , Anthony Putorti Jr
The paper presents the development of a multi-class classification model for detecting early-stage thermal runaway events in button-top single-cell lithium-ion batteries. A signal gate mechanism is introduced to extract relevant acoustic data. A data alignment technique is applied to enhance the model training. A multi-layer, two-dimensional convolutional neural network is employed to learn the key features that distinguish non-thermal runaway events from thermal runaway events. Results show that the proposed model can detect thermal runaway events with an overall accuracy, precision, and recall of 99.8 %, 98.1 %, and 98.3 %, respectively. Sensitivity studies are conducted and the results indicate that the data alignment and data augmentation techniques help to enhance the model performance significantly. The findings presented in this paper aim to contribute to the development of a practical and accurate early-stage thermal runaway detection model, which can enhance process safety and risk engineering strategies by providing earlier warnings before thermal runaway occurs.
{"title":"On the use of machine learning for the development of an acoustic-based detection system for early-stage thermal runaway","authors":"Wai Cheong Tam , MD. Ismail Siddiqi Emon , Jian Chen , Hongqiang Fang , Jun Deng , Anthony Putorti Jr","doi":"10.1016/j.psep.2026.108617","DOIUrl":"10.1016/j.psep.2026.108617","url":null,"abstract":"<div><div>The paper presents the development of a multi-class classification model for detecting early-stage thermal runaway events in button-top single-cell lithium-ion batteries. A signal gate mechanism is introduced to extract relevant acoustic data. A data alignment technique is applied to enhance the model training. A multi-layer, two-dimensional convolutional neural network is employed to learn the key features that distinguish non-thermal runaway events from thermal runaway events. Results show that the proposed model can detect thermal runaway events with an overall accuracy, precision, and recall of 99.8 %, 98.1 %, and 98.3 %, respectively. Sensitivity studies are conducted and the results indicate that the data alignment and data augmentation techniques help to enhance the model performance significantly. The findings presented in this paper aim to contribute to the development of a practical and accurate early-stage thermal runaway detection model, which can enhance process safety and risk engineering strategies by providing earlier warnings before thermal runaway occurs.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108617"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146777596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-10DOI: 10.1016/j.psep.2026.108589
Mitchell Huffman , Chi-Yang Li , Jazmine Aiya D. Marquez , Zihao Wang , Bryant Hendrickson , Thomas Butts , Filippo Gavelli , Qingsheng Wang
Passive Fire Protection (PFP) plays a critical role in minimizing the risks of liquefied natural gas (LNG) spills and fire hazards in processing and storage facilities. One common PFP strategy for LNG structures is the use of fire-resistant coatings. In this study, two one-hour LNG pool fire tests were conducted in a 10-ft by 10-ft pit to evaluate the thermal response of PFP-coated structural beams. The beams were coated with three different PFP systems, including cementitious, foam, and epoxy. The objective was to generate data for evaluating thermal response models and to compare the thermal performance of beams coated on all surfaces (as typically done in standardized testing) versus beams coated only on three sides, leaving the top flange uncoated (as commonly seen in installed equipment). During the tests, the PFP-coated beams were exposed to direct flame impingement from an LNG pool fire for one hour. Results showed that the 3-sided coating configuration led to higher maximum steel temperatures compared to the 4-sided coating due to the absence of thermal resistance on the exposed top flange. When compared to industry-standard thresholds for permissible heating, local temperature readings from the 3-sided coating were found to exceed the limits; however, cross-sectional average temperatures remained within those limits in two out of three cases. These findings have important implications for the insulation of structural beams in the LNG industry, emphasizing the need for comprehensive coating strategies to ensure structural integrity in pool fire scenarios.
