近三年,以大语言模型(large language models,LLMs)为代表的人工智能技术和产业得到快速发展,实际应用也取得显著效果。为应对国家和行业发展的需求,国内许多高校纷纷设置人工智能学院、学科和专业,但与计算机学院、学科及专业的边界仍...近三年,以大语言模型(large language models,LLMs)为代表的人工智能技术和产业得到快速发展,实际应用也取得显著效果。为应对国家和行业发展的需求,国内许多高校纷纷设置人工智能学院、学科和专业,但与计算机学院、学科及专业的边界仍不够明晰。为此,本文描述了人工智能的含义、组成,并试图厘清与计算机学科及专业的关系,提出人工智能学院与计算机学院学科、专业发展的一些建议,期望为后续相关研究提供参考。展开更多
The demand for sensors capable of operating in extreme environment of the fields,such as aerospace vehicles,aeroengines and fire protection,is rapidly increasing.However,developing flexible ceramic fibrous pressure se...The demand for sensors capable of operating in extreme environment of the fields,such as aerospace vehicles,aeroengines and fire protection,is rapidly increasing.However,developing flexible ceramic fibrous pressure sensors that combine high temperature stability with robust mechanical properties remains a significant challenge.Herein,through precise multi-scale process control,high-strength(2.1 MPa)TiC-SiC flexible fibrous membrane is successfully fabricated.The membrane exhibits exceptional thermal resistance(2000℃)and long–term thermal stability(1800℃ for 5 h)in the inert atmosphere.Meanwhile,the TiC-SiC fibrous membrane shows excellent oxidation resistance and still achieves strength of 1.8 MPa after being oxidized at 1200℃ for 1 h in air.Remarkably,TiC-SiC fibrous membrane withstands a load of approximately 1400 times its own weight and the ablation of butane flame(~1300℃)for at least 1 h without breaking.Notably,after heat treatment at 1800℃ for 5 h in an argon atmosphere,the TiC-SiC fibrous membrane even sustains pressure–sensing performance for up to 300 cycles.The membrane exhibits stable resistivity up to 900℃ and shows sensing stability under butane flame.The results of this work provide an effective and feasible solution to fill the research gap of flexible fibrous sensors for extreme environments.展开更多
As joint operations have become a key trend in modern military development,unmanned aerial vehicles(UAVs)play an increasingly important role in enhancing the intelligence and responsiveness of combat systems.However,t...As joint operations have become a key trend in modern military development,unmanned aerial vehicles(UAVs)play an increasingly important role in enhancing the intelligence and responsiveness of combat systems.However,the heterogeneity of aircraft,partial observability,and dynamic uncertainty in operational airspace pose significant challenges to autonomous collision avoidance using traditional methods.To address these issues,this paper proposes an adaptive collision avoidance approach for UAVs based on deep reinforcement learning.First,a unified uncertainty model incorporating dynamic wind fields is constructed to capture the complexity of joint operational environments.Then,to effectively handle the heterogeneity between manned and unmanned aircraft and the limitations of dynamic observations,a sector-based partial observation mechanism is designed.A Dynamic Threat Prioritization Assessment algorithm is also proposed to evaluate potential collision threats from multiple dimensions,including time to closest approach,minimum separation distance,and aircraft type.Furthermore,a Hierarchical Prioritized Experience Replay(HPER)mechanism is introduced,which classifies experience samples into high,medium,and low priority levels to preferentially sample critical experiences,thereby improving learning efficiency and accelerating policy convergence.Simulation results show that the proposed HPER-D3QN algorithm outperforms existing methods in terms of learning speed,environmental adaptability,and robustness,significantly enhancing collision avoidance performance and convergence rate.Finally,transfer experiments on a high-fidelity battlefield airspace simulation platform validate the proposed method's deployment potential and practical applicability in complex,real-world joint operational scenarios.展开更多
Nucleic acid-based therapies have emerged as promising strategies for the regulation of gene expression and the production of therapeutic antigens or proteins for a series of diseases, including cancers, rare diseases...Nucleic acid-based therapies have emerged as promising strategies for the regulation of gene expression and the production of therapeutic antigens or proteins for a series of diseases, including cancers, rare diseases, and infectious diseases. However, their clinical application faces challenges. These include high molecular weight, limited cellular uptake,and susceptibility to enzymatic degradation by nucleases in vivo. Both viral and non-viral delivery vectors have been developed as a means of addressing these limitations, including lipid nanoparticles(LNPs), exosomes, polymers, and inorganic nanoparticles. Among these,LNPs have garnered