Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain limitations.Conversely,deep learning techniques have gained promine...Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain limitations.Conversely,deep learning techniques have gained prominence as a central focus of research in the field of fault diagnosis by strong fault feature extraction ability and end-to-end fault diagnosis efficiency.Recently,utilizing the respective advantages of convolution neural network(CNN)and Transformer in local and global feature extraction,research on cooperating the two have demonstrated promise in the field of fault diagnosis.However,the cross-channel convolution mechanism in CNN and the self-attention calculations in Transformer contribute to excessive complexity in the cooperative model.This complexity results in high computational costs and limited industrial applicability.To tackle the above challenges,this paper proposes a lightweight CNN-Transformer named as SEFormer for rotating machinery fault diagnosis.First,a separable multiscale depthwise convolution block is designed to extract and integrate multiscale feature information from different channel dimensions of vibration signals.Then,an efficient self-attention block is developed to capture critical fine-grained features of the signal from a global perspective.Finally,experimental results on the planetary gearbox dataset and themotor roller bearing dataset prove that the proposed framework can balance the advantages of robustness,generalization and lightweight compared to recent state-of-the-art fault diagnosis models based on CNN and Transformer.This study presents a feasible strategy for developing a lightweight rotating machinery fault diagnosis framework aimed at economical deployment.展开更多
Text in natural scene images usually carries abundant semantic information. However, due to variations of text and complexity of background, detecting text in scene images becomes a critical and challenging task. In t...Text in natural scene images usually carries abundant semantic information. However, due to variations of text and complexity of background, detecting text in scene images becomes a critical and challenging task. In this paper, we present a novel method to detect text from scene images. Firstly, we decompose scene images into background and text components using morphological component analysis(MCA), which will reduce the adverse effects of complex backgrounds on the detection results.In order to improve the performance of image decomposition,two discriminative dictionaries of background and text are learned from the training samples. Moreover, Laplacian sparse regularization is introduced into our proposed dictionary learning method which improves discrimination of dictionary. Based on the text dictionary and the sparse-representation coefficients of text, we can construct the text component. After that, the text in the query image can be detected by applying certain heuristic rules. The results of experiments show the effectiveness of the proposed method.展开更多
Construction of two Ru^(Ⅲ)cations and six lacunary Keggin fragments resulted in a novel Ru_(2)W_(12)-cluster{(RuO_(6))_(2)(WO_(3))_(12)(H_(2)O)_(12)}bridged polyoxometalate,NaH_(11)[(RuO_(6))(AsW_(9)O_(33))_(3){(W_(6...Construction of two Ru^(Ⅲ)cations and six lacunary Keggin fragments resulted in a novel Ru_(2)W_(12)-cluster{(RuO_(6))_(2)(WO_(3))_(12)(H_(2)O)_(12)}bridged polyoxometalate,NaH_(11)[(RuO_(6))(AsW_(9)O_(33))_(3){(W_(6)O_(3))(H_(2)O)_(6)}]_(2)53H_(2)O(NaH_(11)·1·53H_(2)O),which represent the largest cluster in all the Ru-containing polyoxometalates.The most interesting characteristic is that the symmetry-related Ru_(2)W_(12)-cluster-based hexamers contain two windmill-shaped[(RuO_(6))(AsW_(9)O_(33))_(3){(W_(6)O_(3))(H_(2)O)_(6)}]trimers or the Ru_(2)W_(12) cluster was tightly wrapped by six segments of B-β-AsW_(9)O_(33).The other remarkable feature is that there have one intriguing cubane structure:which is composed of the Ru(1,2)and W(1,28,50,51,52,53)atoms.The oxygenation reactions of anilines to azoxybenzenes was evaluated when NaH_(11)·1·53H_(2)O served as effective catalyst by probing various reaction.The inherent redox property of oxygen-rich polyoxometalate surfaces and high photocatalytic activity of the Ru-containing metal cluster imbedded in NaH_(11)·1·53H_(2)O provide sufficient driving force for the photocatalytic transformation from anilines to azoxybenzenes.The oxidation of anilines can be realized with higher selectivity to afford various azoxybenzene compounds.The durability test shows that Ru-doping catalyst displays excellent chemical stability during the photocatalytic process.展开更多
