In recent years,with the increasing attention to issues related to carbon emissions,such as carbon tariffs and government netzero carbon emission policies,carbon emissions have become an important indicator that is be...In recent years,with the increasing attention to issues related to carbon emissions,such as carbon tariffs and government netzero carbon emission policies,carbon emissions have become an important indicator that is being prioritized by governments worldwide.The Google Environmental Insights Explorer(EIE)tool has been developed to facilitate the collection and integration of data in this context.This study focuses on Tainan City and utilizes EIE to analyze greenhouse gas emissions from transportation.By using EIE,the study obtains data on greenhouse gas emissions from transportation activities in Tainan City.EIE utilizes data collected by Google and simulation functions to estimate data based on actual measurements of transportation activities.This tool saves time and resources by eliminating the need for on-site investigations while providing data that closely represent the real emissions from transportation activities in urban areas.Transportation vehicles contribute to greenhouse gas emissions in two ways:through direct combustion of fossil fuels and through the consumption of electricity in electric vehicles(EVs).The level of greenhouse gas emissions in a city’s transportation industry depends on factors such as transportation modes,fuel types,fleet age and energy efficiency,total distance traveled,and annual mileage.EIE estimates the greenhouse gas emissions from Tainan City’s transportation industry in 2022 to be 3,320,000 metric tons,including emissions from buses,motorcycles,cars,walking,railways,bicycles,and other modes of transportation.展开更多
【目的/意义】基于Open Site Explorer平台数据,分析和揭示网站层的我国图书馆共链关系及结构。【过程/方法】从该平台下载我国169所高校图书馆与公共图书馆网站的根域名入链数据,基于网站层的共链关系矩阵绘制共链网络,运用社会网络分...【目的/意义】基于Open Site Explorer平台数据,分析和揭示网站层的我国图书馆共链关系及结构。【过程/方法】从该平台下载我国169所高校图书馆与公共图书馆网站的根域名入链数据,基于网站层的共链关系矩阵绘制共链网络,运用社会网络分析方法对共链网络的节点中心性、块模型与核心-边缘结构进行分析。【结果/结论】该平台数据能够清晰地展现我国图书馆网站之间的共链关系结构,图书馆类型、实力及其所在地区的发达程度,对图书馆网站在共链关系网络中的网络特征及其网络空间分布有较大影响,但在网络中影响力最大、最重要的图书馆网站却是实力处于中上水平和中西部地区的图书馆。该平台既能为图书馆网站建设提供依据,也能拓展网络计量学理论与方法。展开更多
During the final of the International Standardization Youth Star Competition 2025,China Standardization interviewed several teams.In these young students,we see stories of their exploration,perseverance and dreams.The...During the final of the International Standardization Youth Star Competition 2025,China Standardization interviewed several teams.In these young students,we see stories of their exploration,perseverance and dreams.The competition has come to an end,but the real journey has just begun:How will this experience change their future pursuit?展开更多
The Neogene Shawan Formation in the Chepaizi Uplift of the Junggar Basin(NW China)has obtained high oil flow,demonstrating a good potential for oil and gas exploration.The multi-source hydrocarbon generation backgroun...The Neogene Shawan Formation in the Chepaizi Uplift of the Junggar Basin(NW China)has obtained high oil flow,demonstrating a good potential for oil and gas exploration.The multi-source hydrocarbon generation background and strong tectonic activity have led to the simultaneous production of heavy oil and light oil from multi-layer in the area,which makes it very difficult to identify oil origins,presently,the hot debate on the oil origins needs to be clarified.In this paper,due to the selective consumption of different types of compounds in crude oils by severe and intense biodegradation,the commonly used oilsource correlation tools are ineffective or may produce misleading results,this study adopted a biomarker recovery method based on the principle of mass conservation that uses the sum of the mass of the residual biomarkers and their corresponding biodegradation products to obtain the mass of the original biomarkers,improving the reliability of oil origins determination.Based on the nature and occurrence of crude oils,the investigated oils are subdivided into three types,Group A,Group B and Group C.Group A,light oils occurred mainly in lower structure Neogene Shawan Formation in the western Chepaizi Uplift,while Group B,heavy oils occurred mainly in higher structure Neogene Shawan Formation in the western Chepaizi Uplift.The two types of crude oils may come from the mixed source of Jurassic Badaowan Formation source rocks(J_(1)b)and Paleogene Anjihaihe Formation source rocks(E_(2-3)a)in the Sikeshu Sag,and Jurassic Badaowan Formation source rocks(J_(1)b)are the main source of crude oils.Group C,heavy oils occurred mainly in Neogene Shawan Formation in the eastern Chepaizi Uplift,showing good correlation with the Permian(P_(1)f and P_(2)w)source rocks in the Shawan Sag.At the same time,by combining stable carbon isotope and parameters related to triaromatic steroids,the accuracy of the oilsource correlation results by biomarker recovery method was further verified.展开更多
