The fasteners employed in the railway tracks are susceptible to defects arising from their intricate composition.Foreign objects are frequently observed on the track bed in an open environment.These two types of defec...The fasteners employed in the railway tracks are susceptible to defects arising from their intricate composition.Foreign objects are frequently observed on the track bed in an open environment.These two types of defects pose potential threats to high-speed trains,thus necessitating timely and accurate track inspection.The majority of extant automatic inspection methods are predicated on the utilization of single visible light data,and the efficacy of the algorithmic processes is influenced by complex environments.Furthermore,due to the single information dimension,the detection accuracy of defects in similar,occluded,and small object categories is low.To address the aforementioned issues,this paper proposes a track defect detectionmethod based on dynamicmulti-modal fusion and challenging object enhanced perception.First,in light of the variances in the representation dimensions ofmultimodal information,this paper proposes a dynamic weighted multi-modal feature fusion module.The fused multi-modal features are assigned weights,and thenmultiplied with the extracted single-modal features atmultiple levels,achieving adaptive adjustment of the response degree of fusion features.Second,a novel stepwise multi-scale convolution feature aggregation module is proposed for challenging objects.The proposed method employs depth separable convolution and cross-scale aggregation operations of different receptive fields to enhance feature extraction and reuse,thereby reducing the degree of progressive loss of effective information.The experimental results demonstrate the efficacy of the proposed method in comparison to eight established methods,encompassing both single-modal and multi-modal methods,as evidenced by the extensive findings within the constructed RGBD dataset.展开更多
The Wulong lode deposit contains over 80 tonnes of gold with an average grade of 5.35 g.It is one of the largest deposits in Dandong City,Liaoning Province in northeast China.Previous studies on the deposit focused on...The Wulong lode deposit contains over 80 tonnes of gold with an average grade of 5.35 g.It is one of the largest deposits in Dandong City,Liaoning Province in northeast China.Previous studies on the deposit focused on its geological characteristics,geochemistry,fluid inclusions,and the timing of gold mineralization.However,controversy remains regarding the origin of the ore-forming fluids and metals,and the genesis of the gold deposit.This paper presents zircon U-Pb and pyrite Rb-Sr ages and S,Pb,He,and Ar isotopic results along with quartz H and O isotopic data for all litho-units associated with the deposit.Laser ablation inductively coupled mass spectrometry measurements yielded zircon U-Pb dates for samples of pre-mineralization rocks like granite porphyry dike,the Sanguliu granodiorite,fine-grained diorite,and syn-mineralization diorite,as well as post-mineralization dolerite,and lamprophyre;their emplacement ages are 126±1 Ma,124±1 Ma,123±1 Ma,120±1 Ma,119±2 Ma,and 115±2 Ma,respectively.The pyrite Rb-Sr isochron age is 119±1 Ma,indicating that both magmatism and mineralization occurred during the Early Cretaceous.The δ18OH2 O values of ore-forming hydrothermal fluids from the quartz-polymetallic sulfide vein stage vary from 4.8‰to 6.5‰,and the δDv-SMOW values are between-67.7‰and-75.9‰,indicating that the ore-forming fluids were primarily magmatic.The noble gas isotope compositions of fluid inclusions hosted in pyrite suggest that the ore-forming fluids were dominantly derived from crustal sources with minor mantle input.Sulfur isotopic values of pyrite vary between 0.2‰and 3.5‰,suggesting that S was derived from a homogeneous magmatic source or possibly from fluids derived from the crust.The Pb isotopic compositions of sulfides(207 Pb/204 Pb=15.51-15.71,206 Pb/204 Pb=17.35-18.75,208 Pb/204 Pb=38.27-40.03) indicate that the Pb of the Wulong gold deposit is a mixture of crust and mantle components.Geochronological and geochemical data,together with the regional geological history,indicate that Early Cretaceous magmatism and mineralization of the Wulong gold deposit occurred during the rollback of the subducting Paleo-Pacific Plate,which resulted in lithospheric thinning and the destruction of the North China Craton(NCC),which indicates that the deposit is of magmatic-hydrothermal origin.展开更多
