A lightweight malware detection and family classification system for the Internet of Things (IoT) was designed to solve the difficulty of deploying defense models caused by the limited computing and storage resources ...A lightweight malware detection and family classification system for the Internet of Things (IoT) was designed to solve the difficulty of deploying defense models caused by the limited computing and storage resources of IoT devices. By training complex models with IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Additionally, the multi-teacher knowledge distillation method is employed to train KD-LMDNet, which focuses on classifying malware families. The results indicate that the model’s identification speed surpasses that of traditional methods by 23.68%. Moreover, the accuracy achieved on the Malimg dataset for family classification is an impressive 99.07%. Furthermore, with a model size of only 0.45M, it appears to be well-suited for the IoT environment. By training complex models using IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Thus, the presented approach can address the challenges associated with malware detection and family classification in IoT devices.展开更多
Widespread magmatism, metamorphic core complexes(MCCs), and significant lithospheric thinning occurred during the Mesozoic in the North China Craton(NCC). It has been suggested that the coeval exhumation of MCCs with ...Widespread magmatism, metamorphic core complexes(MCCs), and significant lithospheric thinning occurred during the Mesozoic in the North China Craton(NCC). It has been suggested that the coeval exhumation of MCCs with uniform northwest-southeast shear senses and magmatism probably resulted from a decratonization event during the retreat of the paleo-Pacific Plate. Here we used two-dimensional finite element thermomechanical numerical models to investigate critical parameters controlling the formation of MCCs under far-field extensional stress. We observed three end-member deformation modes: the MCC mode, the symmetric-dome mode, and the pure-shear mode. The MCC mode requires a Moho temperature of ≥700 ℃ and an extensional strain rate of ≥5 × 10^(-16)s^(-1), implying that the lithosphere had already thinned when the MCC was formed in the Mesozoic. Considering that the widespread MCCs have the same northwest-southeast extension direction in the NCC, we suggest that the MCCs are surface expressions of both large-scale extension and craton destruction and that rollback of the paleo-Pacific slab might be the common driving force.展开更多
Robots are playing an increasingly important role in engineering applications.Soft robots have promising applications in several fields due to their inherent advantages of compliance,low density,and soft interactions....Robots are playing an increasingly important role in engineering applications.Soft robots have promising applications in several fields due to their inherent advantages of compliance,low density,and soft interactions.A soft gripper based on bio-inspiration is proposed in this study.We analyze the cushioning and energy absorption mechanism of human fingertips in detail and provide insights for designing a soft gripper with a variable stiffness structure.We investigate the grasping modes through a large deformation modeling approach,which is verified through experiments.The characteristics of the three grasping modes are quantified through testing and can provide guidance for robotics manipulation.First,the adaptability of the soft gripper is verified by grasping multi-scale and extremely soft objects.Second,a cushioning model of the soft gripper is proposed,and the effectiveness of cushioning is verified by grasping extremely sharp objects and living organisms.Notably,we validate the advantages of the variable stiffness of the soft gripper,and the results show that the soft robot can robustly complete assemblies with a gap of only 0.1 mm.Owing to the unstructured nature of the engineering environment,the soft gripper can be applied in complex environments based on the abovementioned experimental analysis.Finally,we design the soft robotics system with feedback capture based on the inspiration of human catching behavior.The feasibility of engineering applications is initially verified through fast capture experiments on moving objects.The design concept of this robot can provide new insights for bionic machinery.展开更多