Standardizers keep making theoretical explorations and innovations to advance the standardization process in China.By the end of 2025,the China Standardization Press organized experts to conduct a rigorous review of a...Standardizers keep making theoretical explorations and innovations to advance the standardization process in China.By the end of 2025,the China Standardization Press organized experts to conduct a rigorous review of all articles published on China Standardization (Chinese version) and Standard Science in 2025.The top ten excellent papers were hereby showcased.展开更多
This study investigated the heterogeneous responses of organic matter(OM)in highly-to over-mature source rocks during thermal maturation.An integrated analysis was conducted on the Raman spectroscopic and geochemical ...This study investigated the heterogeneous responses of organic matter(OM)in highly-to over-mature source rocks during thermal maturation.An integrated analysis was conducted on the Raman spectroscopic and geochemical signatures of shales from the Lower Silurian Longmaxi Formation and the Lower Cambrian Qiongzhusi Formation,as well as anthracites from the Lower Permian Shanxi–Formation and the Upper Carboniferous Taiyuan Formation(collectively referred to as the Shanxi Taiyuan Formations).Additionally,burial and thermal evolution modeling was employed to support the analysis.A systematic assessment of Raman spectral parameters(e.g.,the positions and intensity ratio of the D and G bands)revealed robust correlations between the thermal history patterns of source rocks and molecular structural evolution parameters.The subsequent mechanistic quantification demonstrated that the maturation state of the source rocks was subjected to the hierarchical control of three principal factors:Peak heating temperature,the duration of sustained thermal intensity,and effective maturation duration.In addition,comparative analyses demonstrated that the anthracites attained higher structural ordering under sustained thermal conditions.This contrasts with the disordered carbon matrices observed in the intermittently heated shales.Raman spectroscopy further revealed broader variations in the D and G band intensities of the Longmaxi Formation compared to the Qiongzhusi Formation.This difference is associated with their different thermal histories.The thermal burial histories confirm that shales in the Longmaxi Formation underwent thermal exposure at lower peak temperatures over a shorter duration compared to those in the Qiongzhusi Formation.Finally,this study established a maturity calibration model for over-mature source rocks through a systematic correlation between Raman peak height ratios(R_(D/G))and vitrinite reflectance(R_(o)).展开更多
Epithermal deposits are characterized by complex low-temperature hydrothermal alterations, but the links between mineralization and superimposed alteration are obscure and require further elucidation. This study emplo...Epithermal deposits are characterized by complex low-temperature hydrothermal alterations, but the links between mineralization and superimposed alteration are obscure and require further elucidation. This study employs shortwave infrared(SWIR) spectral scalars for alteration mineral mapping and three-dimensional modeling of the Ulan Uzhur Ag polymetallic deposit, a newly discovered epithermal deposit in the Qimantagh. Alteration zoning transitions from illitemuscovite-carbonate-pyrite in the core(Zone Ⅰ), through muscovite ± illite-kaolinite-chlorite-carbonate(Zone Ⅱ), to muscovite-chlorite-biotite(Zone Ⅲ) at the periphery. The Zone Ⅰ with mineralization features long-wavelength white mica(wAlOH > 2207 nm) with a high Illite Crystallinity(IC)(mean > 2.0), suggesting a relatively high-temperature environment conducive to mineralization. Petrographic analyses with fluid inclusion and IC curve characteristics suggests that fluid boiling may be a pivotal mechanism for mineral precipitation. Furthermore, surface mapping and deep threedimensional modeling of spectral characteristics reveal a correlation between long-wavelength white mica, high IC and mineralization zones. These findings indicate that SWIR spectroscopy reveal the evolution of fluids and provide valuable guidance for future exploration efforts.展开更多
The Yuncheng Basin,located in the southern part of the Fenwei Rift,North China,exhibits obvious crust thinning(Moho uplift of 6-8 km)and shallow Curie point depth(less than 18 km)and hence holds great potential for ge...The Yuncheng Basin,located in the southern part of the Fenwei Rift,North China,exhibits obvious crust thinning(Moho uplift of 6-8 km)and shallow Curie point depth(less than 18 km)and hence holds great potential for geothermal resources.However,geothermal exploration within the Yuncheng Basin typically faces significant challenges due to civil and industrial noise from dense populations and industrial activities.To address these challenges,both Controlled-Source Audio-frequency Magnetotellurics(CSAMT)and radon measurements were employed in Baozigou village to investigate the geothermal structures