Safety is essential when building a strong transportation system.As a key development direction in the global railway system,the intelligent railway has safety at its core,making safety a top priority while pursuing t...Safety is essential when building a strong transportation system.As a key development direction in the global railway system,the intelligent railway has safety at its core,making safety a top priority while pursuing the goals of efficiency,convenience,economy,and environmental friendliness.This paper describes the state of the art and proposes a system architecture for intelligent railway systems.It also focuses on the development of railway safety technology at home and abroad,and proposes the active safety method and technology system based on advanced theoretical methods such as the in-depth integration of cyber–physical systems(CPS),data-driven models,and intelligent computing.Finally,several typical applications are demonstrated to verify the advancement and feasibility of active safety technology in intelligent railway systems.展开更多
The large tonnage Maoling gold deposit(25 t@3.2 g)is located in the southwest Liaodong Peninsula,North China Craton.The deposit is hosted in the Paleoproterozoic metamorphic rocks.Four stages of mineralization were id...The large tonnage Maoling gold deposit(25 t@3.2 g)is located in the southwest Liaodong Peninsula,North China Craton.The deposit is hosted in the Paleoproterozoic metamorphic rocks.Four stages of mineralization were identified in the deposit:(stageⅠ)quartz-arsenopyrite±pyrite,(stageⅡ)quartz-goldarsenopyrite-pyrrhotite,(stageⅢ)quartz-gold-polymetallic sulfide,and(stageⅣ)quartz-calcitepyrrhotite.In this paper,we present fluid inclusion,C-H-O-S-Pb-He-Ar isotope data,zircon U-Pb,and gold-bearing sulfide(i.e.arsenopyrite and pyrrhotite)Rb-Sr age of the Maoling gold deposit to constrain its genesis and ore-forming mechanism.Three types of fluid inclusions were distinguished in quartzbearing veins,including liquid-rich two-phase(WL type),gas-rich two-phase(GL type),and daughter mineral-bearing fluid inclusions(S type).Fluid inclusions data show that the homogenization at temperatures 197 to 372°C for stageⅠ,126 to 319°C for stageⅡ,119 to 189°C for stageⅢ,and 115 to 183°C for stageⅣ,with corresponding salinities of 3.7 to 22.6 wt.%,4.7 to 23.2 wt.%,5.3 to 23.2 wt.%,and 1.7 to14.9 wt.%Na Cl equiv.,respectively.Fluid boiling was the critical factor controlling the gold and associated sulfide precipitation at Maoling.Hydrogen and oxygen stable isotopic analyses for quartz yielded δ18O=-5.0‰ to 9.8‰ and δD=-133.5‰ to-77.0‰.Carbon stable isotopic analyses for calcite and ankerite yielded δ13C=-2.3‰to-1.2‰ and O=7.9‰ to 14.1‰.The C-H-O isotope data show that the oreforming fluids were originated from magmatic water with meteoric water input during mineralization.Hydrothermal inclusions in arsenopyrite have 3He/4He ratios of 0.002 Ra to 0.054 Ra,and 40Ar/36Ar rations of 1225 to 3930,indicating that the ore-forming fluids were dominantly derived from crustal sources almost no mantle input.Sulfur isotopic values of Maoling fine-grained granite range from6.‰1 to 9.8‰,with a mean of 7.7‰,δ34S values of arsenopyrite from the mineralized phyllite(host rock)range from 8.9‰ to 10.6‰,with a mean of 10.0‰,by contrast,δ34S values of sulfides from ore vary between 4.3‰and 10.6‰,with a mean of 6.8‰,suggesting that sulfur was mainly originated from both the host rock and magma.Lead radioactive isotopic analyses for sulfides yielded206Pb/204Pb=15.830–17.103,207Pb/204Pb=13.397–15.548,208Pb/204Pb=35.478–36.683,and for Maoling fine-grained granite yielded 206Pb/204Pb=18.757–19.053,207Pb/204Pb=15.596–15.612,and 208Pb/204Pb=38.184–39.309,also suggesting that the ore-forming materials were mainly originated from the host rocks and magma.Zircon U-Pb dating demonstrates that the Maoling fine-grained granite was emplaced at 192.7±1.8 Ma,and the host rock(mineralized phyllite)was emplaced at some time after2065.0±27.0 Ma.Arsenopyrite and pyrrhotite give Rb–Sr isochron age of 188.7±4.5 Ma,indicating that both magmatism and mineralization occurred during the Early Jurassic.Geochronological and geochemical data,together with the regional geological history,indicate that Early Jurassic magmatism and mineralization of the Maoling gold deposit occurred during the subducting Paleo-Pacific Plate beneath Eurasia,and the Maoling gold deposit is of the intrusion-related gold deposit type.展开更多
