Amidst the swift advancement of new power systems and electric vehicles,inverter-fed machines have progressively materialized as a pivotal apparatus for efficient energy conversion.Stator winding turn insulation failu...Amidst the swift advancement of new power systems and electric vehicles,inverter-fed machines have progressively materialized as a pivotal apparatus for efficient energy conversion.Stator winding turn insulation failure is the root cause of inverter-fed machine breakdown.The online monitoring of turn insulation health can detect potential safety risks promptly,but faces the challenge of weak characteristics of turn insulation degradation.This study proposes an innovative method to evaluate the turn insulation state of inverter-fed machines by utilizing the fractional Fourier transform with a Mel filter(FrFT-Mel).First,the sensitivity of the high-frequency(HF)switching oscillation current to variations in turn insulation was analyzed within the fractional domain.Subsequently,an improved Mel filter is introduced,and its structure and parameters are specifically designed based on the features intrinsic to the common-mode impedance resonance point of the electrical machine.Finally,an evaluation index was proposed for the turn insulation state of inverter-fed machines.Experimental results on a 3kW permanent magnet synchronous machine(PMSM)demonstrate that the proposed FrFT-Mel method significantly enhances the sensitivity of turn insulation state perception by approximately five times,compared to the traditional Fourier transform method.展开更多
A potential concept that could be effective for multiple applications is a“cyber-physical system”(CPS).The Internet of Things(IoT)has evolved as a research area,presenting new challenges in obtaining valuable data t...A potential concept that could be effective for multiple applications is a“cyber-physical system”(CPS).The Internet of Things(IoT)has evolved as a research area,presenting new challenges in obtaining valuable data through environmental monitoring.The existing work solely focuses on classifying the audio system of CPS without utilizing feature extraction.This study employs a deep learning method,CNN-LSTM,and two-way feature extraction to classify audio systems within CPS.The primary objective of this system,which is built upon a convolutional neural network(CNN)with Long Short Term Memory(LSTM),is to analyze the vocalization patterns of two different species of anurans.It has been demonstrated that CNNs,when combined with mel-spectrograms for sound analysis,are suitable for classifying ambient noises.Initially,the data is augmented and preprocessed.Next,the mel spectrogram features are extracted through two-way feature extraction.First,Principal Component Analysis(PCA)is utilized for dimensionality reduction,followed by Transfer learning for audio feature extraction.Finally,the classification is performed using the CNN-LSTM process.This methodology can potentially be employed for categorizing various biological acoustic objects and analyzing biodiversity indexes in natural environments,resulting in high classification accuracy.The study highlights that this CNNLSTM approach enables cost-effective and resource-efficient monitoring of large natural regions.The dissemination of updated CNN-LSTM models across distant IoT nodes is facilitated flexibly and dynamically through the utilization of CPS.展开更多
基金supported in part by the National Natural Science Foundation of China under Grant 51907116in part sponsored by Natural Science Foundation of Shanghai 22ZR1425400sponsored by Shanghai Rising-Star Program 23QA1404000.
文摘Amidst the swift advancement of new power systems and electric vehicles,inverter-fed machines have progressively materialized as a pivotal apparatus for efficient energy conversion.Stator winding turn insulation failure is the root cause of inverter-fed machine breakdown.The online monitoring of turn insulation health can detect potential safety risks promptly,but faces the challenge of weak characteristics of turn insulation degradation.This study proposes an innovative method to evaluate the turn insulation state of inverter-fed machines by utilizing the fractional Fourier transform with a Mel filter(FrFT-Mel).First,the sensitivity of the high-frequency(HF)switching oscillation current to variations in turn insulation was analyzed within the fractional domain.Subsequently,an improved Mel filter is introduced,and its structure and parameters are specifically designed based on the features intrinsic to the common-mode impedance resonance point of the electrical machine.Finally,an evaluation index was proposed for the turn insulation state of inverter-fed machines.Experimental results on a 3kW permanent magnet synchronous machine(PMSM)demonstrate that the proposed FrFT-Mel method significantly enhances the sensitivity of turn insulation state perception by approximately five times,compared to the traditional Fourier transform method.
基金Funded by Institutional Fund Projects under Grant No.IFPIP:236-611-1442 by Ministry of Education and King Abdulaziz University,Jeddah,Saudi Arabia(A.O.A.).
文摘A potential concept that could be effective for multiple applications is a“cyber-physical system”(CPS).The Internet of Things(IoT)has evolved as a research area,presenting new challenges in obtaining valuable data through environmental monitoring.The existing work solely focuses on classifying the audio system of CPS without utilizing feature extraction.This study employs a deep learning method,CNN-LSTM,and two-way feature extraction to classify audio systems within CPS.The primary objective of this system,which is built upon a convolutional neural network(CNN)with Long Short Term Memory(LSTM),is to analyze the vocalization patterns of two different species of anurans.It has been demonstrated that CNNs,when combined with mel-spectrograms for sound analysis,are suitable for classifying ambient noises.Initially,the data is augmented and preprocessed.Next,the mel spectrogram features are extracted through two-way feature extraction.First,Principal Component Analysis(PCA)is utilized for dimensionality reduction,followed by Transfer learning for audio feature extraction.Finally,the classification is performed using the CNN-LSTM process.This methodology can potentially be employed for categorizing various biological acoustic objects and analyzing biodiversity indexes in natural environments,resulting in high classification accuracy.The study highlights that this CNNLSTM approach enables cost-effective and resource-efficient monitoring of large natural regions.The dissemination of updated CNN-LSTM models across distant IoT nodes is facilitated flexibly and dynamically through the utilization of CPS.