and identify potential geothermal targets.The CSAMT method effectively delineated the structure of the subsurface hydrothermal system,identifying the reservoir as Paleogene sandstones and Ordovician and Cambrian limestones at elevations ranging from−800 m to−2500 m.In particular,two concealed normal faults(F_(a)and F_(b))were newly revealed by the combination of CSAMT and radon profiling;these previously undetected faults,which exhibit different scales and opposing dips,are likely to be responsible for controlling the convection of thermal water within the Basin’s subsurface hydrothermal system.Moreover,this study developed a preliminary conceptual geothermal model for the Fen River Depression within the Yuncheng Basin,which encompasses geothermal heat sources,cap rocks,reservoirs,and fluid pathways,providing valuable insights for future geothermal exploration.In conjunction with the 3D geological model constructed from CSAMT resistivity structures beneath Baozigou village,test drilling is recommended in the northwestern region of the Baozigou area to intersect the potentially deep fractured carbonates that may contain temperature-elevated geothermal water.This study establishes a good set of guidelines for future geothermal exploration in this region,indicating that high-permeability faults in the central segments of the Fen River Depression are promising targets.展开更多
The integration of Artificial Intelligence(AI)and Machine Learning(ML)into groundwater exploration and water resources management has emerged as a transformative approach to addressing global water challenges.This rev...The integration of Artificial Intelligence(AI)and Machine Learning(ML)into groundwater exploration and water resources management has emerged as a transformative approach to addressing global water challenges.This review explores key AI and ML concepts,methodologies,and their applications in hydrology,focusing on groundwater potential mapping,water quality prediction,and groundwater level forecasting.It discusses various data acquisition techniques,including remote sensing,geospatial analysis,and geophysical surveys,alongside preprocessing methods that are essential for enhancing model accuracy.The study highlights AI-driven solutions in water distribution,allocation optimization,and realtime resource management.Despite their advantages,the application of AI and ML in water sciences faces several challenges,including data scarcity,model reliability,and the integration of these tools with traditional water management systems.Ethical and regulatory concerns also demand careful consideration.The paper also outlines future research directions,emphasizing the need for improved data collection,interpretable models,real-time monitoring capabilities,and interdisciplinary collaboration.By leveraging AI and ML advancements,the water sector can enhance decision-making,optimize resource distribution,and support the development of sustainable water management strategies.展开更多
This paper introduces a novel nature-inspired metaheuristic algorithm called the Gekko japonicus algorithm.The algo-rithm draws inspiration mainly from the predation strategies and survival behaviors of the Gekko japo...This paper introduces a novel nature-inspired metaheuristic algorithm called the Gekko japonicus algorithm.The algo-rithm draws inspiration mainly from the predation strategies and survival behaviors of the Gekko japonicus.The math-ematical model is developed by simulating various biological behaviors of the Gekko japonicus,such as hybrid loco-motion patterns,directional olfactory guidance,implicit group advantage tendencies,and the tail autotomy mechanism.By integrating multi-stage mutual constraints and dynamically adjusting parameters,GJA maintains an optimal balance between global exploration and local exploitation,thereby effectively solving complex optimization problems.To assess the performance of GJA,comparative analyses were performed against fourteen state-of-the-art metaheuristic algorithms using the CEC2017 and CEC2022 benchmark test sets.Additionally,a Friedman test was performed on the experimen-tal results to assess the statistical significance of differences between various algorithms.And GJA was evaluated using multiple qualitative indicators,further confirming its superiority in exploration and exploitation.Finally,GJA was utilized to solve four engineering optimization problems and further implemented in robotic path planning to verify its practical applicability.Experimental results indicate that,compared to other high-performance algorithms,GJA demonstrates excep-tional performance as a powerful optimization algorithm in complex optimization problems.We make the code publicly available at:https://github.com/zhy1109/Gekko-japonicusalgorithm.展开更多
