Considering the interlayer height, luggage, the difference between queuing pedestrians, and walking speed, the pedestrian choice model of vertical walking facilities is established based on a support vector machine. T...Considering the interlayer height, luggage, the difference between queuing pedestrians, and walking speed, the pedestrian choice model of vertical walking facilities is established based on a support vector machine. This model is verified with the pedestrian flow data of Changchun light-rail transfer station and Beijing Xizhimen transfer station. Adding the pedestrian choice model of vertical walking facilities into the pedestrian simulation model which is based on cellular automata, the pedestrian choice behavior is simulated. In the simulation, the effects of the dynamic influence factors are analyzed. To reduce the conflicts between pedestrians in opposite directions, the layout of vertical walking facilities is improved. The simulations indicate that the improved layout of vertical walking facilities can improve the efficiency of pedestrians passing.展开更多
Least-square support vector machines（LS-SVM） are applied for learning the chaotic behavior of Chua＇s circuit.The system is divided into three multiple-input single-output（MISO） structures and the LS-SVM are train...Least-square support vector machines（LS-SVM） are applied for learning the chaotic behavior of Chua＇s circuit.The system is divided into three multiple-input single-output（MISO） structures and the LS-SVM are trained individually.Comparing with classical approaches,the proposed one reduces the structural complexity and the selection of parameters is avoided.Some parameters of the attractor are used to compare the chaotic behavior of the reconstructed and the original systems for model validation.Results show that the LS-SVM combined with the MISO can be trained to identify the underlying link among Chua＇s circuit state variables,and exhibit the chaotic attractors under the autonomous working mode.展开更多
Compatibility is the precondition to ensure the correct interaction among components in composition process, how to make the mismatch components coordinate correctly is a vital problem in component composition. This p...Compatibility is the precondition to ensure the correct interaction among components in composition process, how to make the mismatch components coordinate correctly is a vital problem in component composition. This paper first modeled component behavior by LTS and expressed action mapping as synchronous vector then defined the sequential relationship among synchronous vectors as adaptation contract. Thirdly we analyzed the different mismatch situations and corresponding adaptation strategies. At last designed adaptation algorithm to produce adaptor specification automatically and ensured the mismatch components can correct interaction under the mediation of adaptor and verified the validity of proposed method through an application system.展开更多
[Objectives]This study aimed to quantitatively investigate the dynamic patterns of spore powder of Metarhizium anisopliae and Beauveria bassiana carried on the body surface of Pyemotes zhonghuajia,providing a theoreti...[Objectives]This study aimed to quantitatively investigate the dynamic patterns of spore powder of Metarhizium anisopliae and Beauveria bassiana carried on the body surface of Pyemotes zhonghuajia,providing a theoretical basis for optimizing synergistic biological control strategies utilizing natural enemies and pathogens.[Methods]Under laboratory conditions,the spore load on mites after crawling on a spore-powder-coated surface for different durations(0-4 h),and the spore shedding after crawling on a clean surface for different durations(0-4 h)post-loading were measured.[Results]The spore load of mites for both fungi increased with crawling time,but exhibited distinct kinetic patterns.The load of M.anisopliae spores peaked at 2 h(6.86×104 spores/mite)and then decreased significantly;whereas the load of B.bassiana spores continued to increase to a maximum(12.83×104 spores/mite)within 4 h,with the rate of increase slowing significantly after 2 h.During the shedding phase,the number of both types of spores decreased with crawling time.After 4 h,the residual amount of B.bassiana spores(3.18×104 spores/mite)was significantly higher than that of M.anisopliae(2.51×104 spores/mite).[Conclusions]The process of P.zhonghuajia carrying entomopathogenic fungal spores exhibits significant temporal dynamics and species specificity.The findings identify key time points for spore loading and shedding,providing crucial parameters for determining the optimal pre-release"loading"duration and assessing the mites sustained dispersal capacity in the field,which holds significant importance for advancing the application of synergistic biological control technology.展开更多
为了提高多目标异常行为的识别精度,提出红外图像序列中多运动目标异常行为识别方法。对红外图像展开一次引导滤波,得到其信息细节,对上述图像再次展开引导滤波,得到二次引导滤波后信息细节,同时和一次引导滤波后信息细节作差,并把上述...为了提高多目标异常行为的识别精度,提出红外图像序列中多运动目标异常行为识别方法。对红外图像展开一次引导滤波,得到其信息细节,对上述图像再次展开引导滤波,得到二次引导滤波后信息细节,同时和一次引导滤波后信息细节作差,并把上述获取的图像细节信息与输入图像展开加权求和,得到增强后的红外图像。将人体骨骼关节点作为依据,分别获取关节速度、运动员重心以及关节角度特征,构建双向长短时记忆网络(Bidirectional Long Short-Term Memory Network,Bi-LSTM),并将上述特征输入该模型内,实现多运动目标异常行为识别。经实验验证得出,所提方法对红外图像的增强效果好,多目标异常行为的识别精度高。展开更多
针对电力通信网等工业互联网中非受控终端不能通过安装代理软件进行异常行为监测的问题,采用非侵入式网络监听手段,采集各终端设备进网流量、出网流量、IP组播流量、IP广播流量、会话总数等数据,提出一种基于长短时记忆网络的自适应动...针对电力通信网等工业互联网中非受控终端不能通过安装代理软件进行异常行为监测的问题,采用非侵入式网络监听手段,采集各终端设备进网流量、出网流量、IP组播流量、IP广播流量、会话总数等数据,提出一种基于长短时记忆网络的自适应动态多核单类支持向量机方法(Long ShortTerm Memory Adaptive Dynamic Multiple Kernel One Class Support Vector Machine,LSTM-ADMK-OCSVM),精确刻画各类非受控终端正常工作行为模态,构建异常行为描述和监测模型,实现对非受控终端设备非设定异常行为安全监测。通过电力信息内网非受控终端实际系统实验,得出所提方法可有效对非受控终端异常行为进行监测,精度达到95.36%,满足实际系统应用要求。展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.51278221 and 51378076)the Science Technology Development Project of Jilin Province,China(Grant No.20140204027SF)
文摘Considering the interlayer height, luggage, the difference between queuing pedestrians, and walking speed, the pedestrian choice model of vertical walking facilities is established based on a support vector machine. This model is verified with the pedestrian flow data of Changchun light-rail transfer station and Beijing Xizhimen transfer station. Adding the pedestrian choice model of vertical walking facilities into the pedestrian simulation model which is based on cellular automata, the pedestrian choice behavior is simulated. In the simulation, the effects of the dynamic influence factors are analyzed. To reduce the conflicts between pedestrians in opposite directions, the layout of vertical walking facilities is improved. The simulations indicate that the improved layout of vertical walking facilities can improve the efficiency of pedestrians passing.