{"title":"Evaluation of passive fire protection installation methods: Full-scale LNG fire testing of 3-sided and 4-sided coated structure beams","authors":"Mitchell Huffman , Chi-Yang Li , Jazmine Aiya D. Marquez , Zihao Wang , Bryant Hendrickson , Thomas Butts , Filippo Gavelli , Qingsheng Wang","doi":"10.1016/j.psep.2026.108589","DOIUrl":"10.1016/j.psep.2026.108589","url":null,"abstract":"<div><div>Passive Fire Protection (PFP) plays a critical role in minimizing the risks of liquefied natural gas (LNG) spills and fire hazards in processing and storage facilities. One common PFP strategy for LNG structures is the use of fire-resistant coatings. In this study, two one-hour LNG pool fire tests were conducted in a 10-ft by 10-ft pit to evaluate the thermal response of PFP-coated structural beams. The beams were coated with three different PFP systems, including cementitious, foam, and epoxy. The objective was to generate data for evaluating thermal response models and to compare the thermal performance of beams coated on all surfaces (as typically done in standardized testing) versus beams coated only on three sides, leaving the top flange uncoated (as commonly seen in installed equipment). During the tests, the PFP-coated beams were exposed to direct flame impingement from an LNG pool fire for one hour. Results showed that the 3-sided coating configuration led to higher maximum steel temperatures compared to the 4-sided coating due to the absence of thermal resistance on the exposed top flange. When compared to industry-standard thresholds for permissible heating, local temperature readings from the 3-sided coating were found to exceed the limits; however, cross-sectional average temperatures remained within those limits in two out of three cases. These findings have important implications for the insulation of structural beams in the LNG industry, emphasizing the need for comprehensive coating strategies to ensure structural integrity in pool fire scenarios.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108589"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146153010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-13DOI: 10.1016/j.psep.2026.108595
LIU Fuyao , QIU Fuguo , TIAN Hongyu , LIU Jianwei
This study addressed the issue of performance degradation in toilet wastewater treatment systems under low-temperature conditions by constructing a bio-enhanced anaerobic-aerobic membrane bioreactor (AO-MBR) system.Through inoculation of low-temperature-adapted microbial communities, the system's treatment performance, sludge characteristics, and microbial community response mechanisms were systematically evaluated under gradient temperature reduction (16℃→8℃). The bioaugmented reactor (S1) achieved COD, NH4+ -N, and TN removal efficiencies of 85.5 %, 87.6 %, and 67.8 %, respectively, which were higher than those of the control reactor (S2: 76.6 %, 75.1 %, and 54.8 %). Low-temperature microbial enhancement effectively improved sludge settleability, reduced the sludge volume index (SVI), and enhanced system resilience to low-temperature loads by promoting the secretion of proteins and polysaccharides in extracellular polymeric substances (EPS). Microbial community analysis revealed enrichment of cold-tolerant functional genera such as Acinetobacter and Flavobacterium in S1. Metabolic function prediction indicated significantly increased abundances of genes related to carbohydrate metabolism, nitrogen metabolism, and stress response. These findings elucidate the synergistic metabolic mechanisms and ecological stability of the bioaugmented AO-MBR system. The results provide theoretical and practical support for the engineering application of microbial enhancement technology in low-temperature toilet wastewater treatment.