Robust cooperative unmanned aerial vehicle(UAV)formation in complex 3D environments is hampered by reward sparsity and inefficient collaboration.To address this,we propose context-aware relational agent learning(CORAL...Robust cooperative unmanned aerial vehicle(UAV)formation in complex 3D environments is hampered by reward sparsity and inefficient collaboration.To address this,we propose context-aware relational agent learning(CORAL),a novel multi-agent deep reinforcement learning framework.CORAL synergistically integrates two modules:(1)a novelty-based intrinsic reward module to drive efficient exploration and(2)an explicit relational learning module that allows agents to predict peer intentions and enhance coordination.Built on a multi-agent Actor-Critic architecture,CORAL enables agents to balance self-interest with group objectives.Comprehensive evaluations in a high-fidelity simulation show that our method significantly outperforms state-of-theart baselines like multi-agent deep deterministic policy gradient(MADDPG)and monotonic value function factorisation for deep multi-agent reinforcement learning(QMIX)in path planning efficiency,collision avoidance,and scalability.展开更多
Unmanned Aerial Vehicle(UAV)plays a prominent role in various fields,and autonomous navigation is a crucial component of UAV intelligence.Deep Reinforcement Learning(DRL)has expanded the research avenues for addressin...Unmanned Aerial Vehicle(UAV)plays a prominent role in various fields,and autonomous navigation is a crucial component of UAV intelligence.Deep Reinforcement Learning(DRL)has expanded the research avenues for addressing challenges in autonomous navigation.Nonetheless,challenges persist,including getting stuck in local optima,consuming excessive computations during action space exploration,and neglecting deterministic experience.This paper proposes a noise-driven enhancement strategy.In accordance with the overall learning phases,a global noise control method is designed,while a differentiated local noise control method is developed by analyzing the exploration demands of four typical situations encountered by UAV during navigation.Both methods are integrated into a dual-model for noise control to regulate action space exploration.Furthermore,noise dual experience replay buffers are designed to optimize the rational utilization of both deterministic and noisy experience.In uncertain environments,based on the Twin Delay Deep Deterministic Policy Gradient(TD3)algorithm with Long Short-Term Memory(LSTM)network and Priority Experience Replay(PER),a Noise-Driven Enhancement Priority Memory TD3(NDE-PMTD3)is developed.We established a simulation environment to compare different algorithms,and the performance of the algorithms is analyzed in various scenarios.The training results indicate that the proposed algorithm accelerates the convergence speed and enhances the convergence stability.In test experiments,the proposed algorithm successfully and efficiently performs autonomous navigation tasks in diverse environments,demonstrating superior generalization results.展开更多
文摘In recent years,with the increasing attention to issues related to carbon emissions,such as carbon tariffs and government netzero carbon emission policies,carbon emissions have become an important indicator that is being prioritized by governments worldwide.The Google Environmental Insights Explorer(EIE)tool has been developed to facilitate the collection and integration of data in this context.This study focuses on Tainan City and utilizes EIE to analyze greenhouse gas emissions from transportation.By using EIE,the study obtains data on greenhouse gas emissions from transportation activities in Tainan City.EIE utilizes data collected by Google and simulation functions to estimate data based on actual measurements of transportation activities.This tool saves time and resources by eliminating the need for on-site investigations while providing data that closely represent the real emissions from transportation activities in urban areas.Transportation vehicles contribute to greenhouse gas emissions in two ways:through direct combustion of fossil fuels and through the consumption of electricity in electric vehicles(EVs).The level of greenhouse gas emissions in a city’s transportation industry depends on factors such as transportation modes,fuel types,fleet age and energy efficiency,total distance traveled,and annual mileage.EIE estimates the greenhouse gas emissions from Tainan City’s transportation industry in 2022 to be 3,320,000 metric tons,including emissions from buses,motorcycles,cars,walking,railways,bicycles,and other modes of transportation.
文摘【目的/意义】基于Open Site Explorer平台数据,分析和揭示网站层的我国图书馆共链关系及结构。【过程/方法】从该平台下载我国169所高校图书馆与公共图书馆网站的根域名入链数据,基于网站层的共链关系矩阵绘制共链网络,运用社会网络分析方法对共链网络的节点中心性、块模型与核心-边缘结构进行分析。【结果/结论】该平台数据能够清晰地展现我国图书馆网站之间的共链关系结构,图书馆类型、实力及其所在地区的发达程度,对图书馆网站在共链关系网络中的网络特征及其网络空间分布有较大影响,但在网络中影响力最大、最重要的图书馆网站却是实力处于中上水平和中西部地区的图书馆。该平台既能为图书馆网站建设提供依据,也能拓展网络计量学理论与方法。
文摘During the final of the International Standardization Youth Star Competition 2025,China Standardization interviewed several teams.In these young students,we see stories of their exploration,perseverance and dreams.The competition has come to an end,but the real journey has just begun:How will this experience change their future pursuit?
