In this paper, we proposed a model-based abnormality detection scheme for a class of nonlinear parabolic distributed parameter systems (DPSs). The proposed methodology consists of the design of an observer and an abno...In this paper, we proposed a model-based abnormality detection scheme for a class of nonlinear parabolic distributed parameter systems (DPSs). The proposed methodology consists of the design of an observer and an abnormality detection filter (ADF) based on the backstepping technique and a limited number of in-domain measurements plus one boundary measurement. By taking the difference between the measured and estimated outputs from observer, a residual signal is generated for fault detection. For the detection purpose, the residual is evaluated in a lumped manner and we propose an explicit expression for the time-varying threshold. The convergence properties of the PDE observer and the residual are analyzed by Lyapunov stability theory. Eventually, the proposed abnormality detection scheme is demonstrated on a nonlinear DPS.展开更多
The core technology in an intelligent video surveillance system is that detecting and recognizing abnormal behaviors timely and accurately.The key breakthrough point in recognizing abnormal behaviors is how to obtain ...The core technology in an intelligent video surveillance system is that detecting and recognizing abnormal behaviors timely and accurately.The key breakthrough point in recognizing abnormal behaviors is how to obtain the effective features of the picture,so as to solve the problem of recognizing them.In response to this difficulty,this paper introduces an adjustable jump link coefficients model based on the residual network.The effective coefficients for each layer of the network can be set after using this model to further improving the recognition accuracy of abnormal behavior.A convolution kernel of 1×1 size is added to reduce the number of parameters for the purpose of improving the speed of the model in this paper.In order to reduce the noise of the data edge,and at the same time,improve the accuracy of the data and speed up the training,a BN(Batch Normalization)layer is added before the activation function in this network.This paper trains this network model on the public ImageNet dataset,and then uses the transfer learning method to recognize these abnormal behaviors of human in the UTI behavior dataset processed by the YOLO_v3 target detection network.Under the same experimental conditions,compared with the original ResNet-50 model,the improved model in this paper has a 2.8%higher accuracy in recognition of abnormal behaviors on the public UTI dataset.展开更多
The East Asian subtropical westerly jet(EASWJ)is one of the most important factors modulating the Meiyu rainfall in the Yangtze-Huaihe River Basin,China.This article analyzed periods of the medium-term EASWJ variation...The East Asian subtropical westerly jet(EASWJ)is one of the most important factors modulating the Meiyu rainfall in the Yangtze-Huaihe River Basin,China.This article analyzed periods of the medium-term EASWJ variation,wave packet distribution and energy propagation of Rossby waves along the EASWJ during Meiyu season,and investigated their possible influence on abnormal Meiyu rain.The results showed that during the medium-term scale atmospheric dynamic process,the evolution of the EASWJ in Meiyu season was mainly characterized by the changes of3-8 d synoptic-scale and 10-15 d low-frequency Rossby waves.The strong perturbation wave packet and energy propagation of the 3-8 d synoptic-scale and 10-15 d low-frequency Rossby waves are mostly concentrated in the East Asian region of 90°-150°E,where the two wave trains of perturbation wave packets and wave-activity flux divergence coexist in zonal and meridional directions,and converge on the EASWJ.Besides,the wave trains of perturbation wave packet and wave-activity flux divergence in wet Meiyu years are more systematically westward than those in dry Meiyu years,and they are shown in the inverse phases between each other.In wet(dry)Meiyu year,the perturbation wave packet high-value area of the 10-15 d low-frequency variability is located between the Aral Sea and the Lake Balkhash(in the northeastern part of China),while over eastern China the wave-activity flux is convergent and strong(divergent and weak),and the high-level jets are strong and southward(weak and northward).Because of the coupling of high and low level atmosphere and high-level strong(weak)divergence on the south side of the jet over the Yangtze-Huaihe River Basin,the low-level southwest wind and vertically ascending motion are strengthened(weakened),which is(is not)conducive to precipitation increase in the Yangtze-Huaihe River Basin.These findings would help to better understand the impact mechanisms of the EASWJ activities on abnormal Meiyu from the perspective of medium-term scale Rossby wave energy propagation.展开更多