The Erdaogou fault,also known as the 101 fault,comprises the most important NE strike component in the ore-controlling system in the Qingchengzi orefield,Liaodong Peninsula.Due to the poor outcrop conditions in the fi...The Erdaogou fault,also known as the 101 fault,comprises the most important NE strike component in the ore-controlling system in the Qingchengzi orefield,Liaodong Peninsula.Due to the poor outcrop conditions in the field,the Erdaogou fault lacks activity time constraint.We demonstrate the constraint activity time of the fault since we observe a lamprophyre that was cut through by the Erdaogou fault in the Taoyuan area,central to the Qingchengzi orefield.Zircon grains from the lamprophyre dyke exhibit typical oscillatory zoning and yield weighted mean U–Pb age of 223.8±1.1 Ma(MSWD=1.5).The lower activity time limit of the Erdaogou fault is thus first constrained as 224 Ma or so and is correlated with the crystallization age of the Xinling and Shuangdinggou plutons within the orefield.Taking previous mineralization studies into consideration,a Triassic tectonic-magmatism-mineralization model is approved in the Qingchengzi orefield.展开更多
Despite the big success of transfer learning techniques in anomaly detection,it is still challenging to achieve good transition of detection rules merely based on the preferred data in the anomaly detection with one-c...Despite the big success of transfer learning techniques in anomaly detection,it is still challenging to achieve good transition of detection rules merely based on the preferred data in the anomaly detection with one-class classification,especially for the data with a large distribution difference.To address this challenge,a novel deep one-class transfer learning algorithm with domain-adversarial training is proposed in this paper.First,by integrating a hypersphere adaptation constraint into domainadversarial neural network,a new hypersphere adversarial training mechanism is designed.Second,an alternative optimization method is derived to seek the optimal network parameters while pushing the hyperspheres built in the source domain and target domain to be as identical as possible.Through transferring oneclass detection rule in the adaptive extraction of domain-invariant feature representation,the end-to-end anomaly detection with one-class classification is then enhanced.Furthermore,a theoretical analysis about the model reliability,as well as the strategy of avoiding invalid and negative transfer,is provided.Experiments are conducted on two typical anomaly detection problems,i.e.,image recognition detection and online early fault detection of rolling bearings.The results demonstrate that the proposed algorithm outperforms the state-of-the-art methods in terms of detection accuracy and robustness.展开更多
基金funded by Beijing Natural Science Foundation,grant number L241078.
摘要The fasteners employed in the railway tracks are susceptible to defects arising from their intricate composition.Foreign objects are frequently observed on the track bed in an open environment.These two types of defects pose potential threats to high-speed trains,thus necessitating timely and accurate track inspection.The majority of extant automatic inspection methods are predicated on the utilization of single visible light data,and the efficacy of the algorithmic processes is influenced by complex environments.Furthermore,due to the single information dimension,the detection accuracy of defects in similar,occluded,and small object categories is low.To address the aforementioned issues,this paper proposes a track defect detectionmethod based on dynamicmulti-modal fusion and challenging object enhanced perception.First,in light of the variances in the representation dimensions ofmultimodal information,this paper proposes a dynamic weighted multi-modal feature fusion module.The fused multi-modal features are assigned weights,and thenmultiplied with the extracted single-modal features atmultiple levels,achieving adaptive adjustment of the response degree of fusion features.Second,a novel stepwise multi-scale convolution feature aggregation module is proposed for challenging objects.The proposed method employs depth separable convolution and cross-scale aggregation operations of different receptive fields to enhance feature extraction and reuse,thereby reducing the degree of progressive loss of effective information.The experimental results demonstrate the efficacy of the proposed method in comparison to eight established methods,encompassing both single-modal and multi-modal methods,as evidenced by the extensive findings within the constructed RGBD dataset.