基金co-funded by the National Natural Science Foundation of China(42372160,42072172)。
文摘The Neogene Shawan Formation in the Chepaizi Uplift of the Junggar Basin(NW China)has obtained high oil flow,demonstrating a good potential for oil and gas exploration.The multi-source hydrocarbon generation background and strong tectonic activity have led to the simultaneous production of heavy oil and light oil from multi-layer in the area,which makes it very difficult to identify oil origins,presently,the hot debate on the oil origins needs to be clarified.In this paper,due to the selective consumption of different types of compounds in crude oils by severe and intense biodegradation,the commonly used oilsource correlation tools are ineffective or may produce misleading results,this study adopted a biomarker recovery method based on the principle of mass conservation that uses the sum of the mass of the residual biomarkers and their corresponding biodegradation products to obtain the mass of the original biomarkers,improving the reliability of oil origins determination.Based on the nature and occurrence of crude oils,the investigated oils are subdivided into three types,Group A,Group B and Group C.Group A,light oils occurred mainly in lower structure Neogene Shawan Formation in the western Chepaizi Uplift,while Group B,heavy oils occurred mainly in higher structure Neogene Shawan Formation in the western Chepaizi Uplift.The two types of crude oils may come from the mixed source of Jurassic Badaowan Formation source rocks(J_(1)b)and Paleogene Anjihaihe Formation source rocks(E_(2-3)a)in the Sikeshu Sag,and Jurassic Badaowan Formation source rocks(J_(1)b)are the main source of crude oils.Group C,heavy oils occurred mainly in Neogene Shawan Formation in the eastern Chepaizi Uplift,showing good correlation with the Permian(P_(1)f and P_(2)w)source rocks in the Shawan Sag.At the same time,by combining stable carbon isotope and parameters related to triaromatic steroids,the accuracy of the oilsource correlation results by biomarker recovery method was further verified.
文摘Standardizers keep making theoretical explorations and innovations to advance the standardization process in China.By the end of 2025,the China Standardization Press organized experts to conduct a rigorous review of all articles published on China Standardization (Chinese version) and Standard Science in 2025.The top ten excellent papers were hereby showcased.
基金supported by the National Natural Science Foundation of China(42362022)the Open Fund of the Shaanxi Key Laboratory of Petroleum Accumulation Geology(PAG-202406)the Open Fund of the Mine Geology and Environment Academician and Expert Workstation(2024OITYSZJGZZ-005)。
文摘This study investigated the heterogeneous responses of organic matter(OM)in highly-to over-mature source rocks during thermal maturation.An integrated analysis was conducted on the Raman spectroscopic and geochemical signatures of shales from the Lower Silurian Longmaxi Formation and the Lower Cambrian Qiongzhusi Formation,as well as anthracites from the Lower Permian Shanxi–Formation and the Upper Carboniferous Taiyuan Formation(collectively referred to as the Shanxi Taiyuan Formations).Additionally,burial and thermal evolution modeling was employed to support the analysis.A systematic assessment of Raman spectral parameters(e.g.,the positions and intensity ratio of the D and G bands)revealed robust correlations between the thermal history patterns of source rocks and molecular structural evolution parameters.The subsequent mechanistic quantification demonstrated that the maturation state of the source rocks was subjected to the hierarchical control of three principal factors:Peak heating temperature,the duration of sustained thermal intensity,and effective maturation duration.In addition,comparative analyses demonstrated that the anthracites attained higher structural ordering under sustained thermal conditions.This contrasts with the disordered carbon matrices observed in the intermittently heated shales.Raman spectroscopy further revealed broader variations in the D and G band intensities of the Longmaxi Formation compared to the Qiongzhusi Formation.This difference is associated with their different thermal histories.The thermal burial histories confirm that shales in the Longmaxi Formation underwent thermal exposure at lower peak temperatures over a shorter duration compared to those in the Qiongzhusi Formation.Finally,this study established a maturity calibration model for over-mature source rocks through a systematic correlation between Raman peak height ratios(R_(D/G))and vitrinite reflectance(R_(o)).