In this paper, an application of fault current limiter-thyristor controller series reactor （FCL-TCSR） in the distribution network is presented in order to minimize the peak value of current during the fault conditio...In this paper, an application of fault current limiter-thyristor controller series reactor （FCL-TCSR） in the distribution network is presented in order to minimize the peak value of current during the fault condition. This application considers a single fault current limiting action but can also be applied for a three phase system. The maximum contribution of FCL-TCSR quickly clears the abnormal current in the power distribution system when a fault condition occurs. Using a mathematical model of FCL-TCSR, the impact of series impedance used to adjust the amplitude of the fault current action is demonstrated. The performance of the load with the impedance and load in series is also analyzed.展开更多
Using time-dependent 3D tomography method, the electron density distributions in the low-latitude ionosphere during November 2004 super-storm are reconstructed from GPS observations of joint ground-based IGS network a...Using time-dependent 3D tomography method, the electron density distributions in the low-latitude ionosphere during November 2004 super-storm are reconstructed from GPS observations of joint ground-based IGS network and onboard CHAMP/GRACE satellites. The reconstructed electron densities are validated by satellite in situ measurements of CHAMP and GRACE satellites. It is indicated by computer tomography （CT） reconstructions that the long-lived positive storm phase during the first main phase of the storm （November 8） is mainly attributed to enhancement of electron density in the upper F region above the F2 peak. It is found by the CT imaging that the top-hat-like F2-3 double layers occurred in the equatorial ionization anomaly region during the main phase of the storm （at forenoon of November 8）. The structures of column-like enhanced electron density are found at the time near the minimum of Dst and in the longitudinal sector about 157°E, which extend from the topside ionosphere toward plasmasphere, reaching at least about 2000 km as high. Their footprints stand on the two peaks of the EIA.展开更多
分布式光伏发电系统一般不配备多种类的传感器和监测设备,反映设备运行状态且可用于异常检测的数据有限。提出了基于STL-Bayesian时空模型的光伏异常状态检测方法,利用气象在时空上的传递性,挖掘光伏发电出力的关联性进而完成异常检测...分布式光伏发电系统一般不配备多种类的传感器和监测设备,反映设备运行状态且可用于异常检测的数据有限。提出了基于STL-Bayesian时空模型的光伏异常状态检测方法,利用气象在时空上的传递性,挖掘光伏发电出力的关联性进而完成异常检测。首先,用季节性分解(seasonal and trend decomposition using loess,STL)将光伏发电有功功率时序数据分解为3个分量;然后,研究不同长度数据输入对分解结果的影响和区域内分量的时空分布特性;接着,通过构建贝叶斯模型分别对趋势分量和剩余分量做短期和超短期空间插值,得到区域内光伏出力;最后,计算真实值与回归值的推土机距离(earth move's distance,EMD)用于检测异常状态。算例分析表明,所提模型在分布式光伏场景检测可逆异常和不可逆异常状态均有较高准确率。展开更多
文摘In this paper, we proposed a model-based abnormality detection scheme for a class of nonlinear parabolic distributed parameter systems (DPSs). The proposed methodology consists of the design of an observer and an abnormality detection filter (ADF) based on the backstepping technique and a limited number of in-domain measurements plus one boundary measurement. By taking the difference between the measured and estimated outputs from observer, a residual signal is generated for fault detection. For the detection purpose, the residual is evaluated in a lumped manner and we propose an explicit expression for the time-varying threshold. The convergence properties of the PDE observer and the residual are analyzed by Lyapunov stability theory. Eventually, the proposed abnormality detection scheme is demonstrated on a nonlinear DPS.
基金This research was funded by the Science and Technology Department of Shaanxi Province,China,Grant Number 2019GY-036.
文摘The core technology in an intelligent video surveillance system is that detecting and recognizing abnormal behaviors timely and accurately.The key breakthrough point in recognizing abnormal behaviors is how to obtain the effective features of the picture,so as to solve the problem of recognizing them.In response to this difficulty,this paper introduces an adjustable jump link coefficients model based on the residual network.The effective coefficients for each layer of the network can be set after using this model to further improving the recognition accuracy of abnormal behavior.A convolution kernel of 1×1 size is added to reduce the number of parameters for the purpose of improving the speed of the model in this paper.In order to reduce the noise of the data edge,and at the same time,improve the accuracy of the data and speed up the training,a BN(Batch Normalization)layer is added before the activation function in this network.This paper trains this network model on the public ImageNet dataset,and then uses the transfer learning method to recognize these abnormal behaviors of human in the UTI behavior dataset processed by the YOLO_v3 target detection network.Under the same experimental conditions,compared with the original ResNet-50 model,the improved model in this paper has a 2.8%higher accuracy in recognition of abnormal behaviors on the public UTI dataset.