Ultrasonic peen forming(UPF)is an emerging technology that exhibits great superiority in both its flexible operating modes and the deep residual stress that it produces compared with conventional plastic forming metho...Ultrasonic peen forming(UPF)is an emerging technology that exhibits great superiority in both its flexible operating modes and the deep residual stress that it produces compared with conventional plastic forming methods.Although ultrasonic transducers with longitudinal vibration have been widely studied,they have seldom been incorporated into UPF devices for machining in confined spaces.To meet the requirements of this type of machining,a sandwich-type piezoelectric transducer with coupled longitudinal-flexural vibrational modes is proposed.The basic structure of the transducer is designed to obtain large vibrational amplitudes in both modes.Experimental results obtained with a prototype device demonstrate the feasibility of the proposed transducer.The measured vibrational amplitude for the working face in the longitudinal vibrational mode is 1.0μm,and electrical matching increases this amplitude by 40%.The flexural vibration characteristics of the same prototype transducer are also tested and are found to be slightly smaller than those of longitudinal mode.The resultant working strokes of the UPF impact pins reach 1.7 mm and 1.2 mm in the longitudinal and flexural modes,respectively.The forming capability of the prototype has been evaluated via 15-min machining on standard 2024-T351 aluminum plates.After UPF,an improved surface morphology with lower surface roughness is obtained.The aluminum plate test piece has an apparent upper deformation with an arc height of 0.64 mm.The measured peak value of the compressive residual stress is around 250 MPa,appearing at a depth of 100μm.The proposed longitudinal-flexural hybrid transducer thus provides a high-performance tool for plate peen forming in confined spaces.展开更多
Introduction The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy,precisely measure oscillation parameters and study solar neutrinos,su...Introduction The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy,precisely measure oscillation parameters and study solar neutrinos,supernova neutrinos and geo-neutrinos,etc.The central detector(CD)of JUNO has 20,000 tons liquid scintillator as target mass,which contains inside a huge acrylic sphere with inner diameter of 35.4 m,supported by a stainless steel structure.The whole structure of CD will be installed inside a cylindrical water pool,and the acrylic sphere will be submerged in the center of water pool.The operating temperature of CD is designed to be 21℃ as long as over 20 years,which is determined by the mechanical requirement of the structure and physics consideration.Methods For this operating temperature,a special cooling system will be used to maintain the temperature inside the water pool.The main structure of CD is composed of acrylic and stainless steel,and they have much different thermal expansion coefficients,strengths and life times.Change in temperature may affect the safety of CD.As part of reliability analysis,the effect of cooling system failure on the CD is considered,and finite element method is used in our thermal calculation.In this article,the temperature fields before and after cooling system failure are calculated and analyzed,and the temperatures of different locations of water pool after cooling system failure are compared and discussed in detail.展开更多
Soft silicone films have garnered a great deal of interest for use in dielectric elastomer transducers due to their excellent properties,including high elongation to rupture,low viscoelasticity,and broad application t...Soft silicone films have garnered a great deal of interest for use in dielectric elastomer transducers due to their excellent properties,including high elongation to rupture,low viscoelasticity,and broad application temperature range.However,silicone films generally have higher stiffness and lower dielectric strength than VHB acrylic elastomers,which limits the achievable actuation strain.Devices based on silicone dielectric elastomers always experience high rates of premature dielectric failure when operated at high strains.The premature failure is characterized by the loss of functionality or mechanical rupture of the material when operated below the material’s dielectric strength and elongation to rupture.The use is reported of ultrathin coatings of single-walled carbon nanotubes(SWNTs)as the compliant electrodes,which can overcome the issue of premature failure.The self-clearing of the SWNT electrodes in the event of localized dielectric breakdown improves the apparent dielectric strength of the material by isolating the regions of reduced dielectric strength.The actuators may be operated at higher than 50%area strain with reasonably long lifetimes.High strains were measured between-40 and 80℃and in a broad frequency range up to 100 Hz.The fault tolerance introduced by the SWNT electrodes should broaden the application scope of silicone dielectric elastomers.展开更多
