Belt conveyors are extensively utilized in mining and power industries.In a typical coal mine conveyor system,coal is transported over distances exceeding 2 km,involving more than 20000 idlers,which far exceeds a reas...Belt conveyors are extensively utilized in mining and power industries.In a typical coal mine conveyor system,coal is transported over distances exceeding 2 km,involving more than 20000 idlers,which far exceeds a reasonable manual inspection capacity.Given that idlers typically have a lifespan of 1-2 years,there is an urgent need for a rapid,cost-effective,and intelligent safety monitoring system.However,current embedded systems face prohibitive replacement costs,while conventional monitoring technologies suffer from inefficiency at low rotational speeds and lack systematic structural optimization frameworks for diverse idler types and parameters.To address these challenges,this paper introduces an integrated,on-site detachable self-powered idler condition monitoring system(ICMS).This system combines energy harvesting based on the magnetic modulation technology with wireless condition monitoring capabilities.Specifically,it develops a data-driven model integrating convolutional neural networks(CNNs) with genetic algorithms(GAs).The conventional testing results show that the data-driven model not only significantly accelerates the parameter response time,but also achieves a prediction accuracy of 92.95%.The in-situ experiments conducted in coal mines demonstrate the system's reliability and monitoring functionality under both no-load and fullload conditions.This research provides an innovative self-powered condition monitoring solution and develops an efficient data-driven model,offering feasible online monitoring approaches for smart mine construction.展开更多
A designed visual geometry group(VGG)-based convolutional neural network(CNN)model with small computational cost and high accuracy is utilized to monitor pulse amplitude modulation-based intensity modulation and direc...A designed visual geometry group(VGG)-based convolutional neural network(CNN)model with small computational cost and high accuracy is utilized to monitor pulse amplitude modulation-based intensity modulation and direct detection channel performance using eye diagram measurements.Experimental results show that the proposed technique can achieve a high accuracy in jointly monitoring modulation format,probabilistic shaping,roll-off factor,baud rate,optical signal-to-noise ratio,and chromatic dispersion.The designed VGG-based CNN model outperforms the other four traditional machine-learning methods in different scenarios.Furthermore,the multitask learning model combined with MobileNet CNN is designed to improve the flexibility of the network.Compared with the designed VGG-based CNN,the MobileNet-based MTL does not need to train all the classes,and it can simultaneously monitor single parameter or multiple parameters without sacrificing accuracy,indicating great potential in various monitoring scenarios.展开更多
Multi-functioning in virtual monitoring and assessment of ultimate dynamics of thin-walled bridges is treated in present paper. The approach enables multiple functions in virtual monitoring of the bridges made of inte...Multi-functioning in virtual monitoring and assessment of ultimate dynamics of thin-walled bridges is treated in present paper. The approach enables multiple functions in virtual monitoring of the bridges made of integrated thin-walled members with their hierarchical configuration. Theoretical, numerical and experimental in situ assessments of the problem are presented. Some results of structural application are submitted.展开更多
High-temperature electromagnetic(EM) protection materials integrated of multiple EM protection mechanisms and functions are regarded as desirable candidates for solving EM interference over a wide temperature range.In...High-temperature electromagnetic(EM) protection materials integrated of multiple EM protection mechanisms and functions are regarded as desirable candidates for solving EM interference over a wide temperature range.In this work,a novel microwave modulator is fabricated by introducing carbonyl iron particles(CIP)/resin into channels of carbonized wood(C-wood).Innovatively,the spaced arrangement of two microwave absorbents not only achieves a synergistic enhancement of magnetic and dielectric losses,but also breaks the translational invariance of EM characteristics in the horizontal direction to obtain multiple phase discontinuities in the frequency range of 8.2-18.0 GHz achieving modulation of reflected wave radiation direction.Accordingly,CIP/C-wood microwave modulator demonstrates the maximum effective bandwidth of 5.2 GHz and the maximum EM protection efficiency over 97% with a thickness of only 1.5 mm in the temperature range 298-673 K.Besides,CIP/C-wood microwave modulator shows stable and low thermal conductivities,as well as monotonic electrical conductivity-temperature characteristics,therefore it can also achieve thermal infrared stealth and working temperature monitoring in wide temperature ranges.This work provides an inspiration for the design of high-temperature EM protection materials with multiple EM protection mechanisms and functions.展开更多
