This paper focuses on the new approach of on line monitoring of grinding burn and wheel wear based on the rough set theory. This method adopts the grinding chip flow thermal signal, and acquires identification rules ...This paper focuses on the new approach of on line monitoring of grinding burn and wheel wear based on the rough set theory. This method adopts the grinding chip flow thermal signal, and acquires identification rules by establishing sensitive characteristic parameters and constructing the knowledge system through continuum attribute discretization, attribute reduction and knowledge acquisition, and then monitors grinding burn and wheel wear in accordance with the acquired rules. The experiment results show that the new method is effective.展开更多
On the basis of analyzing the influencing factors and harmfulness of stray current, and discussing the existing problems of monitoring and prevention system for stray current, the integrated monitoring and prevention ...On the basis of analyzing the influencing factors and harmfulness of stray current, and discussing the existing problems of monitoring and prevention system for stray current, the integrated monitoring and prevention system for stray current in metro was developed. A net system of distributed computers for monitoring was set up. It can monitor the distribution of stray current in metro and the corrosion of the metal structure in the whole line. According to the situation of monitoring it can also control the drainage of its tank to reach the best effect and eliminate the negative effect of polarity drainage. By using the new type unilateral electric device, the problem of burning the rail by electric arc can be avoided. The unilateral electric device can be connected with the monitoring net system directly to realize the monitor in line and improve the reliability of the device.展开更多
Power grid is an indispensable infrastructure in modern society,in order to ensure the normal operation of the grid,online non-contact monitoring of high-voltage lines is essential.In this work,a‘capacitor-laser diod...Power grid is an indispensable infrastructure in modern society,in order to ensure the normal operation of the grid,online non-contact monitoring of high-voltage lines is essential.In this work,a‘capacitor-laser diode(LD)-capacitor’structure,namely,laser diode in capacitors(LDIC),that can be used for non-contact monitoring of high-voltage(HV)line status by directly transferring the status information of the HV line to modulated laser pulses is proposed.The proposed LDIC can accurately extract the real-time voltage changes on the HV line with an accuracy level of 0.959%.Because the LDIC is sensitive to high-frequency electromagnetic field,the LDIC is successfully utilised to detect the external electromagnetic interference(EMI)to obtain the intensity and frequency of the external EMI.Additionally,the amplitude and frequency of the HV line vibration can be accurately monitored by using the LDIC.For the third-order curve fitting of vibration amplitude,the average error is only 0.00867,and the average error of linear fitting of vibration frequency is as low as 0.00655.This work provides a novel approach for the online monitoring of the HV line status and a new supplement for the development of power grid technology.展开更多
In recent years,several efforts have been made to develop power transmission line abnormal target detection models based on edge devices.Typically,updates to these models rely on participation of the cloud,which means...In recent years,several efforts have been made to develop power transmission line abnormal target detection models based on edge devices.Typically,updates to these models rely on participation of the cloud,which means that network resource shortages can lead to update failures,followed by unsatisfactory recognition and detection performance in practical use.To address this problem,this article proposes an edge visual incremental perception framework,based on deep semisupervised learning,for monitoring power transmission lines.After generation of the initial model using a small amount of labeled data,models trained using this framework can update themselves based on unlabeled data.A teacher-student joint training strategy,a data augmentation strategy,and a model updating strategy are also designed and adopted to improve the performance of the models trained with this framework.The proposed framework is then examined with various transmission line datasets with 1%,2%,5%,and 10%labeled data.General performance enhancement is thus confirmed against traditional supervised learning strategies.With the 10%labeled data training set,the recognition accuracy of the model is improved to exceed 80%,meeting the practical needs of power system operation,and thus clearly validating the effectiveness of the framework.展开更多
While non-destructive in-line monitoring at manufacturing sites is essential for safe distribution cycles of pharmaceuticals,efforts are still insufficient to develop analytical systems for detailed dynamic visualisat...While non-destructive in-line monitoring at manufacturing sites is essential for safe distribution cycles of pharmaceuticals,efforts are still insufficient to develop analytical systems for detailed dynamic visualisation of foreign substances and material composition in target pills.Although spectroscopies,expected towards pharma testing,have faced technical challenges in in-line setups for bulky equipment housing,this work demonstrates compact dynamic photo-monitoring systems by selectively extracting informative irradiation-wavelengths from comprehensive optical references of target pills.This work develops a non-destructive in-line dynamic inspection system for pharma agent pills with carbon nanotube(CNT)photo-thermoelectric imagers and the associated ultrabroadband sub-terahertz(THz)–infrared(IR)multi-wavelength monitoring.The CNT imager in the proposed system functions in ultrabroadband regions over existing sensors,facilitating multi-wavelength photo-monitoring against external sub-THz–IR-irradiation.Under recent advances in the investigation of functional optical materials(e.g.,gallium arsenide,vanadium oxide,graphene,polymers,transition metal dichalcogenides),CNTs play advantageous leading roles in collectively satisfying informative efficient photo-absorption and solution-processable configurations for printable device fabrication into freely attachable thin-film imagers in pharma monitoring sites.The above non-destructive dynamic monitoring system maintains in-line experimental setups by integrating the functional thin-film imager sheets and compact multiple photo-sources.Furthermore,permeable sub-THz–IR-irradiation,which provides different transmittance values specific to non-metallic materials per wavelength or composition,identifies constituent materials for pharma agents themselves and concealed foreign substances in a non-contact manner.This work finally inspects invisible detailed features of pharma pills with the non-destructive in-line dynamic photo-monitoring system by incorporating performances of CNT imagers and compact optical setups.展开更多
文摘This paper focuses on the new approach of on line monitoring of grinding burn and wheel wear based on the rough set theory. This method adopts the grinding chip flow thermal signal, and acquires identification rules by establishing sensitive characteristic parameters and constructing the knowledge system through continuum attribute discretization, attribute reduction and knowledge acquisition, and then monitors grinding burn and wheel wear in accordance with the acquired rules. The experiment results show that the new method is effective.
