With the increasing intelligence and integration,a great number of two-valued variables(generally stored in the form of 0 or 1)often exist in large-scale industrial processes.However,these variables cannot be effectiv...With the increasing intelligence and integration,a great number of two-valued variables(generally stored in the form of 0 or 1)often exist in large-scale industrial processes.However,these variables cannot be effectively handled by traditional monitoring methods such as linear discriminant analysis(LDA),principal component analysis(PCA)and partial least square(PLS)analysis.Recently,a mixed hidden naive Bayesian model(MHNBM)is developed for the first time to utilize both two-valued and continuous variables for abnormality monitoring.Although the MHNBM is effective,it still has some shortcomings that need to be improved.For the MHNBM,the variables with greater correlation to other variables have greater weights,which can not guarantee greater weights are assigned to the more discriminating variables.In addition,the conditional P(x j|x j′,y=k)probability must be computed based on historical data.When the training data is scarce,the conditional probability between continuous variables tends to be uniformly distributed,which affects the performance of MHNBM.Here a novel feature weighted mixed naive Bayes model(FWMNBM)is developed to overcome the above shortcomings.For the FWMNBM,the variables that are more correlated to the class have greater weights,which makes the more discriminating variables contribute more to the model.At the same time,FWMNBM does not have to calculate the conditional probability between variables,thus it is less restricted by the number of training data samples.Compared with the MHNBM,the FWMNBM has better performance,and its effectiveness is validated through numerical cases of a simulation example and a practical case of the Zhoushan thermal power plant(ZTPP),China.展开更多
The failure rate of crankpin bearing bush of diesel engine under complex working conditions such as high temperature,dynamic load and variable speed is high.After serious wear,it is easy to deteriorate the stress stat...The failure rate of crankpin bearing bush of diesel engine under complex working conditions such as high temperature,dynamic load and variable speed is high.After serious wear,it is easy to deteriorate the stress state of connecting rod body and connecting rod bolt,resulting in serious accidents such as connecting rod fracture and body damage.Based on the mixed lubrication characteristics of connecting rod big endbearing shell of diesel engine under high explosion pressure impact load,an improved mixed lubrication mechanism model is established,which considers the influence of viscoelastic micro deformation of bearing bush material,integrates the full film lubrication model and dry friction model,couples dynamic equation of connecting rod.Then the actual lubrication state of big end bearing shell is simulated numerically.Further,the correctness of the theoretical research results is verified by fault simulation experiments.The results show that the high-frequency impact signal with fixed angle domain characteristics will be generated after the serious wear of bearing bush and the deterioration of lubrication state.The fault feature capture and alarm can be realized through the condition monitoring system,which can be applied to the fault monitoring of connecting rod bearing bush of diesel engine in the future.展开更多
Aimed at the problem that Fourier decomposition method(FDM)is sensitive to noise and existing mode mixing cannot accurately extract gearbox fault features,a gear fault feature extraction method combining compound dict...Aimed at the problem that Fourier decomposition method(FDM)is sensitive to noise and existing mode mixing cannot accurately extract gearbox fault features,a gear fault feature extraction method combining compound dictionary noise reduction and optimized FDM(OFDM)is proposed.Firstly,the characteristics of the gear signals are used to construct a compound dictionary,and the orthogonal matching pursuit algorithm(OMP)is combined to reduce the noise of the vibration signal.Secondly,in order to overcome the mode mixing phenomenon occuring during the decomposition of FDM,a method of frequency band division based on the extremum of the spectrum is proposed to optimize the decomposition quality.Then,the OFDM is used to decompose the signal into several analytic Fourier intrinsic band functions(AFIBFs).Finally,the AFIBF with the largest correlation coefficient is selected for Hilbert envelope spectrum analysis.The fault feature frequencies of the vibration signal can be accurately extracted.The proposed method is validated through analyzing the gearbox fault simulation signal and the real vibration signals collected from an experimental gearbox.展开更多