Dynamic temperature monitoring at critical locations of IGBT modules is a key means to improve the reliability of high-power converters.However,most ex-isting thermal model-based methods suffer from temper-ature estim...Dynamic temperature monitoring at critical locations of IGBT modules is a key means to improve the reliability of high-power converters.However,most ex-isting thermal model-based methods suffer from temper-ature estimation errors due to model parameter varia-tions and loss calculation errors.To address this problem,based on the reduced-order thermal model,an H_(∞)ob-server-based robust 3-D thermal monitoring method for IGBT modules is proposed in this paper.Through the optimized design of the observer feedback gain,the thermal model and real-time temperature information are effectively combined,which reduces the temperature estimation error in the worst case.Thus,the proposed method is more robust to model parameter uncertainty and loss error than the conventional temperature ob-servers.Experiment validations of the proposed H_(∞)ob-server and conventional observers are provided.The results demonstrate that the proposed observer achieves the highest temperature estimation accuracy under vari-ous system uncertainties,making it an effective solution for reliable online thermal monitoring of IGBT modules over the whole life cycle.展开更多
The rapid development of network technology and its evolution toward heterogeneous networks has increased the demand to support automatic monitoring and the management of heterogeneous wireless communication networks....The rapid development of network technology and its evolution toward heterogeneous networks has increased the demand to support automatic monitoring and the management of heterogeneous wireless communication networks.This paper presents a multilevel pattern mining architecture to support automatic network management by discovering interesting patterns from telecom network monitoring data.This architecture leverages and combines existing frequent itemset discovery over data streams,association rule deduction,frequent sequential pattern mining,and frequent temporal pattern mining techniques while also making use of distributed processing platforms to achieve high-volume throughput.展开更多
A technique using artificial neural networks trained with parameters derived from delay tap plots for optical performance monitoring in 40 Gbit/s duobinary system is demonstrated. Firstly, the optical signal is delay ...A technique using artificial neural networks trained with parameters derived from delay tap plots for optical performance monitoring in 40 Gbit/s duobinary system is demonstrated. Firstly, the optical signal is delay tap sampled to obtain two-dimensional histogram, known as delay tap plots. Secondly, the features of delay tap plots are extracted to train the feed forward, three-layer preceptor structure artificial neural networks. Finally, the outputs of trained neural network are used to monitor optical duobinary signal impairments. Simulation of optical signal noise ratio （ OSNR）, chromatic dispersion （CD）, and differential group delay （DGD） monitoring in 40 Gbit/s optical duo- binary system is presented. The proposed monitoring scheme can accurately identify simultaneous impairments without requiring synchronous sampling or data clock recovery. The proposed technique is simple, cost-effective and suitable for in-service distributed OPM.展开更多
A signal-amplified mercury sensing biosensor with desired sensitivity was developed through firstly using the GFP mutant with fluorescence increasing response towards Hg^2+ as the reporter module.The developed biosens...A signal-amplified mercury sensing biosensor with desired sensitivity was developed through firstly using the GFP mutant with fluorescence increasing response towards Hg^2+ as the reporter module.The developed biosensor showed response for Hg^2+ in a relatively wide range of 1–10,000 nmol/L,and the detection limit was improved one or two orders of magnitude in comparison with most metal-sensing biosensors in similar constructs.In addition,the biosensor could distinguish Hg^2+ easily from multiple metal ions and displayed strong adaptability to extensive p H conditions (pH 4.0–10.0).More importantly,the developed biosensor was able to provide an initial assessment of Hg^2+ spiked in the environmental water with the recoveries between 85.70%and 112.50%.The signal-amplified strategy performed by the modified reporter module will be widely applicable to many other wholecell biosensors,meeting the practical requirements with sufficient sensing performance.展开更多