基金supported by the National Key Research and Development Program (Grand Nos. 2018YFC0603804, 2016YFC0600108-09)the National Natural Science Foundation of China (41502093)the Geological Investigation Project of the China Geological Survey (DD20160049)
摘要The Wulong lode deposit contains over 80 tonnes of gold with an average grade of 5.35 g.It is one of the largest deposits in Dandong City,Liaoning Province in northeast China.Previous studies on the deposit focused on its geological characteristics,geochemistry,fluid inclusions,and the timing of gold mineralization.However,controversy remains regarding the origin of the ore-forming fluids and metals,and the genesis of the gold deposit.This paper presents zircon U-Pb and pyrite Rb-Sr ages and S,Pb,He,and Ar isotopic results along with quartz H and O isotopic data for all litho-units associated with the deposit.Laser ablation inductively coupled mass spectrometry measurements yielded zircon U-Pb dates for samples of pre-mineralization rocks like granite porphyry dike,the Sanguliu granodiorite,fine-grained diorite,and syn-mineralization diorite,as well as post-mineralization dolerite,and lamprophyre;their emplacement ages are 126±1 Ma,124±1 Ma,123±1 Ma,120±1 Ma,119±2 Ma,and 115±2 Ma,respectively.The pyrite Rb-Sr isochron age is 119±1 Ma,indicating that both magmatism and mineralization occurred during the Early Cretaceous.The δ18OH2 O values of ore-forming hydrothermal fluids from the quartz-polymetallic sulfide vein stage vary from 4.8‰to 6.5‰,and the δDv-SMOW values are between-67.7‰and-75.9‰,indicating that the ore-forming fluids were primarily magmatic.The noble gas isotope compositions of fluid inclusions hosted in pyrite suggest that the ore-forming fluids were dominantly derived from crustal sources with minor mantle input.Sulfur isotopic values of pyrite vary between 0.2‰and 3.5‰,suggesting that S was derived from a homogeneous magmatic source or possibly from fluids derived from the crust.The Pb isotopic compositions of sulfides(207 Pb/204 Pb=15.51-15.71,206 Pb/204 Pb=17.35-18.75,208 Pb/204 Pb=38.27-40.03) indicate that the Pb of the Wulong gold deposit is a mixture of crust and mantle components.Geochronological and geochemical data,together with the regional geological history,indicate that Early Cretaceous magmatism and mineralization of the Wulong gold deposit occurred during the rollback of the subducting Paleo-Pacific Plate,which resulted in lithospheric thinning and the destruction of the North China Craton(NCC),which indicates that the deposit is of magmatic-hydrothermal origin.
基金supported by the 2021 Chinese Academy of Engineering(CAE)International Top-level Forum on Engineering Science and Technology,“Safety and Governance of the High-Speed Railway”。
摘要Safety is essential when building a strong transportation system.As a key development direction in the global railway system,the intelligent railway has safety at its core,making safety a top priority while pursuing the goals of efficiency,convenience,economy,and environmental friendliness.This paper describes the state of the art and proposes a system architecture for intelligent railway systems.It also focuses on the development of railway safety technology at home and abroad,and proposes the active safety method and technology system based on advanced theoretical methods such as the in-depth integration of cyber–physical systems(CPS),data-driven models,and intelligent computing.Finally,several typical applications are demonstrated to verify the advancement and feasibility of active safety technology in intelligent railway systems.
基金funded by the NSF of China(No.42072104,41502093)the National Key R&D Plan of China(Grant Nos.2016YFC0600108,2018YFC0603804)。
摘要The large tonnage Maoling gold deposit(25 t@3.2 g)is located in the southwest Liaodong Peninsula,North China Craton.The deposit is hosted in the Paleoproterozoic metamorphic rocks.Four stages of mineralization were identified in the deposit:(stageⅠ)quartz-arsenopyrite±pyrite,(stageⅡ)quartz-goldarsenopyrite-pyrrhotite,(stageⅢ)quartz-gold-polymetallic sulfide,and(stageⅣ)quartz-calcitepyrrhotite.In this paper,we present fluid inclusion,C-H-O-S-Pb-He-Ar isotope data,zircon U-Pb,and gold-bearing sulfide(i.e.arsenopyrite and pyrrhotite)Rb-Sr age of the Maoling gold deposit to constrain its genesis and ore-forming mechanism.Three types of fluid inclusions were distinguished in quartzbearing veins,including liquid-rich two-phase(WL type),gas-rich two-phase(GL type),and daughter mineral-bearing fluid inclusions(S type).Fluid inclusions data show that the homogenization at temperatures 197 to 372°C for stageⅠ,126 to 319°C for stageⅡ,119 to 189°C for stageⅢ,and 115 to 183°C for stageⅣ,with corresponding salinities of 3.7 to 22.6 wt.%,4.7 to 23.2 wt.%,5.3 to 23.2 wt.%,and 1.7 to14.9 wt.