significant attention due to their superior biocompatibility, high delivery efficiency and customizable design potential, as demonstrated by the clinical success of the FDA-approved si RNA drug Onpattro®. The critical role of nucleic acid drug carriers is discussed in this review. It also outlines the major types of carriers under development and examines the advancements and applications in LNP-based systems for nucleic acid delivery. By conducting a review of recent advancements in LNP design, delivery mechanisms, and clinical applications, this article aims to clarify the ways in which LNPs overcome delivery barriers, compare LNPs with other carriers, and identify key trends that can inform the development of next-generation LNP platforms for nucleic acid therapeutics.展开更多
To understand the smoke level and NO_(x)emission characteristics of in-use construction machinery in Beijing,we selected 905 construction machines in Beijing from August 2022 to April 2023 to monitor the emission leve...To understand the smoke level and NO_(x)emission characteristics of in-use construction machinery in Beijing,we selected 905 construction machines in Beijing from August 2022 to April 2023 to monitor the emission level of smoke and NO_(x).The exhaust smoke level and excessive emission situation of different machinery types were identified,and their NO_(x)emission levels were monitored according to the free acceleration method.We investigated the correlation of NO_(x)and smoke emission,and proposed suggestions for controlling pollution discharge from construction machinery in the future.The results show that the exhaust smoke level was 0–2.62 m^(−1),followed a log-normal distribution(μ=-1.73,δ=1.09,R^(2)=0.99),with a 5.64%exceedance rate.Differenceswere observed amongmachinery types,with low-power engine forklifts showing higher smoke levels.The NO_(x)emission range was 71–1516 ppm,followed a normal distribution(μ=565.54,δ=309.51,R^(2)=0.83).Differences among machinery types were relatively small.Engine rated net power had the most significant impact on NO_(x)emissions.Thus,NO_(x)emissions from construction machinery need further attention.Furthermore,we found a weak negative correlation(p<0.05)between the emission level of smoke and NO_(x),that is the synergic emission reduction effect is poor,emphasizing the need for NO_(x)emission limits.In the future,the oversight in Beijing should prioritize phasing out ChinaⅠand ChinaⅡmachinery,and monitor emissions from highpower engine ChinaⅢmachinery.展开更多
Ionic Liquid Electrospray Thrusters(ILETs)are well suited for micro-nano satellite applications due to their small size,low power consumption,and high specific impulse.However,the limited thrust of a single-emitter IL...Ionic Liquid Electrospray Thrusters(ILETs)are well suited for micro-nano satellite applications due to their small size,low power consumption,and high specific impulse.However,the limited thrust of a single-emitter ILET restricts its use in space missions.To optimize the performance of ILETs and make them suitable for a wider range of space missions,we designed a Circular-emitter ILET(CILET)to convert a one-dimensional(point)emission into a twodimensional(line)emission.The CILET can self-organize multiple Taylor cones simultaneously.The cones were photographed and the axial emission currents were measured under different voltage and pressure difference conditions with a CILET experimental system.The emission can be divided into two stable states and one unstable state based on the flow and current characteristics.The current in Stable state Ⅰ increases non-linearly with the voltage,while that in Stable state Ⅱ is nearly linear with respect to the voltage.The number of cones increases with the voltage in stable states,while the cones become short and crowded under high-voltage conditions.The variation law of the number of cones can be explained with the self-organization theory.The variation in the current exhibits a good correlation with the number of cones.This study demonstrates the feasibility of circular emitters and experimentally indicates that the emission current is improved by approximately two orders of magnitude compared to that of a single capillary.展开更多
文摘近三年,以大语言模型(large language models,LLMs)为代表的人工智能技术和产业得到快速发展,实际应用也取得显著效果。为应对国家和行业发展的需求,国内许多高校纷纷设置人工智能学院、学科和专业,但与计算机学院、学科及专业的边界仍不够明晰。为此,本文描述了人工智能的含义、组成,并试图厘清与计算机学科及专业的关系,提出人工智能学院与计算机学院学科、专业发展的一些建议,期望为后续相关研究提供参考。
基金supported by National Natural Science Foundation of China(Grant No.52272100)the Fund of Science and Technology on Advanced Ceramic Fibers and Composites Laboratory(Grant No.WDZC20215250507)the Fund of National Key Laboratory of Nuclear Reactor Technology of Nuclear Power Institute of China(KGSW-0324-0301-08)。