基金supported by the Natural Science Foundation of China(Grant No.42372346,41802080,42030809,41873043).
文摘Epithermal deposits are characterized by complex low-temperature hydrothermal alterations, but the links between mineralization and superimposed alteration are obscure and require further elucidation. This study employs shortwave infrared(SWIR) spectral scalars for alteration mineral mapping and three-dimensional modeling of the Ulan Uzhur Ag polymetallic deposit, a newly discovered epithermal deposit in the Qimantagh. Alteration zoning transitions from illitemuscovite-carbonate-pyrite in the core(Zone Ⅰ), through muscovite ± illite-kaolinite-chlorite-carbonate(Zone Ⅱ), to muscovite-chlorite-biotite(Zone Ⅲ) at the periphery. The Zone Ⅰ with mineralization features long-wavelength white mica(wAlOH > 2207 nm) with a high Illite Crystallinity(IC)(mean > 2.0), suggesting a relatively high-temperature environment conducive to mineralization. Petrographic analyses with fluid inclusion and IC curve characteristics suggests that fluid boiling may be a pivotal mechanism for mineral precipitation. Furthermore, surface mapping and deep threedimensional modeling of spectral characteristics reveal a correlation between long-wavelength white mica, high IC and mineralization zones. These findings indicate that SWIR spectroscopy reveal the evolution of fluids and provide valuable guidance for future exploration efforts.
基金supported by the Shanxi Province Basic Research Program(No.20210302123374)Yuncheng University Doctoral Research Initiation Fund(No.YQ-2021008)+3 种基金Excellent doctors come to Shanxi to reward scientific research projects(No.QZX-2023020)Open Fund of State Key Laboratory of Precision Geodesy(No.SKLPG2025-1-1)Joint Open Fund of the Research Platforms of School of Computer Science,China University of Geosciences,Wuhan(No.PTLH2024-B-03)Hubei Provincial Natural Science Foundation Project(No.2025AFC095).
文摘The Yuncheng Basin,located in the southern part of the Fenwei Rift,North China,exhibits obvious crust thinning(Moho uplift of 6-8 km)and shallow Curie point depth(less than 18 km)and hence holds great potential for geothermal resources.However,geothermal exploration within the Yuncheng Basin typically faces significant challenges due to civil and industrial noise from dense populations and industrial activities.To address these challenges,both Controlled-Source Audio-frequency Magnetotellurics(CSAMT)and radon measurements were employed in Baozigou village to investigate the geothermal structures and identify potential geothermal targets.The CSAMT method effectively delineated the structure of the subsurface hydrothermal system,identifying the reservoir as Paleogene sandstones and Ordovician and Cambrian limestones at elevations ranging from−800 m to−2500 m.In particular,two concealed normal faults(F_(a)and F_(b))were newly revealed by the combination of CSAMT and radon profiling;these previously undetected faults,which exhibit different scales and opposing dips,are likely to be responsible for controlling the convection of thermal water within the Basin’s subsurface hydrothermal system.Moreover,this study developed a preliminary conceptual geothermal model for the Fen River Depression within the Yuncheng Basin,which encompasses geothermal heat sources,cap rocks,reservoirs,and fluid pathways,providing valuable insights for future geothermal exploration.In conjunction with the 3D geological model constructed from CSAMT resistivity structures beneath Baozigou village,test drilling is recommended in the northwestern region of the Baozigou area to intersect the potentially deep fractured carbonates that may contain temperature-elevated geothermal water.This study establishes a good set of guidelines for future geothermal exploration in this region,indicating that high-permeability faults in the central segments of the Fen River Depression are promising targets.
文摘The integration of Artificial Intelligence(AI)and Machine Learning(ML)into groundwater exploration and water resources management has emerged as a transformative approach to addressing global water challenges.This review explores key AI and ML concepts,methodologies,and their applications in hydrology,focusing on groundwater potential mapping,water quality prediction,and groundwater level forecasting.It discusses various data acquisition techniques,including remote sensing,geospatial analysis,and geophysical surveys,alongside preprocessing methods that are essential for enhancing model accuracy.The study highlights AI-driven solutions in water distribution,allocation optimization,and realtime resource management.Despite their advantages,the application of AI and ML in water sciences faces several challenges,including data scarcity,model reliability,and the integration of these tools with traditional water management systems.Ethical and regulatory concerns also demand careful consideration.The paper also outlines future research directions,emphasizing the need for improved data collection,interpretable models,real-time monitoring capabilities,and interdisciplinary collaboration.By leveraging AI and ML advancements,the water sector can enhance decision-making,optimize resource distribution,and support the development of sustainable water management strategies.