文摘On the basis of analyzing the influencing factors and harmfulness of stray current, and discussing the existing problems of monitoring and prevention system for stray current, the integrated monitoring and prevention system for stray current in metro was developed. A net system of distributed computers for monitoring was set up. It can monitor the distribution of stray current in metro and the corrosion of the metal structure in the whole line. According to the situation of monitoring it can also control the drainage of its tank to reach the best effect and eliminate the negative effect of polarity drainage. By using the new type unilateral electric device, the problem of burning the rail by electric arc can be avoided. The unilateral electric device can be connected with the monitoring net system directly to realize the monitor in line and improve the reliability of the device.
基金supported by National Key R&D Program of China under Grant No.2021YFB3600400Mindu Innovation Laboratory Project under Grant No.2020ZZ113.
文摘Power grid is an indispensable infrastructure in modern society,in order to ensure the normal operation of the grid,online non-contact monitoring of high-voltage lines is essential.In this work,a‘capacitor-laser diode(LD)-capacitor’structure,namely,laser diode in capacitors(LDIC),that can be used for non-contact monitoring of high-voltage(HV)line status by directly transferring the status information of the HV line to modulated laser pulses is proposed.The proposed LDIC can accurately extract the real-time voltage changes on the HV line with an accuracy level of 0.959%.Because the LDIC is sensitive to high-frequency electromagnetic field,the LDIC is successfully utilised to detect the external electromagnetic interference(EMI)to obtain the intensity and frequency of the external EMI.Additionally,the amplitude and frequency of the HV line vibration can be accurately monitored by using the LDIC.For the third-order curve fitting of vibration amplitude,the average error is only 0.00867,and the average error of linear fitting of vibration frequency is as low as 0.00655.This work provides a novel approach for the online monitoring of the HV line status and a new supplement for the development of power grid technology.
基金supported by the National Key R&D Program of China (2020YFB0905900).
文摘In recent years,several efforts have been made to develop power transmission line abnormal target detection models based on edge devices.Typically,updates to these models rely on participation of the cloud,which means that network resource shortages can lead to update failures,followed by unsatisfactory recognition and detection performance in practical use.To address this problem,this article proposes an edge visual incremental perception framework,based on deep semisupervised learning,for monitoring power transmission lines.After generation of the initial model using a small amount of labeled data,models trained using this framework can update themselves based on unlabeled data.A teacher-student joint training strategy,a data augmentation strategy,and a model updating strategy are also designed and adopted to improve the performance of the models trained with this framework.The proposed framework is then examined with various transmission line datasets with 1%,2%,5%,and 10%labeled data.General performance enhancement is thus confirmed against traditional supervised learning strategies.With the 10%labeled data training set,the recognition accuracy of the model is improved to exceed 80%,meeting the practical needs of power system operation,and thus clearly validating the effectiveness of the framework.
基金supported by a part of ACT-X(JPMJAX23KL)and Mirai Program(JPMJMI23G1):JST,KAKENHI(JP22H01553,JP22H01555,JP23H00169,JP23K19125,JP24K01288,JP24K17325,and JP25H02154).
文摘While non-destructive in-line monitoring at manufacturing sites is essential for safe distribution cycles of pharmaceuticals,efforts are still insufficient to develop analytical systems for detailed dynamic visualisation of foreign substances and material composition in target pills.Although spectroscopies,expected towards pharma testing,have faced technical challenges in in-line setups for bulky equipment housing,this work demonstrates compact dynamic photo-monitoring systems by selectively extracting informative irradiation-wavelengths from comprehensive optical references of target pills.This work develops a non-destructive in-line dynamic inspection system for pharma agent pills with carbon nanotube(CNT)photo-thermoelectric imagers and the associated ultrabroadband sub-terahertz(THz)–infrared(IR)multi-wavelength monitoring.The CNT imager in the proposed system functions in ultrabroadband regions over existing sensors,facilitating multi-wavelength photo-monitoring against external sub-THz–IR-irradiation.Under recent advances in the investigation of functional optical materials(e.g.,gallium arsenide,vanadium oxide,graphene,polymers,transition metal dichalcogenides),CNTs play advantageous leading roles in collectively satisfying informative efficient photo-absorption and solution-processable configurations for printable device fabrication into freely attachable thin-film imagers in pharma monitoring sites.The above non-destructive dynamic monitoring system maintains in-line experimental setups by integrating the functional thin-film imager sheets and compact multiple photo-sources.Furthermore,permeable sub-THz–IR-irradiation,which provides different transmittance values specific to non-metallic materials per wavelength or composition,identifies constituent materials for pharma agents themselves and concealed foreign substances in a non-contact manner.This work finally inspects invisible detailed features of pharma pills with the non-destructive in-line dynamic photo-monitoring system by incorporating performances of CNT imagers and compact optical setups.