基金Project supported by the National Natural Science Foundation of China (Grant No. 61072103)the Jiangxi Province Training Program for Younger Scientists
文摘Least-square support vector machines（LS-SVM） are applied for learning the chaotic behavior of Chua＇s circuit.The system is divided into three multiple-input single-output（MISO） structures and the LS-SVM are trained individually.Comparing with classical approaches,the proposed one reduces the structural complexity and the selection of parameters is avoided.Some parameters of the attractor are used to compare the chaotic behavior of the reconstructed and the original systems for model validation.Results show that the LS-SVM combined with the MISO can be trained to identify the underlying link among Chua＇s circuit state variables,and exhibit the chaotic attractors under the autonomous working mode.
文摘Compatibility is the precondition to ensure the correct interaction among components in composition process, how to make the mismatch components coordinate correctly is a vital problem in component composition. This paper first modeled component behavior by LTS and expressed action mapping as synchronous vector then defined the sequential relationship among synchronous vectors as adaptation contract. Thirdly we analyzed the different mismatch situations and corresponding adaptation strategies. At last designed adaptation algorithm to produce adaptor specification automatically and ensured the mismatch components can correct interaction under the mediation of adaptor and verified the validity of proposed method through an application system.
基金Supported by HAAFS Science and Technology Innovation Special Project(2022KJCXZX-CGS-8).
文摘[Objectives]This study aimed to quantitatively investigate the dynamic patterns of spore powder of Metarhizium anisopliae and Beauveria bassiana carried on the body surface of Pyemotes zhonghuajia,providing a theoretical basis for optimizing synergistic biological control strategies utilizing natural enemies and pathogens.[Methods]Under laboratory conditions,the spore load on mites after crawling on a spore-powder-coated surface for different durations(0-4 h),and the spore shedding after crawling on a clean surface for different durations(0-4 h)post-loading were measured.[Results]The spore load of mites for both fungi increased with crawling time,but exhibited distinct kinetic patterns.The load of M.anisopliae spores peaked at 2 h(6.86×104 spores/mite)and then decreased significantly;whereas the load of B.bassiana spores continued to increase to a maximum(12.83×104 spores/mite)within 4 h,with the rate of increase slowing significantly after 2 h.During the shedding phase,the number of both types of spores decreased with crawling time.After 4 h,the residual amount of B.bassiana spores(3.18×104 spores/mite)was significantly higher than that of M.anisopliae(2.51×104 spores/mite).[Conclusions]The process of P.zhonghuajia carrying entomopathogenic fungal spores exhibits significant temporal dynamics and species specificity.The findings identify key time points for spore loading and shedding,providing crucial parameters for determining the optimal pre-release"loading"duration and assessing the mites sustained dispersal capacity in the field,which holds significant importance for advancing the application of synergistic biological control technology.
文摘为了提高多目标异常行为的识别精度,提出红外图像序列中多运动目标异常行为识别方法。对红外图像展开一次引导滤波,得到其信息细节,对上述图像再次展开引导滤波,得到二次引导滤波后信息细节,同时和一次引导滤波后信息细节作差,并把上述获取的图像细节信息与输入图像展开加权求和,得到增强后的红外图像。将人体骨骼关节点作为依据,分别获取关节速度、运动员重心以及关节角度特征,构建双向长短时记忆网络(Bidirectional Long Short-Term Memory Network,Bi-LSTM),并将上述特征输入该模型内,实现多运动目标异常行为识别。经实验验证得出,所提方法对红外图像的增强效果好,多目标异常行为的识别精度高。
文摘针对电力通信网等工业互联网中非受控终端不能通过安装代理软件进行异常行为监测的问题,采用非侵入式网络监听手段,采集各终端设备进网流量、出网流量、IP组播流量、IP广播流量、会话总数等数据,提出一种基于长短时记忆网络的自适应动态多核单类支持向量机方法(Long ShortTerm Memory Adaptive Dynamic Multiple Kernel One Class Support Vector Machine,LSTM-ADMK-OCSVM),精确刻画各类非受控终端正常工作行为模态,构建异常行为描述和监测模型,实现对非受控终端设备非设定异常行为安全监测。通过电力信息内网非受控终端实际系统实验,得出所提方法可有效对非受控终端异常行为进行监测,精度达到95.36%,满足实际系统应用要求。