%Na Cl equiv.,respectively.Fluid boiling was the critical factor controlling the gold and associated sulfide precipitation at Maoling.Hydrogen and oxygen stable isotopic analyses for quartz yielded δ18O=-5.0‰ to 9.8‰ and δD=-133.5‰ to-77.0‰.Carbon stable isotopic analyses for calcite and ankerite yielded δ13C=-2.3‰to-1.2‰ and O=7.9‰ to 14.1‰.The C-H-O isotope data show that the oreforming fluids were originated from magmatic water with meteoric water input during mineralization.Hydrothermal inclusions in arsenopyrite have 3He/4He ratios of 0.002 Ra to 0.054 Ra,and 40Ar/36Ar rations of 1225 to 3930,indicating that the ore-forming fluids were dominantly derived from crustal sources almost no mantle input.Sulfur isotopic values of Maoling fine-grained granite range from6.‰1 to 9.8‰,with a mean of 7.7‰,δ34S values of arsenopyrite from the mineralized phyllite(host rock)range from 8.9‰ to 10.6‰,with a mean of 10.0‰,by contrast,δ34S values of sulfides from ore vary between 4.3‰and 10.6‰,with a mean of 6.8‰,suggesting that sulfur was mainly originated from both the host rock and magma.Lead radioactive isotopic analyses for sulfides yielded206Pb/204Pb=15.830–17.103,207Pb/204Pb=13.397–15.548,208Pb/204Pb=35.478–36.683,and for Maoling fine-grained granite yielded 206Pb/204Pb=18.757–19.053,207Pb/204Pb=15.596–15.612,and 208Pb/204Pb=38.184–39.309,also suggesting that the ore-forming materials were mainly originated from the host rocks and magma.Zircon U-Pb dating demonstrates that the Maoling fine-grained granite was emplaced at 192.7±1.8 Ma,and the host rock(mineralized phyllite)was emplaced at some time after2065.0±27.0 Ma.Arsenopyrite and pyrrhotite give Rb–Sr isochron age of 188.7±4.5 Ma,indicating that both magmatism and mineralization occurred during the Early Jurassic.Geochronological and geochemical data,together with the regional geological history,indicate that Early Jurassic magmatism and mineralization of the Maoling gold deposit occurred during the subducting Paleo-Pacific Plate beneath Eurasia,and the Maoling gold deposit is of the intrusion-related gold deposit type.
基金financially supported by the National Key Research and Development Program of China (Grant No. 2018YFC0603804, 2016YFC0600108)Geological Survey Project of China (Grant No. DD20190156)the China Scholarship Council
摘要The Erdaogou fault,also known as the 101 fault,comprises the most important NE strike component in the ore-controlling system in the Qingchengzi orefield,Liaodong Peninsula.Due to the poor outcrop conditions in the field,the Erdaogou fault lacks activity time constraint.We demonstrate the constraint activity time of the fault since we observe a lamprophyre that was cut through by the Erdaogou fault in the Taoyuan area,central to the Qingchengzi orefield.Zircon grains from the lamprophyre dyke exhibit typical oscillatory zoning and yield weighted mean U–Pb age of 223.8±1.1 Ma(MSWD=1.5).The lower activity time limit of the Erdaogou fault is thus first constrained as 224 Ma or so and is correlated with the crystallization age of the Xinling and Shuangdinggou plutons within the orefield.Taking previous mineralization studies into consideration,a Triassic tectonic-magmatism-mineralization model is approved in the Qingchengzi orefield.
基金supported by the National Natural Science Foundation of China(NSFC)(U1704158)Henan Province Technologies Research and Development Project of China(212102210103)+1 种基金the NSFC Development Funding of Henan Normal University(2020PL09)the University of Manitoba Research Grants Program(URGP)。
摘要Despite the big success of transfer learning techniques in anomaly detection,it is still challenging to achieve good transition of detection rules merely based on the preferred data in the anomaly detection with one-class classification,especially for the data with a large distribution difference.To address this challenge,a novel deep one-class transfer learning algorithm with domain-adversarial training is proposed in this paper.First,by integrating a hypersphere adaptation constraint into domainadversarial neural network,a new hypersphere adversarial training mechanism is designed.Second,an alternative optimization method is derived to seek the optimal network parameters while pushing the hyperspheres built in the source domain and target domain to be as identical as possible.Through transferring oneclass detection rule in the adaptive extraction of domain-invariant feature representation,the end-to-end anomaly detection with one-class classification is then enhanced.Furthermore,a theoretical analysis about the model reliability,as well as the strategy of avoiding invalid and negative transfer,is provided.Experiments are conducted on two typical anomaly detection problems,i.e.,image recognition detection and online early fault detection of rolling bearings.The results demonstrate that the proposed algorithm outperforms the state-of-the-art methods in terms of detection accuracy and robustness.