{"title":"Metabolic basis and ecological stability of bioaugmented AO-MBR treating low-temperature toilet wastewater: Linking functional genes, pathway efficiency, and microbial dynamics","authors":"LIU Fuyao , QIU Fuguo , TIAN Hongyu , LIU Jianwei","doi":"10.1016/j.psep.2026.108595","DOIUrl":"10.1016/j.psep.2026.108595","url":null,"abstract":"<div><div>This study addressed the issue of performance degradation in toilet wastewater treatment systems under low-temperature conditions by constructing a bio-enhanced anaerobic-aerobic membrane bioreactor (AO-MBR) system.Through inoculation of low-temperature-adapted microbial communities, the system's treatment performance, sludge characteristics, and microbial community response mechanisms were systematically evaluated under gradient temperature reduction (16℃→8℃). The bioaugmented reactor (S1) achieved COD, NH<sub>4</sub><sup>+</sup> -N, and TN removal efficiencies of 85.5 %, 87.6 %, and 67.8 %, respectively, which were higher than those of the control reactor (S2: 76.6 %, 75.1 %, and 54.8 %). Low-temperature microbial enhancement effectively improved sludge settleability, reduced the sludge volume index (SVI), and enhanced system resilience to low-temperature loads by promoting the secretion of proteins and polysaccharides in extracellular polymeric substances (EPS). Microbial community analysis revealed enrichment of cold-tolerant functional genera such as Acinetobacter and Flavobacterium in S1. Metabolic function prediction indicated significantly increased abundances of genes related to carbohydrate metabolism, nitrogen metabolism, and stress response. These findings elucidate the synergistic metabolic mechanisms and ecological stability of the bioaugmented AO-MBR system. The results provide theoretical and practical support for the engineering application of microbial enhancement technology in low-temperature toilet wastewater treatment.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108595"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-12DOI: 10.1016/j.psep.2026.108576
Jihui Zhao , Jie Liu , Hongjin Zhong , Lu Luo , Huali He , Haijiao Xie
Steel slag (SS), a major solid waste from the iron and steel industry, possesses a chemical composition similar to Portland cement and exhibits latent hydraulic activity. This study investigated the effects of triethanolamine (TEA) and triisopropanolamine (TIPA), used as complexing agents, on SS dissolution and the performance of SS-cement based materials in a simulated cement pore solution (Ca(OH)2 and 0.2 mol/L NaOH). Results demonstrated that a) TEA exhibited stronger complexation capabilities towards Ca2 + and Fe3+, b) TIPA accelerated the precipitation of hydration products, notably promoting the formation of ettringite. The fundamental difference in Ca2+ complexation ability between TEA and TIPA stems from the influence of their spatial configurations. In the SS-cement system, both complexing agents significantly enhanced the early-age mechanical properties. However, TIPA had a more pronounced effect on later-age strength development. This was attributed to the unique effects by TIPA in improving and optimizing SS-PC pastes pore structure. Furthermore, both agents not only promoted the hydration of mineral phases within both the SS and cement but also facilitated the pozzolanic reaction between the amorphous phase of SS and CH.
{"title":"Dissolution-complexation of steel slag containing alcoholic amine compounds and its hydration properties in composite cement","authors":"Jihui Zhao , Jie Liu , Hongjin Zhong , Lu Luo , Huali He , Haijiao Xie","doi":"10.1016/j.psep.2026.108576","DOIUrl":"10.1016/j.psep.2026.108576","url":null,"abstract":"<div><div>Steel slag (SS), a major solid waste from the iron and steel industry, possesses a chemical composition similar to Portland cement and exhibits latent hydraulic activity. This study investigated the effects of triethanolamine (TEA) and triisopropanolamine (TIPA), used as complexing agents, on SS dissolution and the performance of SS-cement based materials in a simulated cement pore solution (Ca(OH)<sub>2</sub> and 0.2 mol/L NaOH). Results demonstrated that a) TEA exhibited stronger complexation capabilities towards Ca<sup>2 +</sup> and Fe<sup>3+</sup>, b) TIPA accelerated the precipitation of hydration products, notably promoting the formation of ettringite. The fundamental difference in Ca<sup>2+</sup> complexation ability between TEA and TIPA stems from the influence of their spatial configurations. In the SS-cement system, both complexing agents significantly enhanced the early-age mechanical properties. However, TIPA had a more pronounced effect on later-age strength development. This was attributed to the unique effects by TIPA in improving and optimizing SS-PC pastes pore structure. Furthermore, both agents not only promoted the hydration of mineral phases within both the SS and cement but also facilitated the pozzolanic reaction between the amorphous phase of SS and CH.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108576"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-15Epub Date: 2026-02-12DOI: 10.1016/j.psep.2026.108599
Zhengbo Li , Dingtao Peng
The brewing industry annually generates 38.6 million tons of solid waste, with 85 % comprising brewery residues, creating significant sustainability challenges. Under pressure to achieve carbon neutrality, wastewater treatment facilities must enhance efficiency while operating with low carbon emissions. Traditional optimization methods cannot effectively handle the high-dimensional nonlinear problems involving 32 coupled variables across multiple scales in brewery wastewater systems. This study proposes a tensor network-driven framework that decomposes the 32-dimensional decision space into six factor matrices with 5–6 dimensions each, enabling parallel computation and multiscale optimization at molecular, process, and system levels. Six-month industrial validation at a 200,000-ton/year brewery demonstrated substantial improvements: energy consumption decreased from 0.90 to 0.56 kWh/m3 (37.8 % reduction), COD removal increased from 85.2 % to 94.6 %, and carbon emissions dropped from 0.65 to 0.13 kg CO2-eq/m3 (80 % reduction). Economic analysis revealed a 14-month payback period and $2.62 million five-year NPV. The quantum-inspired approach provides a replicable solution for intelligent transformation and carbon-neutral transition in wastewater treatment, advancing circular economy implementation in the brewing sector.