Many safety-critical applications that utilize the global navigation satellite system (GNSS) demand highly accurate positioning information, as well as highly integrity and reliability. Due to GNSS signals are easily ...Many safety-critical applications that utilize the global navigation satellite system (GNSS) demand highly accurate positioning information, as well as highly integrity and reliability. Due to GNSS signals are easily distorted by the interferences or disturbances, the signal quality monitoring (SQM) is necessary to detect the presence of dangerous signal distortions. In this paper, we developed an SQM software for binary offset carrier (BOC) modulated navigation signals. Firstly, the models of BOC signal with ideal and distortion are presented respectively. Then the architecture of SQM software is proposed. Moreover, the effect of the white gaussian noise (WGN) and the front-end filter on the correlation peak of the receiver is analyzed. Finally, the biases induced by the signal distortion are evaluated. The experiments simulate the relationships between the code phase shift and the normalized correlation value in the case of the signal digital distortion and the analog distortion. The simulation results demonstrate that the proposed SQM method can effectively monitor the signal distortion and accurately estimate the correlation peak deviation caused by the distortion.展开更多
A bunch arrival-time monitor(BAM) based on an electro-optical intensity modulation scheme is currently under development at Shanghai Soft X-ray Free-Electron Laser to meet the high-resolution requirements for bunch st...A bunch arrival-time monitor(BAM) based on an electro-optical intensity modulation scheme is currently under development at Shanghai Soft X-ray Free-Electron Laser to meet the high-resolution requirements for bunch stability. The BAM uses a radio frequency signal generated by a pickup cavity to modulate the reference laser pulses in an electro-optical intensity modulator(EOM), and the bunch arrival-time information is derived from the amplitude change of the laser pulse after laser pulse modulation.EOM is a key optical component in the BAM system.Through the basic principle analysis of BAM, many parameters of the EOM are observed to affect the measurement resolution of the BAM system. Therefore, a systematic analysis of the EOM is crucial. In this paper, we present two schemes to compare and analyze an EOM and provide a reference for selecting a new version of the EOM.展开更多
Pursuing small critical dimensions(i.e.14 nm or below)and high integration bring us lots of physical defects causing low yield and functionality failures for foundries.Under this circumstance,inspection,metrology and ...Pursuing small critical dimensions(i.e.14 nm or below)and high integration bring us lots of physical defects causing low yield and functionality failures for foundries.Under this circumstance,inspection,metrology and monitoring technologies are unprecedentedly vital for development of semiconductor industry.Optical and electron beam solutions are the most common two methods in semiconductor manufacturing.Hamamatsu Photonics is now aiming at optical inspection,metrology and monitoring systems market by providing light sources,photodetectors and failure analysis systems for semiconductor equipment manufacturers,foundries and research institutions.In this paper,features and potential applications of light sources,photodetectors(like image sensors,photomultiplier tubes/modules,silicon photomultipliers(modules),(avalanche)photodiodes(arrays)and so on),with the wavelengths ranging from UV to Infrared,are mainly discussed.In addition,Hamamatsu’s star product– failure analysis system to quickly locate faults or defects are introduced.In conclusion,Hamamatsu Photonics is dedicated to develop large varieties of light sources and optical sensors/detector/modules along with failure analysis systems and willing to improve the development of semiconductor and related industries,especially in China.展开更多