文摘The demand for sensors capable of operating in extreme environment of the fields,such as aerospace vehicles,aeroengines and fire protection,is rapidly increasing.However,developing flexible ceramic fibrous pressure sensors that combine high temperature stability with robust mechanical properties remains a significant challenge.Herein,through precise multi-scale process control,high-strength(2.1 MPa)TiC-SiC flexible fibrous membrane is successfully fabricated.The membrane exhibits exceptional thermal resistance(2000℃)and long–term thermal stability(1800℃ for 5 h)in the inert atmosphere.Meanwhile,the TiC-SiC fibrous membrane shows excellent oxidation resistance and still achieves strength of 1.8 MPa after being oxidized at 1200℃ for 1 h in air.Remarkably,TiC-SiC fibrous membrane withstands a load of approximately 1400 times its own weight and the ablation of butane flame(~1300℃)for at least 1 h without breaking.Notably,after heat treatment at 1800℃ for 5 h in an argon atmosphere,the TiC-SiC fibrous membrane even sustains pressure–sensing performance for up to 300 cycles.The membrane exhibits stable resistivity up to 900℃ and shows sensing stability under butane flame.The results of this work provide an effective and feasible solution to fill the research gap of flexible fibrous sensors for extreme environments.
基金supported by the National Key Research and Development Program of China(No.2022YFB4300902).
文摘As joint operations have become a key trend in modern military development,unmanned aerial vehicles(UAVs)play an increasingly important role in enhancing the intelligence and responsiveness of combat systems.However,the heterogeneity of aircraft,partial observability,and dynamic uncertainty in operational airspace pose significant challenges to autonomous collision avoidance using traditional methods.To address these issues,this paper proposes an adaptive collision avoidance approach for UAVs based on deep reinforcement learning.First,a unified uncertainty model incorporating dynamic wind fields is constructed to capture the complexity of joint operational environments.Then,to effectively handle the heterogeneity between manned and unmanned aircraft and the limitations of dynamic observations,a sector-based partial observation mechanism is designed.A Dynamic Threat Prioritization Assessment algorithm is also proposed to evaluate potential collision threats from multiple dimensions,including time to closest approach,minimum separation distance,and aircraft type.Furthermore,a Hierarchical Prioritized Experience Replay(HPER)mechanism is introduced,which classifies experience samples into high,medium,and low priority levels to preferentially sample critical experiences,thereby improving learning efficiency and accelerating policy convergence.Simulation results show that the proposed HPER-D3QN algorithm outperforms existing methods in terms of learning speed,environmental adaptability,and robustness,significantly enhancing collision avoidance performance and convergence rate.Finally,transfer experiments on a high-fidelity battlefield airspace simulation platform validate the proposed method's deployment potential and practical applicability in complex,real-world joint operational scenarios.
基金supported by the Regional University-Industry Technology Transfer Center for Biopharmaceuticals (Nanjing,Jiangsu) Early-Stage Translational Grant (JB2025211)。
文摘Nucleic acid-based therapies have emerged as promising strategies for the regulation of gene expression and the production of therapeutic antigens or proteins for a series of diseases, including cancers, rare diseases, and infectious diseases. However, their clinical application faces challenges. These include high molecular weight, limited cellular uptake,and susceptibility to enzymatic degradation by nucleases in vivo. Both viral and non-viral delivery vectors have been developed as a means of addressing these limitations, including lipid nanoparticles(LNPs), exosomes, polymers, and inorganic nanoparticles. Among these,LNPs have garnered significant attention due to their superior biocompatibility, high delivery efficiency and customizable design potential, as demonstrated by the clinical success of the FDA-approved si RNA drug Onpattro®. The critical role of nucleic acid drug carriers is discussed in this review. It also outlines the major types of carriers under development and examines the advancements and applications in LNP-based systems for nucleic acid delivery. By conducting a review of recent advancements in LNP design, delivery mechanisms, and clinical applications, this article aims to clarify the ways in which LNPs overcome delivery barriers, compare LNPs with other carriers, and identify key trends that can inform the development of next-generation LNP platforms for nucleic acid therapeutics.