Background The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy and precisely measure oscillation parameters and study the solar neutri...Background The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy and precisely measure oscillation parameters and study the solar neutrino,supernova neutrino,geo-neutrino,etc.JUNO's central detector(CD)has 20 kilo-ton liquid scintillator as target mass,which is contained by a huge acrylic sphere with the inner diameter of 35.4 m,and the acrylic sphere is supported by a stainless steel structure through 590 connecting bars.Motivation Part of the connecting bars bear pull force and the other bars bear push force.There is a direct relationship between the stress of connecting bars and that of acrylic sphere.For the installation process of the CD,the pretightening force and axial force of the connecting bars should be monitored with accuracy,and for the filling process and running condition,the precise measurement of axial force can indicate the safety of structure of the CD.Methods Statistical method was used to evaluate the performance of measurement schemes,and 4-fiber Bragg grating measurement scheme was determined to be the final scheme,which can get the measurement uncertainty better than 0.7 kN.Performance of different measurement schemes are discussed in detail,and some related finite element analysis and evalu-ation method are also introduced in this paper.展开更多
Background The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy,precisely measure oscillation parameters and study solar neutrinos,supe...Background The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy,precisely measure oscillation parameters and study solar neutrinos,supernova neutrinos and geo-neutrinos.The JUNO Central Detector 20 kton liquid scintillator target mass is contained by a huge acrylic sphere with a 35.4 m inner diameter,supported by a stainless steel structure,and the sphere is eventually submerged in pure water.Motivation Before the JUNO Central Detector is built,a small prototype has been designed and static loading experiments will be carried out to verify the consistency of the finite-element calculations and static loading experiments and test the subsystems performances,such as the monitoring system and liquid scintillator filling system.Methods The small prototype is composed of an acrylic sphere with a 3 m inner diameter and an aluminum alloy support structure.In this article,the structure of the small prototype is briefly described.A detailed simulation study using finite element analysis is conducted to account for liquid-filled condition,temperature variation and existence of high load(due to vacuum pumping)and demonstrate a satisfying mechanical performance of the small prototype.The experimental plan about the prototype on the basis of the simulation will also be mentioned.展开更多
This study aimed to find out the blood data characteristics of patients and explore the correlation between severe preeclampsia and blood index value.Provide assistance for the early attention direction of severe pree...This study aimed to find out the blood data characteristics of patients and explore the correlation between severe preeclampsia and blood index value.Provide assistance for the early attention direction of severe preeclampsia diagnosis and treatment.19,653 pregnant women presenting to the West China Second University Hospital,Sichuan University from January 2017 to April 2019.After screening,a total of 248 patients,124 severe preeclampsia cases,and 124 controls were selected for this study.Forty-three blood examination variables were obtained from routine blood work,hepatic,renal and coagulation function examination.Light gradient boosting machine(light GBM),decision tree and random forest were used for date diving.We randomly divided 35%of the original data as a testing set to conduct internal validation of the performance of the prediction model.The area under receiver operating characteristic curve(AUC)was used as the main score to compare the three methods.Finally,a binary classification light GBM model based on aspartate aminotransferase,direct bilirubin and activated partial thromboplastin time ratio can predict severe preeclampsia with sensitivity of 88.37%,specificity of 77.27%,AUC of 89.74%and positive predictive value of 65.96%.We believe relevant quantifiable indicators can establish an effective prediction model,which can provide guidance for early detection and prevention of severe preeclampsia.展开更多
基金supported by the National Natural Science Foundation of China(No.52277055).