非侵入式负荷监测(non-intrusive load monitoring,NILM)技术对于实现智慧用电与管理具有重要意义。针对现有的非侵入式负荷监测方法在高噪声环境下对特征相似电器以及微小负荷变化监测精度不足的难题,提出了一种基于单位力操作视觉变...非侵入式负荷监测(non-intrusive load monitoring,NILM)技术对于实现智慧用电与管理具有重要意义。针对现有的非侵入式负荷监测方法在高噪声环境下对特征相似电器以及微小负荷变化监测精度不足的难题,提出了一种基于单位力操作视觉变换器的非侵入式负荷监测(non-intrusive load monitoring based on unit force operated vision transformer,UFONILM)模型的非侵入式负荷监测的深度学习框架。UFONILM模型的单位力操作(unit force operated,UFO)模块通过层归一化和一系列卷积层有效地提取和利用了多尺度的时间序列数据,特征。在标准的UK-DALE数据集上进行的实验显示,UFONILM模型在准确性和F1得分上均优于现有方法,特别是在细粒度的负荷监测场景中。研制了基于UFONILM模型的嵌入式系统,实现了边缘计算的非侵入式负荷监测,可实时监测和响应电网中的异常用电行为,如违规充电事件。实验检测证明,UFONILM模型嵌入式非侵入式负荷监测方法在监测效率方面具有显著的提升,具有高效、便捷安装、可扩展等特点。展开更多
为提升FPGA(Field Programmable Gate Array)高速Serdes通讯稳定性,实时监控其通讯状态,文中设计了一种基于K码控制字符的通讯协议。创建标志用户数据帧起始的动态SOF(Start of Frame)和标志结束的静态EOF(End of Frame)两种K码控制字符...为提升FPGA(Field Programmable Gate Array)高速Serdes通讯稳定性,实时监控其通讯状态,文中设计了一种基于K码控制字符的通讯协议。创建标志用户数据帧起始的动态SOF(Start of Frame)和标志结束的静态EOF(End of Frame)两种K码控制字符,有利于通讯的连续性检测。创建TLINK(Transmit Link)、BLINK(Back Link)的K码控制字符,其中TLINK控制字符在Serdes的发送端进行定期发送,接收端收到TLINK控制字符后,控制本方的Serdes发送端优先输出BLINK控制字符进行应答,以建立通讯双方之间的握手关系,有利于通讯的超时和状态检测。校验独立于SOF、EOF之间的用户数据,进行CRC32(Cyclic Redundancy Check32)计算,有利于通讯的误码检测。实验结果表明,该协议可实现对Serdes链路的丢帧数量、误码数量、超时数量及通讯断开时长进行准确监控,最小时间精度为10μs。展开更多
基金supported by the National Natural Science Foundation of China(Nos.12172248,12302022,12021002,and 12132010)the Tianjin Research Program of Application Foundation and Advanced Technology of China(No.23JCZDJC00950)。
文摘Belt conveyors are extensively utilized in mining and power industries.In a typical coal mine conveyor system,coal is transported over distances exceeding 2 km,involving more than 20000 idlers,which far exceeds a reasonable manual inspection capacity.Given that idlers typically have a lifespan of 1-2 years,there is an urgent need for a rapid,cost-effective,and intelligent safety monitoring system.However,current embedded systems face prohibitive replacement costs,while conventional monitoring technologies suffer from inefficiency at low rotational speeds and lack systematic structural optimization frameworks for diverse idler types and parameters.To address these challenges,this paper introduces an integrated,on-site detachable self-powered idler condition monitoring system(ICMS).This system combines energy harvesting based on the magnetic modulation technology with wireless condition monitoring capabilities.Specifically,it develops a data-driven model integrating convolutional neural networks(CNNs) with genetic algorithms(GAs).The conventional testing results show that the data-driven model not only significantly accelerates the parameter response time,but also achieves a prediction accuracy of 92.95%.The in-situ experiments conducted in coal mines demonstrate the system's reliability and monitoring functionality under both no-load and fullload conditions.This research provides an innovative self-powered condition monitoring solution and develops an efficient data-driven model,offering feasible online monitoring approaches for smart mine construction.
基金supported by the National Key Research and Development Program of China (Grant No.2019YFB1803700)the Key Technologies Research and Development Program of Tianjin (Grant No.20YFZCGX00440).
文摘A designed visual geometry group(VGG)-based convolutional neural network(CNN)model with small computational cost and high accuracy is utilized to monitor pulse amplitude modulation-based intensity modulation and direct detection channel performance using eye diagram measurements.Experimental results show that the proposed technique can achieve a high accuracy in jointly monitoring modulation format,probabilistic shaping,roll-off factor,baud rate,optical signal-to-noise ratio,and chromatic dispersion.The designed VGG-based CNN model outperforms the other four traditional machine-learning methods in different scenarios.Furthermore,the multitask learning model combined with MobileNet CNN is designed to improve the flexibility of the network.Compared with the designed VGG-based CNN,the MobileNet-based MTL does not need to train all the classes,and it can simultaneously monitor single parameter or multiple parameters without sacrificing accuracy,indicating great potential in various monitoring scenarios.
文摘Multi-functioning in virtual monitoring and assessment of ultimate dynamics of thin-walled bridges is treated in present paper. The approach enables multiple functions in virtual monitoring of the bridges made of integrated thin-walled members with their hierarchical configuration. Theoretical, numerical and experimental in situ assessments of the problem are presented. Some results of structural application are submitted.
基金Supported by Program for the National Natural Science Foundation of China(No.52071053,U1704253)the Fundamental Research Funds for the Central Universities(DUT20GF111)the China Postdoctoral Science Foundation(2020M670748,2020M680946).