With the progress in global demography of aging, dementia has become a great world health-care issue that require urgent attention and settlement. Demen-tia can arise from a variety of factors, such as neuronal degene...With the progress in global demography of aging, dementia has become a great world health-care issue that require urgent attention and settlement. Demen-tia can arise from a variety of factors, such as neuronal degeneration for Alzheimer’s disease （AD）, vascular risk factors and multiple infarcts for vascular dementia （VaD）, and both degeneration and vascular factors for mixed de-mentia （MD）. Pathophysiology of AD includes the amy-loid and tau protein hypothesis, and infammation-related mechanisms are also widespread mentioned. Subcortical ischemic vascular dementia （SIVD）, a subtype of VaD, is commonly caused by complete or incomplete lacunar infarction of VaD pathology. MD involves both degenera-tion and vascular factors, and the interaction between the two results in the complication of the pathological mech-anism and clinical phenotype. The clinical manifestationsof AD are often divided into four stages according to the progress of the disease, while the phenotypes of SIVD usually has two categories. As for MD, the phenotypes are complex and diverse. Several clinical studies showedthat its symptoms and signs are more similar to SIVD than AD. This article aims to analyze and compare the differ-ent aspects of the three kinds of dementia.展开更多
传统稀疏表示方法因其在冲击类信号特征提取中的独特优势而在故障诊断领域被广泛研究。然而,传统稀疏表示理论基于对干扰噪声的高斯分布假设,导致其难以适用于多种噪声分布混合的实际现场。针对上述问题,提出一种混合高斯噪声条件下的...传统稀疏表示方法因其在冲击类信号特征提取中的独特优势而在故障诊断领域被广泛研究。然而,传统稀疏表示理论基于对干扰噪声的高斯分布假设,导致其难以适用于多种噪声分布混合的实际现场。针对上述问题,提出一种混合高斯噪声条件下的冲击类故障特征稀疏表示方法。基于传统稀疏表示理论的贝叶斯框架,借助混合高斯分布的万有逼近性质,建立了基于db4小波字典的混合高斯噪声稀疏分解模型,并推导了基于EM(Expectation-maximum,EM)和ADMM(Alternating direction method of multipliers,ADMM)的优化求解算法用于模型求解。仿真和实验结果表明,所提出的方法能够有效提取混合噪声干扰下的冲击类微弱故障特征信号。展开更多
基金supported by the National Natural Science Foundation of China(62033008,61873143)。
文摘With the increasing intelligence and integration,a great number of two-valued variables(generally stored in the form of 0 or 1)often exist in large-scale industrial processes.However,these variables cannot be effectively handled by traditional monitoring methods such as linear discriminant analysis(LDA),principal component analysis(PCA)and partial least square(PLS)analysis.Recently,a mixed hidden naive Bayesian model(MHNBM)is developed for the first time to utilize both two-valued and continuous variables for abnormality monitoring.Although the MHNBM is effective,it still has some shortcomings that need to be improved.For the MHNBM,the variables with greater correlation to other variables have greater weights,which can not guarantee greater weights are assigned to the more discriminating variables.In addition,the conditional P(x j|x j′,y=k)probability must be computed based on historical data.When the training data is scarce,the conditional probability between continuous variables tends to be uniformly distributed,which affects the performance of MHNBM.Here a novel feature weighted mixed naive Bayes model(FWMNBM)is developed to overcome the above shortcomings.For the FWMNBM,the variables that are more correlated to the class have greater weights,which makes the more discriminating variables contribute more to the model.At the same time,FWMNBM does not have to calculate the conditional probability between variables,thus it is less restricted by the number of training data samples.Compared with the MHNBM,the FWMNBM has better performance,and its effectiveness is validated through numerical cases of a simulation example and a practical case of the Zhoushan thermal power plant(ZTPP),China.
基金Supported by the National Natural Science Foundation of China(No.52101343)the Aeronautical Science Foundation(No.201834S9002).