基金National Natural Science Foundation of China(41575066)National Science and Technology Support Program of China(2015BAC03B04)
文摘The East Asian subtropical westerly jet(EASWJ)is one of the most important factors modulating the Meiyu rainfall in the Yangtze-Huaihe River Basin,China.This article analyzed periods of the medium-term EASWJ variation,wave packet distribution and energy propagation of Rossby waves along the EASWJ during Meiyu season,and investigated their possible influence on abnormal Meiyu rain.The results showed that during the medium-term scale atmospheric dynamic process,the evolution of the EASWJ in Meiyu season was mainly characterized by the changes of3-8 d synoptic-scale and 10-15 d low-frequency Rossby waves.The strong perturbation wave packet and energy propagation of the 3-8 d synoptic-scale and 10-15 d low-frequency Rossby waves are mostly concentrated in the East Asian region of 90°-150°E,where the two wave trains of perturbation wave packets and wave-activity flux divergence coexist in zonal and meridional directions,and converge on the EASWJ.Besides,the wave trains of perturbation wave packet and wave-activity flux divergence in wet Meiyu years are more systematically westward than those in dry Meiyu years,and they are shown in the inverse phases between each other.In wet(dry)Meiyu year,the perturbation wave packet high-value area of the 10-15 d low-frequency variability is located between the Aral Sea and the Lake Balkhash(in the northeastern part of China),while over eastern China the wave-activity flux is convergent and strong(divergent and weak),and the high-level jets are strong and southward(weak and northward).Because of the coupling of high and low level atmosphere and high-level strong(weak)divergence on the south side of the jet over the Yangtze-Huaihe River Basin,the low-level southwest wind and vertically ascending motion are strengthened(weakened),which is(is not)conducive to precipitation increase in the Yangtze-Huaihe River Basin.These findings would help to better understand the impact mechanisms of the EASWJ activities on abnormal Meiyu from the perspective of medium-term scale Rossby wave energy propagation.
文摘In this paper, an application of fault current limiter-thyristor controller series reactor （FCL-TCSR） in the distribution network is presented in order to minimize the peak value of current during the fault condition. This application considers a single fault current limiting action but can also be applied for a three phase system. The maximum contribution of FCL-TCSR quickly clears the abnormal current in the power distribution system when a fault condition occurs. Using a mathematical model of FCL-TCSR, the impact of series impedance used to adjust the amplitude of the fault current action is demonstrated. The performance of the load with the impedance and load in series is also analyzed.
基金supported by the National Natural Science Foundation of China (Grant Nos. 40974087, 40674078)
文摘Using time-dependent 3D tomography method, the electron density distributions in the low-latitude ionosphere during November 2004 super-storm are reconstructed from GPS observations of joint ground-based IGS network and onboard CHAMP/GRACE satellites. The reconstructed electron densities are validated by satellite in situ measurements of CHAMP and GRACE satellites. It is indicated by computer tomography （CT） reconstructions that the long-lived positive storm phase during the first main phase of the storm （November 8） is mainly attributed to enhancement of electron density in the upper F region above the F2 peak. It is found by the CT imaging that the top-hat-like F2-3 double layers occurred in the equatorial ionization anomaly region during the main phase of the storm （at forenoon of November 8）. The structures of column-like enhanced electron density are found at the time near the minimum of Dst and in the longitudinal sector about 157°E, which extend from the topside ionosphere toward plasmasphere, reaching at least about 2000 km as high. Their footprints stand on the two peaks of the EIA.
文摘分布式光伏发电系统一般不配备多种类的传感器和监测设备,反映设备运行状态且可用于异常检测的数据有限。提出了基于STL-Bayesian时空模型的光伏异常状态检测方法,利用气象在时空上的传递性,挖掘光伏发电出力的关联性进而完成异常检测。首先,用季节性分解(seasonal and trend decomposition using loess,STL)将光伏发电有功功率时序数据分解为3个分量;然后,研究不同长度数据输入对分解结果的影响和区域内分量的时空分布特性;接着,通过构建贝叶斯模型分别对趋势分量和剩余分量做短期和超短期空间插值,得到区域内光伏出力;最后,计算真实值与回归值的推土机距离(earth move's distance,EMD)用于检测异常状态。算例分析表明,所提模型在分布式光伏场景检测可逆异常和不可逆异常状态均有较高准确率。