文摘Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain limitations.Conversely,deep learning techniques have gained prominence as a central focus of research in the field of fault diagnosis by strong fault feature extraction ability and end-to-end fault diagnosis efficiency.Recently,utilizing the respective advantages of convolution neural network(CNN)and Transformer in local and global feature extraction,research on cooperating the two have demonstrated promise in the field of fault diagnosis.However,the cross-channel convolution mechanism in CNN and the self-attention calculations in Transformer contribute to excessive complexity in the cooperative model.This complexity results in high computational costs and limited industrial applicability.To tackle the above challenges,this paper proposes a lightweight CNN-Transformer named as SEFormer for rotating machinery fault diagnosis.First,a separable multiscale depthwise convolution block is designed to extract and integrate multiscale feature information from different channel dimensions of vibration signals.Then,an efficient self-attention block is developed to capture critical fine-grained features of the signal from a global perspective.Finally,experimental results on the planetary gearbox dataset and themotor roller bearing dataset prove that the proposed framework can balance the advantages of robustness,generalization and lightweight compared to recent state-of-the-art fault diagnosis models based on CNN and Transformer.This study presents a feasible strategy for developing a lightweight rotating machinery fault diagnosis framework aimed at economical deployment.
基金supported in part by the National Natural Science Foundation of China(61302041,61363044,61562053,61540042)the Applied Basic Research Foundation of Yunnan Provincial Science and Technology Department(2013FD011,2016FD039)
文摘Text in natural scene images usually carries abundant semantic information. However, due to variations of text and complexity of background, detecting text in scene images becomes a critical and challenging task. In this paper, we present a novel method to detect text from scene images. Firstly, we decompose scene images into background and text components using morphological component analysis(MCA), which will reduce the adverse effects of complex backgrounds on the detection results.In order to improve the performance of image decomposition,two discriminative dictionaries of background and text are learned from the training samples. Moreover, Laplacian sparse regularization is introduced into our proposed dictionary learning method which improves discrimination of dictionary. Based on the text dictionary and the sparse-representation coefficients of text, we can construct the text component. After that, the text in the query image can be detected by applying certain heuristic rules. The results of experiments show the effectiveness of the proposed method.
基金supported bythe National Natural Science Foundationof China(Nos.22171071,22071044,21771054,21571050)。
文摘Construction of two Ru^(Ⅲ)cations and six lacunary Keggin fragments resulted in a novel Ru_(2)W_(12)-cluster{(RuO_(6))_(2)(WO_(3))_(12)(H_(2)O)_(12)}bridged polyoxometalate,NaH_(11)[(RuO_(6))(AsW_(9)O_(33))_(3){(W_(6)O_(3))(H_(2)O)_(6)}]_(2)53H_(2)O(NaH_(11)·1·53H_(2)O),which represent the largest cluster in all the Ru-containing polyoxometalates.The most interesting characteristic is that the symmetry-related Ru_(2)W_(12)-cluster-based hexamers contain two windmill-shaped[(RuO_(6))(AsW_(9)O_(33))_(3){(W_(6)O_(3))(H_(2)O)_(6)}]trimers or the Ru_(2)W_(12) cluster was tightly wrapped by six segments of B-β-AsW_(9)O_(33).The other remarkable feature is that there have one intriguing cubane structure:which is composed of the Ru(1,2)and W(1,28,50,51,52,53)atoms.The oxygenation reactions of anilines to azoxybenzenes was evaluated when NaH_(11)·1·53H_(2)O served as effective catalyst by probing various reaction.The inherent redox property of oxygen-rich polyoxometalate surfaces and high photocatalytic activity of the Ru-containing metal cluster imbedded in NaH_(11)·1·53H_(2)O provide sufficient driving force for the photocatalytic transformation from anilines to azoxybenzenes.The oxidation of anilines can be realized with higher selectivity to afford various azoxybenzene compounds.The durability test shows that Ru-doping catalyst displays excellent chemical stability during the photocatalytic process.