文摘The failure rate of crankpin bearing bush of diesel engine under complex working conditions such as high temperature,dynamic load and variable speed is high.After serious wear,it is easy to deteriorate the stress state of connecting rod body and connecting rod bolt,resulting in serious accidents such as connecting rod fracture and body damage.Based on the mixed lubrication characteristics of connecting rod big endbearing shell of diesel engine under high explosion pressure impact load,an improved mixed lubrication mechanism model is established,which considers the influence of viscoelastic micro deformation of bearing bush material,integrates the full film lubrication model and dry friction model,couples dynamic equation of connecting rod.Then the actual lubrication state of big end bearing shell is simulated numerically.Further,the correctness of the theoretical research results is verified by fault simulation experiments.The results show that the high-frequency impact signal with fixed angle domain characteristics will be generated after the serious wear of bearing bush and the deterioration of lubrication state.The fault feature capture and alarm can be realized through the condition monitoring system,which can be applied to the fault monitoring of connecting rod bearing bush of diesel engine in the future.
基金The National Natural Science Foundation of China(No.51975117)the Key Research&Development Program of Jiangsu Province(No.BE2019086).
文摘Aimed at the problem that Fourier decomposition method(FDM)is sensitive to noise and existing mode mixing cannot accurately extract gearbox fault features,a gear fault feature extraction method combining compound dictionary noise reduction and optimized FDM(OFDM)is proposed.Firstly,the characteristics of the gear signals are used to construct a compound dictionary,and the orthogonal matching pursuit algorithm(OMP)is combined to reduce the noise of the vibration signal.Secondly,in order to overcome the mode mixing phenomenon occuring during the decomposition of FDM,a method of frequency band division based on the extremum of the spectrum is proposed to optimize the decomposition quality.Then,the OFDM is used to decompose the signal into several analytic Fourier intrinsic band functions(AFIBFs).Finally,the AFIBF with the largest correlation coefficient is selected for Hilbert envelope spectrum analysis.The fault feature frequencies of the vibration signal can be accurately extracted.The proposed method is validated through analyzing the gearbox fault simulation signal and the real vibration signals collected from an experimental gearbox.
文摘With the progress in global demography of aging, dementia has become a great world health-care issue that require urgent attention and settlement. Demen-tia can arise from a variety of factors, such as neuronal degeneration for Alzheimer’s disease （AD）, vascular risk factors and multiple infarcts for vascular dementia （VaD）, and both degeneration and vascular factors for mixed de-mentia （MD）. Pathophysiology of AD includes the amy-loid and tau protein hypothesis, and infammation-related mechanisms are also widespread mentioned. Subcortical ischemic vascular dementia （SIVD）, a subtype of VaD, is commonly caused by complete or incomplete lacunar infarction of VaD pathology. MD involves both degenera-tion and vascular factors, and the interaction between the two results in the complication of the pathological mech-anism and clinical phenotype. The clinical manifestationsof AD are often divided into four stages according to the progress of the disease, while the phenotypes of SIVD usually has two categories. As for MD, the phenotypes are complex and diverse. Several clinical studies showedthat its symptoms and signs are more similar to SIVD than AD. This article aims to analyze and compare the differ-ent aspects of the three kinds of dementia.
文摘传统稀疏表示方法因其在冲击类信号特征提取中的独特优势而在故障诊断领域被广泛研究。然而,传统稀疏表示理论基于对干扰噪声的高斯分布假设,导致其难以适用于多种噪声分布混合的实际现场。针对上述问题,提出一种混合高斯噪声条件下的冲击类故障特征稀疏表示方法。基于传统稀疏表示理论的贝叶斯框架,借助混合高斯分布的万有逼近性质,建立了基于db4小波字典的混合高斯噪声稀疏分解模型,并推导了基于EM(Expectation-maximum,EM)和ADMM(Alternating direction method of multipliers,ADMM)的优化求解算法用于模型求解。仿真和实验结果表明,所提出的方法能够有效提取混合噪声干扰下的冲击类微弱故障特征信号。
