Centrifugal Pumps(CPs)are critical machine components in many industries,and their efficient operation and reliable Fault Diagnosis(FD)are essential for minimizing downtime and maintenance costs.This paper introduces ...Centrifugal Pumps(CPs)are critical machine components in many industries,and their efficient operation and reliable Fault Diagnosis(FD)are essential for minimizing downtime and maintenance costs.This paper introduces a novel FD method to improve both the accuracy and reliability of detecting potential faults in such pumps.Theproposed method combinesWaveletCoherent Analysis(WCA)and Stockwell Transform(S-transform)scalograms with Sobel and non-local means filters,effectively capturing complex fault signatures from vibration signals.Using Convolutional Neural Network(CNN)for feature extraction,the method transforms these scalograms into image inputs,enabling the recognition of patterns that span both time and frequency domains.The CNN extracts essential discriminative features,which are then merged and passed into a Kolmogorov-Arnold Network(KAN)classifier,ensuring precise fault identification.The proposed approach was experimentally validated on diverse datasets collected under varying conditions,demonstrating its robustness and generalizability.Achieving classification accuracy of 100%,99.86%,and 99.92%across the datasets,this method significantly outperforms traditional fault detection approaches.These results underscore the potential to enhance CP FD,providing an effective solution for predictive maintenance and improving overall system reliability.展开更多
Changes in surface temperature extremes have become a global concern.Based on the daily lowest temperature(TN)and daily highest temperature(TX)data from 2138weather stations in China from 1961 to 2020,we calculated 14...Changes in surface temperature extremes have become a global concern.Based on the daily lowest temperature(TN)and daily highest temperature(TX)data from 2138weather stations in China from 1961 to 2020,we calculated 14 extreme temperature indices to analyze the characteristics of extreme temperature events.The widespread changes observed in all extreme temperature indices suggest that China experienced significant warming during this period.Specifically,the cold extreme indices,such as cold nights,cold days,frost days,icing days,and the cold spell duration index,decreased significantly by-6.64,-2.67,-2.96,-0.97,and-1.01 days/decade,respectively.In contrast,we observed significant increases in warm extreme indices.The number of warm nights,warm days,summer days,tropical nights,and warm spell duration index increased by 8.44,5.18,2.81,2.50,and 1.66d/decade,respectively.In addition,the lowest TN,highest TN,lowest TX,and highest TX over the entire period rose by 0.47,0.22,0.26,and 0.16℃/decade,respectively.Furthermore,using Pearson's correlation and wavelet coherence analyses,this study identified a strong association between extreme temperature indices and atmospheric circulation factors,with varying correlation strengths and resonance periods across different time-frequency domains.展开更多
The pressure and temperature significantly influence precipitable water vapor(PWV) retrieval. Global Navigation Satellite System(GNSS) PWV retrieval is limited because the GNSS stations lack meteorological sensors. Fi...The pressure and temperature significantly influence precipitable water vapor(PWV) retrieval. Global Navigation Satellite System(GNSS) PWV retrieval is limited because the GNSS stations lack meteorological sensors. First, this article evaluated the accuracy of pressure and temperature in 68 radiosonde stations in China based on ERA5 Reanalysis data from 2015 to 2019 and compared them with GPT3model. Then, the accuracy of pressure and temperature calculated by ERA5 were estimated in 5 representative IGS stations in China. And the PWV calculated by these meteorological parameters from ERA5(ERA5-PWV) were analyzed. Finally, the relation between ERA5-PWV and precipitation was deeply explored using wavelet coherence analysis in IGS stations. These results indicate that the accuracy of pressure and temperature of ERA5 is better than the GPT3 model. In radiosonde stations, the mean BIAS and MAE of pressure and temperature in ERA5 are-0.41/1.15 hpa and-0.97/2.12 K. And the mean RMSEs are 1.35 hpa and 2.87 K, which improve 74.77% and 40.58% compared with GPT3 model. The errors of pressure and temperature of ERA5 are smaller than the GPT3 model in bjfs, hksl and wuh2, and the accuracy of ERA5-PWV is improved by 18.77% compared with the GPT3 model. In addition, there is a significant positive correlation between ERA5-PWV and precipitation. And precipitation is always associated with the sharp rise of ERA5-PWV, which provides important references for rainfall prediction.展开更多
基金supported by the Technology Innovation Program(20023566,‘Development and Demonstration of Industrial IoT and AI-Based Process Facility Intelligence Support System in Small and Medium Manufacturing Sites’)funded by the Ministry of Trade,Industry,&Energy(MOTIE,Republic of Korea).