Cadmium(Cd)is a common toxic heavy metal in the environment.Taking Cd(II)as a target contaminant,we systematically compared the performances of three Fe-based nanomaterials(nano zero valent iron,nZVI;sulfidated nZVI,S...Cadmium(Cd)is a common toxic heavy metal in the environment.Taking Cd(II)as a target contaminant,we systematically compared the performances of three Fe-based nanomaterials(nano zero valent iron,nZVI;sulfidated nZVI,S-nZVI;and nano FeS,nFeS)for Cd immobilization under anaerobic conditions.Effects of nanomaterials doses,initial pH,co-existing ions,and humic acid(HA)were examined.Under identical conditions,at varied doses or initial pH,Cd(II)removal by three materials followed the order of S-nZVI>nFeS>nZVI.At pH 6,the Cd(II)removal within 24 hours for S-nZVI,nFeS,and nZVI(dose of 20 mg/L)were 93.50%,89.12%and 4.10%,respectively.The fast initial reaction rate of nZVI did not lead to a high removal capacity.The Cd removal was slightly impacted or even improved with co-existing ions(at 50 mg/L or 200 mg/L)or HA(at 2 mg/L or 20 mg/L).Characterization results revealed that nZVI immobilized Cd through coprecipitation,surface complexation,and reduction,whereas the mechanisms for sulfidated materials involved replacement,coprecipitation,and surface complexation,with replacement as the predominant reaction.A strong linear correlation between Cd(II)removal and Fe(II)dissolution was observed,and we proposed a novel pseudo-second-order kinetic model to simulate Fe(II)dissolution.展开更多
Research about farmland pollution by heavy metals/metalloids in China has drawn growing attention.However,there was rare information on spatiotemporal evolution and pollution levels of heavy metals in the major grainp...Research about farmland pollution by heavy metals/metalloids in China has drawn growing attention.However,there was rare information on spatiotemporal evolution and pollution levels of heavy metals in the major grainproducing areas.We extracted and examined data from 276 publications between 2010 and 2021 covering five major grain-producing regions in China from 2010 to 2021.Spatiotemporal evolution characteristics of main heavy metals/metalloids was obtained by meta-analysis.In addition,subgroup analyses were carried out to study preliminary correlations related to accumulation of the pollutants.Cadmium(Cd)was found to be the most prevailing pollutant in the regions in terms of both spatial distribution and temporal accumulation.The Huang-Huai-Hai Plain was the most severely polluted.Accumulation of Cd,mercury(Hg)and copper(Cu)increased from 2010 to 2015 when compared with the 1990 background data.Further,the levels of five key heavy metals(Cd,Cu,Hg,lead[Pb]and zinc[Zn])showed increasing trends from 2016 to 2021 in all five regions.Soil pH and mean annual precipitation had variable influences on heavy metal accumulation.Alkaline soil and areas with less rainfall faced higher pollution levels.Farmlands cropped with mixed species showed smaller effect sizes of heavy metals than those with single upland crop,suggesting that mixed farmland use patterns could alleviate the levels of heavy metals in soil.Of various soil remediation efforts,farmland projects only held a small market share.The findings are important to support the research of risk assessment,regulatory development,pollution prevention,fund allocation and remediation actions.展开更多
In this study,(Cr_(1/3)/Ta_(2/3))non-equivalent co-doped Bi_(4)Ti_(3)O_(12)(BIT)ceramics were prepared to solve the problem that high piezoelectric performance,high Curie temperature,and high-temperature resistivity c...In this study,(Cr_(1/3)/Ta_(2/3))non-equivalent co-doped Bi_(4)Ti_(3)O_(12)(BIT)ceramics were prepared to solve the problem that high piezoelectric performance,high Curie temperature,and high-temperature resistivity could not be achieved simultaneously in BIT-based ceramics.A series of Bi_(4)Ti_(3-x)(Cr_(1/3)Ta_(2/3))_(x)O_(12)(x=0-0.04)ceramics were synthesized by the solid-state reaction method.The phase structure,microstructure,piezoelectric performance,and conductive mechanism of the samples were systematically investigated.The B-site non-equivalent co-doping strategy combining high-valence Ta^(5+)and low-valence Cr^(3+)significantly enhances electrical properties due to a decrease in oxygen vacancy concentration.Bi_(4)Ti_(2.97)(Cr_(1/3)Ta_(2/3))_(0.03)O_(12)ceramics exhibit a high piezoelectric coefficient(d_(33)=26 pC·N^(-1))and a high Curie temperature(TC=687℃).Moreover,the significantly increased resistivity(ρ=2.8×10^(6)Ω·cm at 500℃)and good piezoelectric stability up to 600℃are also obtained for this composition.All the results demonstrate that Cr/Ta co-doped BIT-based ceramics have great potential to be applied in high-temperature piezoelectric applications.展开更多
文摘A lightweight malware detection and family classification system for the Internet of Things (IoT) was designed to solve the difficulty of deploying defense models caused by the limited computing and storage resources of IoT devices. By training complex models with IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Additionally, the multi-teacher knowledge distillation method is employed to train KD-LMDNet, which focuses on classifying malware families. The results indicate that the model’s identification speed surpasses that of traditional methods by 23.68%. Moreover, the accuracy achieved on the Malimg dataset for family classification is an impressive 99.07%. Furthermore, with a model size of only 0.45M, it appears to be well-suited for the IoT environment. By training complex models using IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Thus, the presented approach can address the challenges associated with malware detection and family classification in IoT devices.