文摘High-temperature electromagnetic(EM) protection materials integrated of multiple EM protection mechanisms and functions are regarded as desirable candidates for solving EM interference over a wide temperature range.In this work,a novel microwave modulator is fabricated by introducing carbonyl iron particles(CIP)/resin into channels of carbonized wood(C-wood).Innovatively,the spaced arrangement of two microwave absorbents not only achieves a synergistic enhancement of magnetic and dielectric losses,but also breaks the translational invariance of EM characteristics in the horizontal direction to obtain multiple phase discontinuities in the frequency range of 8.2-18.0 GHz achieving modulation of reflected wave radiation direction.Accordingly,CIP/C-wood microwave modulator demonstrates the maximum effective bandwidth of 5.2 GHz and the maximum EM protection efficiency over 97% with a thickness of only 1.5 mm in the temperature range 298-673 K.Besides,CIP/C-wood microwave modulator shows stable and low thermal conductivities,as well as monotonic electrical conductivity-temperature characteristics,therefore it can also achieve thermal infrared stealth and working temperature monitoring in wide temperature ranges.This work provides an inspiration for the design of high-temperature EM protection materials with multiple EM protection mechanisms and functions.
基金supported by the National Key Research and Development Program of China(No.2022YFE0138400)Zhejiang Provincial Key R&D Program Project(No.2023C01061).
文摘Dynamic temperature monitoring at critical locations of IGBT modules is a key means to improve the reliability of high-power converters.However,most ex-isting thermal model-based methods suffer from temper-ature estimation errors due to model parameter varia-tions and loss calculation errors.To address this problem,based on the reduced-order thermal model,an H_(∞)ob-server-based robust 3-D thermal monitoring method for IGBT modules is proposed in this paper.Through the optimized design of the observer feedback gain,the thermal model and real-time temperature information are effectively combined,which reduces the temperature estimation error in the worst case.Thus,the proposed method is more robust to model parameter uncertainty and loss error than the conventional temperature ob-servers.Experiment validations of the proposed H_(∞)ob-server and conventional observers are provided.The results demonstrate that the proposed observer achieves the highest temperature estimation accuracy under vari-ous system uncertainties,making it an effective solution for reliable online thermal monitoring of IGBT modules over the whole life cycle.
基金funded by the Enterprise Ireland Innovation Partnership Programme with Ericsson under grant agreement IP/2011/0135[6]supported by the National Natural Science Foundation of China(No.61373131,61303039,61232016,61501247)+1 种基金the PAPDCICAEET funds
文摘The rapid development of network technology and its evolution toward heterogeneous networks has increased the demand to support automatic monitoring and the management of heterogeneous wireless communication networks.This paper presents a multilevel pattern mining architecture to support automatic network management by discovering interesting patterns from telecom network monitoring data.This architecture leverages and combines existing frequent itemset discovery over data streams,association rule deduction,frequent sequential pattern mining,and frequent temporal pattern mining techniques while also making use of distributed processing platforms to achieve high-volume throughput.
基金Supported by the National Natural Science Foundation of China (60978007 61027007 61177067)
文摘A technique using artificial neural networks trained with parameters derived from delay tap plots for optical performance monitoring in 40 Gbit/s duobinary system is demonstrated. Firstly, the optical signal is delay tap sampled to obtain two-dimensional histogram, known as delay tap plots. Secondly, the features of delay tap plots are extracted to train the feed forward, three-layer preceptor structure artificial neural networks. Finally, the outputs of trained neural network are used to monitor optical duobinary signal impairments. Simulation of optical signal noise ratio （ OSNR）, chromatic dispersion （CD）, and differential group delay （DGD） monitoring in 40 Gbit/s optical duo- binary system is presented. The proposed monitoring scheme can accurately identify simultaneous impairments without requiring synchronous sampling or data clock recovery. The proposed technique is simple, cost-effective and suitable for in-service distributed OPM.
基金supported by the National Natural Science Foundation of China (No.31800631)the Natural Science Foundation of Guangxi Province (No.2018JJB120049)+1 种基金the Middleaged and Young Teachers’ Basic Ability Promotion Project of Guangxi (No.2018KY0361)the BAGUI Scholar Program of Guangxi Province of China。
文摘A signal-amplified mercury sensing biosensor with desired sensitivity was developed through firstly using the GFP mutant with fluorescence increasing response towards Hg^2+ as the reporter module.The developed biosensor showed response for Hg^2+ in a relatively wide range of 1–10,000 nmol/L,and the detection limit was improved one or two orders of magnitude in comparison with most metal-sensing biosensors in similar constructs.In addition,the biosensor could distinguish Hg^2+ easily from multiple metal ions and displayed strong adaptability to extensive p H conditions (pH 4.0–10.0).More importantly,the developed biosensor was able to provide an initial assessment of Hg^2+ spiked in the environmental water with the recoveries between 85.70%and 112.50%.The signal-amplified strategy performed by the modified reporter module will be widely applicable to many other wholecell biosensors,meeting the practical requirements with sufficient sensing performance.