基金CHINA POSTDOCTORAL SCIENCE FOUNDATION(Grant No.2025M771925)Young Scientists Fund(C Class)(Grant No.32501636)Special Fund of Fundamental Scientific Research Business Expense for Higher School of Central Government(Grant No.2572025JT04).
文摘This paper introduces a novel nature-inspired metaheuristic algorithm called the Gekko japonicus algorithm.The algo-rithm draws inspiration mainly from the predation strategies and survival behaviors of the Gekko japonicus.The math-ematical model is developed by simulating various biological behaviors of the Gekko japonicus,such as hybrid loco-motion patterns,directional olfactory guidance,implicit group advantage tendencies,and the tail autotomy mechanism.By integrating multi-stage mutual constraints and dynamically adjusting parameters,GJA maintains an optimal balance between global exploration and local exploitation,thereby effectively solving complex optimization problems.To assess the performance of GJA,comparative analyses were performed against fourteen state-of-the-art metaheuristic algorithms using the CEC2017 and CEC2022 benchmark test sets.Additionally,a Friedman test was performed on the experimen-tal results to assess the statistical significance of differences between various algorithms.And GJA was evaluated using multiple qualitative indicators,further confirming its superiority in exploration and exploitation.Finally,GJA was utilized to solve four engineering optimization problems and further implemented in robotic path planning to verify its practical applicability.Experimental results indicate that,compared to other high-performance algorithms,GJA demonstrates excep-tional performance as a powerful optimization algorithm in complex optimization problems.We make the code publicly available at:https://github.com/zhy1109/Gekko-japonicusalgorithm.
基金supported by the STI 2030 Major Projects(No.2022ZD0208804)the National Natural Science Foundation of China(No.62473017)。
文摘Robust cooperative unmanned aerial vehicle(UAV)formation in complex 3D environments is hampered by reward sparsity and inefficient collaboration.To address this,we propose context-aware relational agent learning(CORAL),a novel multi-agent deep reinforcement learning framework.CORAL synergistically integrates two modules:(1)a novelty-based intrinsic reward module to drive efficient exploration and(2)an explicit relational learning module that allows agents to predict peer intentions and enhance coordination.Built on a multi-agent Actor-Critic architecture,CORAL enables agents to balance self-interest with group objectives.Comprehensive evaluations in a high-fidelity simulation show that our method significantly outperforms state-of-theart baselines like multi-agent deep deterministic policy gradient(MADDPG)and monotonic value function factorisation for deep multi-agent reinforcement learning(QMIX)in path planning efficiency,collision avoidance,and scalability.
基金the Collaborative Innovation Project of Shanghai,China for the financial support。
文摘Unmanned Aerial Vehicle(UAV)plays a prominent role in various fields,and autonomous navigation is a crucial component of UAV intelligence.Deep Reinforcement Learning(DRL)has expanded the research avenues for addressing challenges in autonomous navigation.Nonetheless,challenges persist,including getting stuck in local optima,consuming excessive computations during action space exploration,and neglecting deterministic experience.This paper proposes a noise-driven enhancement strategy.In accordance with the overall learning phases,a global noise control method is designed,while a differentiated local noise control method is developed by analyzing the exploration demands of four typical situations encountered by UAV during navigation.Both methods are integrated into a dual-model for noise control to regulate action space exploration.Furthermore,noise dual experience replay buffers are designed to optimize the rational utilization of both deterministic and noisy experience.In uncertain environments,based on the Twin Delay Deep Deterministic Policy Gradient(TD3)algorithm with Long Short-Term Memory(LSTM)network and Priority Experience Replay(PER),a Noise-Driven Enhancement Priority Memory TD3(NDE-PMTD3)is developed.We established a simulation environment to compare different algorithms,and the performance of the algorithms is analyzed in various scenarios.The training results indicate that the proposed algorithm accelerates the convergence speed and enhances the convergence stability.In test experiments,the proposed algorithm successfully and efficiently performs autonomous navigation tasks in diverse environments,demonstrating superior generalization results.