基金supported by the National Natural Science Foundation of China(Grant Nos.51975278 and 52277055)the Qing Lan Project,the Research Fund of the State Key Laboratory of Mechanics and Control of Mechanical Structures(Nanjing University of Aeronautics and Astronautics)under Grant No.MCMS-I-0321G01+2 种基金the Biomedical Engineering Fusion Laboratory of the affiliated Jiangning Hospital of Nanjing Medical University(Grant No.JNYYZXKY202217)the Postgraduate Research&Practice Innovation Program of NUAA(Grant Nos.xcxjh20220114 and xcxjh20220111)the Postgraduate Research&Practice Innovation Program of Jiangsu Province(Grant No.KYCX23_0353).
文摘Ultrasonic peen forming(UPF)is an emerging technology that exhibits great superiority in both its flexible operating modes and the deep residual stress that it produces compared with conventional plastic forming methods.Although ultrasonic transducers with longitudinal vibration have been widely studied,they have seldom been incorporated into UPF devices for machining in confined spaces.To meet the requirements of this type of machining,a sandwich-type piezoelectric transducer with coupled longitudinal-flexural vibrational modes is proposed.The basic structure of the transducer is designed to obtain large vibrational amplitudes in both modes.Experimental results obtained with a prototype device demonstrate the feasibility of the proposed transducer.The measured vibrational amplitude for the working face in the longitudinal vibrational mode is 1.0μm,and electrical matching increases this amplitude by 40%.The flexural vibration characteristics of the same prototype transducer are also tested and are found to be slightly smaller than those of longitudinal mode.The resultant working strokes of the UPF impact pins reach 1.7 mm and 1.2 mm in the longitudinal and flexural modes,respectively.The forming capability of the prototype has been evaluated via 15-min machining on standard 2024-T351 aluminum plates.After UPF,an improved surface morphology with lower surface roughness is obtained.The aluminum plate test piece has an apparent upper deformation with an arc height of 0.64 mm.The measured peak value of the compressive residual stress is around 250 MPa,appearing at a depth of 100μm.The proposed longitudinal-flexural hybrid transducer thus provides a high-performance tool for plate peen forming in confined spaces.
基金This work is supported by the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA100102).
文摘Introduction The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy,precisely measure oscillation parameters and study solar neutrinos,supernova neutrinos and geo-neutrinos,etc.The central detector(CD)of JUNO has 20,000 tons liquid scintillator as target mass,which contains inside a huge acrylic sphere with inner diameter of 35.4 m,supported by a stainless steel structure.The whole structure of CD will be installed inside a cylindrical water pool,and the acrylic sphere will be submerged in the center of water pool.The operating temperature of CD is designed to be 21℃ as long as over 20 years,which is determined by the mechanical requirement of the structure and physics consideration.Methods For this operating temperature,a special cooling system will be used to maintain the temperature inside the water pool.The main structure of CD is composed of acrylic and stainless steel,and they have much different thermal expansion coefficients,strengths and life times.Change in temperature may affect the safety of CD.As part of reliability analysis,the effect of cooling system failure on the CD is considered,and finite element method is used in our thermal calculation.In this article,the temperature fields before and after cooling system failure are calculated and analyzed,and the temperatures of different locations of water pool after cooling system failure are compared and discussed in detail.
基金financial support from the General Motor Corporation,and the University of California Discovery Program.