基金supported by the National Natural Science Foundation of China(61771393 61571368)
文摘Many safety-critical applications that utilize the global navigation satellite system (GNSS) demand highly accurate positioning information, as well as highly integrity and reliability. Due to GNSS signals are easily distorted by the interferences or disturbances, the signal quality monitoring (SQM) is necessary to detect the presence of dangerous signal distortions. In this paper, we developed an SQM software for binary offset carrier (BOC) modulated navigation signals. Firstly, the models of BOC signal with ideal and distortion are presented respectively. Then the architecture of SQM software is proposed. Moreover, the effect of the white gaussian noise (WGN) and the front-end filter on the correlation peak of the receiver is analyzed. Finally, the biases induced by the signal distortion are evaluated. The experiments simulate the relationships between the code phase shift and the normalized correlation value in the case of the signal digital distortion and the analog distortion. The simulation results demonstrate that the proposed SQM method can effectively monitor the signal distortion and accurately estimate the correlation peak deviation caused by the distortion.
基金supported by the National Key R&D Plan(No.2016YFA0401900)
文摘A bunch arrival-time monitor(BAM) based on an electro-optical intensity modulation scheme is currently under development at Shanghai Soft X-ray Free-Electron Laser to meet the high-resolution requirements for bunch stability. The BAM uses a radio frequency signal generated by a pickup cavity to modulate the reference laser pulses in an electro-optical intensity modulator(EOM), and the bunch arrival-time information is derived from the amplitude change of the laser pulse after laser pulse modulation.EOM is a key optical component in the BAM system.Through the basic principle analysis of BAM, many parameters of the EOM are observed to affect the measurement resolution of the BAM system. Therefore, a systematic analysis of the EOM is crucial. In this paper, we present two schemes to compare and analyze an EOM and provide a reference for selecting a new version of the EOM.
文摘Pursuing small critical dimensions(i.e.14 nm or below)and high integration bring us lots of physical defects causing low yield and functionality failures for foundries.Under this circumstance,inspection,metrology and monitoring technologies are unprecedentedly vital for development of semiconductor industry.Optical and electron beam solutions are the most common two methods in semiconductor manufacturing.Hamamatsu Photonics is now aiming at optical inspection,metrology and monitoring systems market by providing light sources,photodetectors and failure analysis systems for semiconductor equipment manufacturers,foundries and research institutions.In this paper,features and potential applications of light sources,photodetectors(like image sensors,photomultiplier tubes/modules,silicon photomultipliers(modules),(avalanche)photodiodes(arrays)and so on),with the wavelengths ranging from UV to Infrared,are mainly discussed.In addition,Hamamatsu’s star product– failure analysis system to quickly locate faults or defects are introduced.In conclusion,Hamamatsu Photonics is dedicated to develop large varieties of light sources and optical sensors/detector/modules along with failure analysis systems and willing to improve the development of semiconductor and related industries,especially in China.
文摘非侵入式负荷监测(non-intrusive load monitoring,NILM)技术对于实现智慧用电与管理具有重要意义。针对现有的非侵入式负荷监测方法在高噪声环境下对特征相似电器以及微小负荷变化监测精度不足的难题,提出了一种基于单位力操作视觉变换器的非侵入式负荷监测(non-intrusive load monitoring based on unit force operated vision transformer,UFONILM)模型的非侵入式负荷监测的深度学习框架。UFONILM模型的单位力操作(unit force operated,UFO)模块通过层归一化和一系列卷积层有效地提取和利用了多尺度的时间序列数据,特征。在标准的UK-DALE数据集上进行的实验显示,UFONILM模型在准确性和F1得分上均优于现有方法,特别是在细粒度的负荷监测场景中。研制了基于UFONILM模型的嵌入式系统,实现了边缘计算的非侵入式负荷监测,可实时监测和响应电网中的异常用电行为,如违规充电事件。实验检测证明,UFONILM模型嵌入式非侵入式负荷监测方法在监测效率方面具有显著的提升,具有高效、便捷安装、可扩展等特点。