基金supported by the National Natural Science Foundation of China(Grant No.41774112)。
文摘Widespread magmatism, metamorphic core complexes(MCCs), and significant lithospheric thinning occurred during the Mesozoic in the North China Craton(NCC). It has been suggested that the coeval exhumation of MCCs with uniform northwest-southeast shear senses and magmatism probably resulted from a decratonization event during the retreat of the paleo-Pacific Plate. Here we used two-dimensional finite element thermomechanical numerical models to investigate critical parameters controlling the formation of MCCs under far-field extensional stress. We observed three end-member deformation modes: the MCC mode, the symmetric-dome mode, and the pure-shear mode. The MCC mode requires a Moho temperature of ≥700 ℃ and an extensional strain rate of ≥5 × 10^(-16)s^(-1), implying that the lithosphere had already thinned when the MCC was formed in the Mesozoic. Considering that the widespread MCCs have the same northwest-southeast extension direction in the NCC, we suggest that the MCCs are surface expressions of both large-scale extension and craton destruction and that rollback of the paleo-Pacific slab might be the common driving force.
基金supported by the General Program(Grant No.12272222)Key Program(Grant No.11932001)of the National Natural Science Foundation of China,for which the authors are grateful.
文摘Robots are playing an increasingly important role in engineering applications.Soft robots have promising applications in several fields due to their inherent advantages of compliance,low density,and soft interactions.A soft gripper based on bio-inspiration is proposed in this study.We analyze the cushioning and energy absorption mechanism of human fingertips in detail and provide insights for designing a soft gripper with a variable stiffness structure.We investigate the grasping modes through a large deformation modeling approach,which is verified through experiments.The characteristics of the three grasping modes are quantified through testing and can provide guidance for robotics manipulation.First,the adaptability of the soft gripper is verified by grasping multi-scale and extremely soft objects.Second,a cushioning model of the soft gripper is proposed,and the effectiveness of cushioning is verified by grasping extremely sharp objects and living organisms.Notably,we validate the advantages of the variable stiffness of the soft gripper,and the results show that the soft robot can robustly complete assemblies with a gap of only 0.1 mm.Owing to the unstructured nature of the engineering environment,the soft gripper can be applied in complex environments based on the abovementioned experimental analysis.Finally,we design the soft robotics system with feedback capture based on the inspiration of human catching behavior.The feasibility of engineering applications is initially verified through fast capture experiments on moving objects.The design concept of this robot can provide new insights for bionic machinery.
基金support by the National Natural Science Foundation of China(No.51809267)the Chinese Universities Scientific Fund(No.109018).