文摘Soft silicone films have garnered a great deal of interest for use in dielectric elastomer transducers due to their excellent properties,including high elongation to rupture,low viscoelasticity,and broad application temperature range.However,silicone films generally have higher stiffness and lower dielectric strength than VHB acrylic elastomers,which limits the achievable actuation strain.Devices based on silicone dielectric elastomers always experience high rates of premature dielectric failure when operated at high strains.The premature failure is characterized by the loss of functionality or mechanical rupture of the material when operated below the material’s dielectric strength and elongation to rupture.The use is reported of ultrathin coatings of single-walled carbon nanotubes(SWNTs)as the compliant electrodes,which can overcome the issue of premature failure.The self-clearing of the SWNT electrodes in the event of localized dielectric breakdown improves the apparent dielectric strength of the material by isolating the regions of reduced dielectric strength.The actuators may be operated at higher than 50%area strain with reasonably long lifetimes.High strains were measured between-40 and 80℃and in a broad frequency range up to 100 Hz.The fault tolerance introduced by the SWNT electrodes should broaden the application scope of silicone dielectric elastomers.
基金supported by the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDA100102)
文摘Background The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy and precisely measure oscillation parameters and study the solar neutrino,supernova neutrino,geo-neutrino,etc.JUNO's central detector(CD)has 20 kilo-ton liquid scintillator as target mass,which is contained by a huge acrylic sphere with the inner diameter of 35.4 m,and the acrylic sphere is supported by a stainless steel structure through 590 connecting bars.Motivation Part of the connecting bars bear pull force and the other bars bear push force.There is a direct relationship between the stress of connecting bars and that of acrylic sphere.For the installation process of the CD,the pretightening force and axial force of the connecting bars should be monitored with accuracy,and for the filling process and running condition,the precise measurement of axial force can indicate the safety of structure of the CD.Methods Statistical method was used to evaluate the performance of measurement schemes,and 4-fiber Bragg grating measurement scheme was determined to be the final scheme,which can get the measurement uncertainty better than 0.7 kN.Performance of different measurement schemes are discussed in detail,and some related finite element analysis and evalu-ation method are also introduced in this paper.
文摘Background The Jiangmen Underground Neutrino Observatory(JUNO)is a multipurpose neutrino experiment designed to determine neutrino mass hierarchy,precisely measure oscillation parameters and study solar neutrinos,supernova neutrinos and geo-neutrinos.The JUNO Central Detector 20 kton liquid scintillator target mass is contained by a huge acrylic sphere with a 35.4 m inner diameter,supported by a stainless steel structure,and the sphere is eventually submerged in pure water.Motivation Before the JUNO Central Detector is built,a small prototype has been designed and static loading experiments will be carried out to verify the consistency of the finite-element calculations and static loading experiments and test the subsystems performances,such as the monitoring system and liquid scintillator filling system.Methods The small prototype is composed of an acrylic sphere with a 3 m inner diameter and an aluminum alloy support structure.In this article,the structure of the small prototype is briefly described.A detailed simulation study using finite element analysis is conducted to account for liquid-filled condition,temperature variation and existence of high load(due to vacuum pumping)and demonstrate a satisfying mechanical performance of the small prototype.The experimental plan about the prototype on the basis of the simulation will also be mentioned.
基金National Key R&D Program of China(2018YFC2001800).
文摘This study aimed to find out the blood data characteristics of patients and explore the correlation between severe preeclampsia and blood index value.Provide assistance for the early attention direction of severe preeclampsia diagnosis and treatment.19,653 pregnant women presenting to the West China Second University Hospital,Sichuan University from January 2017 to April 2019.After screening,a total of 248 patients,124 severe preeclampsia cases,and 124 controls were selected for this study.Forty-three blood examination variables were obtained from routine blood work,hepatic,renal and coagulation function examination.Light gradient boosting machine(light GBM),decision tree and random forest were used for date diving.We randomly divided 35%of the original data as a testing set to conduct internal validation of the performance of the prediction model.The area under receiver operating characteristic curve(AUC)was used as the main score to compare the three methods.Finally,a binary classification light GBM model based on aspartate aminotransferase,direct bilirubin and activated partial thromboplastin time ratio can predict severe preeclampsia with sensitivity of 88.37%,specificity of 77.27%,AUC of 89.74%and positive predictive value of 65.96%.We believe relevant quantifiable indicators can establish an effective prediction model,which can provide guidance for early detection and prevention of severe preeclampsia.