酿酒业每年产生3860万吨固体废物,其中85% 为啤酒残留物,这对可持续发展构成了重大挑战。在实现碳中和的压力下,污水处理设施必须在低碳排放的同时提高效率。传统的优化方法不能有效地处理啤酒废水系统中涉及32个耦合变量的多尺度高维非线性问题。本研究提出了一个张量网络驱动的框架,该框架将32维决策空间分解为6个因子矩阵,每个因子矩阵有5-6个维度,可以在分子、过程和系统层面进行并行计算和多尺度优化。在一家年产20万吨的啤酒厂进行的为期6个月的工业验证显示出了实质性的改善:能耗从0.90千瓦时/立方米下降到0.56千瓦时/立方米(减少37.8% %),COD去除率从85.2% %增加到94.6 %,碳排放量从0.65 kg co2当量/立方米下降到0.13 kg co2当量/立方米(减少80% %)。经济分析显示,投资回收期为14个月,5年净现值为262万美元。量子启发的方法为废水处理的智能转换和碳中和过渡提供了可复制的解决方案,推动了酿造行业循环经济的实施。
{"title":"Tensor network-driven pathway optimization for brewery waste recycling: A multi-scale decision framework towards carbon neutrality","authors":"Zhengbo Li , Dingtao Peng","doi":"10.1016/j.psep.2026.108599","DOIUrl":"10.1016/j.psep.2026.108599","url":null,"abstract":"<div><div>The brewing industry annually generates 38.6 million tons of solid waste, with 85 % comprising brewery residues, creating significant sustainability challenges. Under pressure to achieve carbon neutrality, wastewater treatment facilities must enhance efficiency while operating with low carbon emissions. Traditional optimization methods cannot effectively handle the high-dimensional nonlinear problems involving 32 coupled variables across multiple scales in brewery wastewater systems. This study proposes a tensor network-driven framework that decomposes the 32-dimensional decision space into six factor matrices with 5–6 dimensions each, enabling parallel computation and multiscale optimization at molecular, process, and system levels. Six-month industrial validation at a 200,000-ton/year brewery demonstrated substantial improvements: energy consumption decreased from 0.90 to 0.56 kWh/m<sup>3</sup> (37.8 % reduction), COD removal increased from 85.2 % to 94.6 %, and carbon emissions dropped from 0.65 to 0.13 kg CO<sub>2</sub>-eq/m<sup>3</sup> (80 % reduction). Economic analysis revealed a 14-month payback period and $2.62 million five-year NPV. The quantum-inspired approach provides a replicable solution for intelligent transformation and carbon-neutral transition in wastewater treatment, advancing circular economy implementation in the brewing sector.</div></div>","PeriodicalId":20743,"journal":{"name":"Process Safety and Environmental Protection","volume":"209 ","pages":"Article 108599"},"PeriodicalIF":7.8,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146209517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}