文摘Centrifugal Pumps(CPs)are critical machine components in many industries,and their efficient operation and reliable Fault Diagnosis(FD)are essential for minimizing downtime and maintenance costs.This paper introduces a novel FD method to improve both the accuracy and reliability of detecting potential faults in such pumps.Theproposed method combinesWaveletCoherent Analysis(WCA)and Stockwell Transform(S-transform)scalograms with Sobel and non-local means filters,effectively capturing complex fault signatures from vibration signals.Using Convolutional Neural Network(CNN)for feature extraction,the method transforms these scalograms into image inputs,enabling the recognition of patterns that span both time and frequency domains.The CNN extracts essential discriminative features,which are then merged and passed into a Kolmogorov-Arnold Network(KAN)classifier,ensuring precise fault identification.The proposed approach was experimentally validated on diverse datasets collected under varying conditions,demonstrating its robustness and generalizability.Achieving classification accuracy of 100%,99.86%,and 99.92%across the datasets,this method significantly outperforms traditional fault detection approaches.These results underscore the potential to enhance CP FD,providing an effective solution for predictive maintenance and improving overall system reliability.
基金National Key Research and Development Program of China,No.2021YFB3900900。
文摘Changes in surface temperature extremes have become a global concern.Based on the daily lowest temperature(TN)and daily highest temperature(TX)data from 2138weather stations in China from 1961 to 2020,we calculated 14 extreme temperature indices to analyze the characteristics of extreme temperature events.The widespread changes observed in all extreme temperature indices suggest that China experienced significant warming during this period.Specifically,the cold extreme indices,such as cold nights,cold days,frost days,icing days,and the cold spell duration index,decreased significantly by-6.64,-2.67,-2.96,-0.97,and-1.01 days/decade,respectively.In contrast,we observed significant increases in warm extreme indices.The number of warm nights,warm days,summer days,tropical nights,and warm spell duration index increased by 8.44,5.18,2.81,2.50,and 1.66d/decade,respectively.In addition,the lowest TN,highest TN,lowest TX,and highest TX over the entire period rose by 0.47,0.22,0.26,and 0.16℃/decade,respectively.Furthermore,using Pearson's correlation and wavelet coherence analyses,this study identified a strong association between extreme temperature indices and atmospheric circulation factors,with varying correlation strengths and resonance periods across different time-frequency domains.
基金supported by Open Fund of Key Laboratory of Mine Environmental Monitoring and Improving around Poyang Lake of Ministry of Natural Resources (Grant MEMI-2021-2022-27)funded by the National Natural Science Foundation of China (Grants 41904031,42374040,42061077)+2 种基金the Jiangxi Provincial Natural Science Foundation (Grants 20202BABL213033)the State Key Laboratory of Geodesy and Earth's Dynamics (Grants SKLGED2021-2-2)the Graduate Innovation Foundation of East China University of Technology (Grants YC2022-s604,YC2022-s609)。
文摘The pressure and temperature significantly influence precipitable water vapor(PWV) retrieval. Global Navigation Satellite System(GNSS) PWV retrieval is limited because the GNSS stations lack meteorological sensors. First, this article evaluated the accuracy of pressure and temperature in 68 radiosonde stations in China based on ERA5 Reanalysis data from 2015 to 2019 and compared them with GPT3model. Then, the accuracy of pressure and temperature calculated by ERA5 were estimated in 5 representative IGS stations in China. And the PWV calculated by these meteorological parameters from ERA5(ERA5-PWV) were analyzed. Finally, the relation between ERA5-PWV and precipitation was deeply explored using wavelet coherence analysis in IGS stations. These results indicate that the accuracy of pressure and temperature of ERA5 is better than the GPT3 model. In radiosonde stations, the mean BIAS and MAE of pressure and temperature in ERA5 are-0.41/1.15 hpa and-0.97/2.12 K. And the mean RMSEs are 1.35 hpa and 2.87 K, which improve 74.77% and 40.58% compared with GPT3 model. The errors of pressure and temperature of ERA5 are smaller than the GPT3 model in bjfs, hksl and wuh2, and the accuracy of ERA5-PWV is improved by 18.77% compared with the GPT3 model. In addition, there is a significant positive correlation between ERA5-PWV and precipitation. And precipitation is always associated with the sharp rise of ERA5-PWV, which provides important references for rainfall prediction.