文摘分子筛材料在吸附分离领域具有重要的应用价值,吸附质在分子筛中的晶内扩散行为是评价其吸附分离效果的重要依据之一。零长柱方法(zero length column,ZLC)可准确评价晶内扩散行为,但经典的ZLC实验利用鼓泡法获得不同分压的有机蒸气,无法适用于饱和蒸气压较低的吸附质,同时起泡器温度的微小变化也会影响蒸气压。鉴于此,基于反气相色谱(inverse gas chromatography,IGC),结合微量注射泵搭建了一套IGC-ZLC装置评价晶内扩散行为,以饱和蒸气压较低的正辛烷作为吸附质分子,ZSM-5作为吸附剂,在不同分压、载气流量、进样体积和温度下验证了改进IGC-ZLC方法的准确性,同时对比了正己烷、正庚烷、正壬烷、异辛烷和环辛烷的扩散行为,为在极端场景下测定晶内扩散系数提供了一种潜在的新方法。
基金supported by the Energy Foundation(No.G-2203-33693).
文摘To understand the smoke level and NO_(x)emission characteristics of in-use construction machinery in Beijing,we selected 905 construction machines in Beijing from August 2022 to April 2023 to monitor the emission level of smoke and NO_(x).The exhaust smoke level and excessive emission situation of different machinery types were identified,and their NO_(x)emission levels were monitored according to the free acceleration method.We investigated the correlation of NO_(x)and smoke emission,and proposed suggestions for controlling pollution discharge from construction machinery in the future.The results show that the exhaust smoke level was 0–2.62 m^(−1),followed a log-normal distribution(μ=-1.73,δ=1.09,R^(2)=0.99),with a 5.64%exceedance rate.Differenceswere observed amongmachinery types,with low-power engine forklifts showing higher smoke levels.The NO_(x)emission range was 71–1516 ppm,followed a normal distribution(μ=565.54,δ=309.51,R^(2)=0.83).Differences among machinery types were relatively small.Engine rated net power had the most significant impact on NO_(x)emissions.Thus,NO_(x)emissions from construction machinery need further attention.Furthermore,we found a weak negative correlation(p<0.05)between the emission level of smoke and NO_(x),that is the synergic emission reduction effect is poor,emphasizing the need for NO_(x)emission limits.In the future,the oversight in Beijing should prioritize phasing out ChinaⅠand ChinaⅡmachinery,and monitor emissions from highpower engine ChinaⅢmachinery.
基金co-supported by the National Key R&D Program of China(No.2020YFC2201001)the Shenzhen Science and Technology Program,China(No.20210623091808026)。
文摘Ionic Liquid Electrospray Thrusters(ILETs)are well suited for micro-nano satellite applications due to their small size,low power consumption,and high specific impulse.However,the limited thrust of a single-emitter ILET restricts its use in space missions.To optimize the performance of ILETs and make them suitable for a wider range of space missions,we designed a Circular-emitter ILET(CILET)to convert a one-dimensional(point)emission into a twodimensional(line)emission.The CILET can self-organize multiple Taylor cones simultaneously.The cones were photographed and the axial emission currents were measured under different voltage and pressure difference conditions with a CILET experimental system.The emission can be divided into two stable states and one unstable state based on the flow and current characteristics.The current in Stable state Ⅰ increases non-linearly with the voltage,while that in Stable state Ⅱ is nearly linear with respect to the voltage.The number of cones increases with the voltage in stable states,while the cones become short and crowded under high-voltage conditions.The variation law of the number of cones can be explained with the self-organization theory.The variation in the current exhibits a good correlation with the number of cones.This study demonstrates the feasibility of circular emitters and experimentally indicates that the emission current is improved by approximately two orders of magnitude compared to that of a single capillary.