文摘Cadmium(Cd)is a common toxic heavy metal in the environment.Taking Cd(II)as a target contaminant,we systematically compared the performances of three Fe-based nanomaterials(nano zero valent iron,nZVI;sulfidated nZVI,S-nZVI;and nano FeS,nFeS)for Cd immobilization under anaerobic conditions.Effects of nanomaterials doses,initial pH,co-existing ions,and humic acid(HA)were examined.Under identical conditions,at varied doses or initial pH,Cd(II)removal by three materials followed the order of S-nZVI>nFeS>nZVI.At pH 6,the Cd(II)removal within 24 hours for S-nZVI,nFeS,and nZVI(dose of 20 mg/L)were 93.50%,89.12%and 4.10%,respectively.The fast initial reaction rate of nZVI did not lead to a high removal capacity.The Cd removal was slightly impacted or even improved with co-existing ions(at 50 mg/L or 200 mg/L)or HA(at 2 mg/L or 20 mg/L).Characterization results revealed that nZVI immobilized Cd through coprecipitation,surface complexation,and reduction,whereas the mechanisms for sulfidated materials involved replacement,coprecipitation,and surface complexation,with replacement as the predominant reaction.A strong linear correlation between Cd(II)removal and Fe(II)dissolution was observed,and we proposed a novel pseudo-second-order kinetic model to simulate Fe(II)dissolution.
基金support by the National Natural Science Foundation of China(No.51809267)the Chinese Universities Scientific Fund(No.00109018)the 2115 Talent Development Program of China Agricultural University(No.00109018).
文摘Research about farmland pollution by heavy metals/metalloids in China has drawn growing attention.However,there was rare information on spatiotemporal evolution and pollution levels of heavy metals in the major grainproducing areas.We extracted and examined data from 276 publications between 2010 and 2021 covering five major grain-producing regions in China from 2010 to 2021.Spatiotemporal evolution characteristics of main heavy metals/metalloids was obtained by meta-analysis.In addition,subgroup analyses were carried out to study preliminary correlations related to accumulation of the pollutants.Cadmium(Cd)was found to be the most prevailing pollutant in the regions in terms of both spatial distribution and temporal accumulation.The Huang-Huai-Hai Plain was the most severely polluted.Accumulation of Cd,mercury(Hg)and copper(Cu)increased from 2010 to 2015 when compared with the 1990 background data.Further,the levels of five key heavy metals(Cd,Cu,Hg,lead[Pb]and zinc[Zn])showed increasing trends from 2016 to 2021 in all five regions.Soil pH and mean annual precipitation had variable influences on heavy metal accumulation.Alkaline soil and areas with less rainfall faced higher pollution levels.Farmlands cropped with mixed species showed smaller effect sizes of heavy metals than those with single upland crop,suggesting that mixed farmland use patterns could alleviate the levels of heavy metals in soil.Of various soil remediation efforts,farmland projects only held a small market share.The findings are important to support the research of risk assessment,regulatory development,pollution prevention,fund allocation and remediation actions.
基金This work is financially supported by the National Natural Science Foundation of China(No.52172135)the Youth Top Talent Project of the National Special Support Program(No.2021-527-07)+1 种基金the Leading Talent Project of the National Special Support Program(No.2022WRLJ003)the Guangdong Basic and Applied Basic Research Foundation for Distinguished Young Scholars(Nos.2022B1515020070 and 2021B1515020083).
文摘In this study,(Cr_(1/3)/Ta_(2/3))non-equivalent co-doped Bi_(4)Ti_(3)O_(12)(BIT)ceramics were prepared to solve the problem that high piezoelectric performance,high Curie temperature,and high-temperature resistivity could not be achieved simultaneously in BIT-based ceramics.A series of Bi_(4)Ti_(3-x)(Cr_(1/3)Ta_(2/3))_(x)O_(12)(x=0-0.04)ceramics were synthesized by the solid-state reaction method.The phase structure,microstructure,piezoelectric performance,and conductive mechanism of the samples were systematically investigated.The B-site non-equivalent co-doping strategy combining high-valence Ta^(5+)and low-valence Cr^(3+)significantly enhances electrical properties due to a decrease in oxygen vacancy concentration.Bi_(4)Ti_(2.97)(Cr_(1/3)Ta_(2/3))_(0.03)O_(12)ceramics exhibit a high piezoelectric coefficient(d_(33)=26 pC·N^(-1))and a high Curie temperature(TC=687℃).Moreover,the significantly increased resistivity(ρ=2.8×10^(6)Ω·cm at 500℃)and good piezoelectric stability up to 600℃are also obtained for this composition.All the results demonstrate that Cr/Ta co-doped BIT-based ceramics have great potential to be applied in high-temperature piezoelectric applications.