The performance of traditional vibration based fault diagnosis methods greatly depends on those hand- crafted features extracted using signal processing algo- rithms, which require significant amounts of domain knowle...The performance of traditional vibration based fault diagnosis methods greatly depends on those hand- crafted features extracted using signal processing algo- rithms, which require significant amounts of domain knowledge and human labor, and do not generalize well to new diagnosis domains. Recently, unsupervised represen- tation learning provides an alternative promising solution to feature extraction in traditional fault diagnosis due to its superior learning ability from unlabeled data. Given that vibration signals usually contain multiple temporal struc- tures, this paper proposes a multiscale representation learning (MSRL) framework to learn useful features directly from raw vibration signals, with the aim to capture rich and complementary fault pattern information at dif- ferent scales. In our proposed approach, a coarse-grained procedure is first employed to obtain multiple scale signals from an original vibration signal. Then, sparse filtering, a newly developed unsupervised learning algorithm, is applied to automatically learn useful features from each scale signal, respectively, and then the learned features at each scale to be concatenated one by one to obtain multi- scale representations. Finally, the multiscale representa- tions are fed into a supervised classifier to achieve diagnosis results. Our proposed approach is evaluated using two different case studies: motor bearing and wind turbine gearbox fault diagnosis. Experimental results show that the proposed MSRL approach can take full advantages of the availability of unlabeled data to learn discriminative features and achieved better performance with higher accuracy and stability compared to the traditional approaches.展开更多
The success of deep transfer learning in fault diagnosis is attributed to the collection of high-quality labeled data from the source domain.However,in engineering scenarios,achieving such high-quality label annotatio...The success of deep transfer learning in fault diagnosis is attributed to the collection of high-quality labeled data from the source domain.However,in engineering scenarios,achieving such high-quality label annotation is difficult and expensive.The incorrect label annotation produces two negative effects:1)the complex decision boundary of diagnosis models lowers the generalization performance on the target domain,and2)the distribution of target domain samples becomes misaligned with the false-labeled samples.To overcome these negative effects,this article proposes a solution called the label recovery and trajectory designable network(LRTDN).LRTDN consists of three parts.First,a residual network with dual classifiers is to learn features from cross-domain samples.Second,an annotation check module is constructed to generate a label anomaly indicator that could modify the abnormal labels of false-labeled samples in the source domain.With the training of relabeled samples,the complexity of diagnosis model is reduced via semi-supervised learning.Third,the adaptation trajectories are designed for sample distributions across domains.This ensures that the target domain samples are only adapted with the pure-labeled samples.The LRTDN is verified by two case studies,in which the diagnosis knowledge of bearings is transferred across different working conditions as well as different yet related machines.The results show that LRTDN offers a high diagnosis accuracy even in the presence of incorrect annotation.展开更多
Deep learning has achieved much success in mechanical intelligent fault diagnosis in recent years. However, many deep learning methods cannot fully extract fault information to recognize mechanical health states when ...Deep learning has achieved much success in mechanical intelligent fault diagnosis in recent years. However, many deep learning methods cannot fully extract fault information to recognize mechanical health states when processing high-dimensional samples. Therefore, a multi-model ensemble deep learning method based on deep convolutional neural network (DCNN) is proposed in this study to accomplish fault recognition of high-dimensional samples. First, several 1D DCNN models with different activation functions are trained through dimension reduction learning to obtain different fault features from high-dimensional samples. Second, the obtained features are constructed into 2D images with multiple channels through a conversion method. The integrated 2D feature images can effectively represent the fault characteristic contained in raw high-dimension vibration signals. Lastly, a 2D DCNN model with multi-layer convolution and pooling is used to automatically learn features from the 2D images and identify the fault mode of the mechanical equipment by adopting a softmax classifier. The proposed method, which is validated using the bearing public dataset of Case Western Reserve University, USA and a one-stage reduction gearbox dataset, has high recognition accuracy. Compared with other classical deep learning methods, the proposed fault diagnosis method has considerable improvements.展开更多
After analyzing the structure and characteristics of the hybrid intelligent diagnosis system of CNC machine toolsCNC-HIDS), we describe the intelligent hybrid mechanism of the CNC-HIDS, and present the evaluation and ...After analyzing the structure and characteristics of the hybrid intelligent diagnosis system of CNC machine toolsCNC-HIDS), we describe the intelligent hybrid mechanism of the CNC-HIDS, and present the evaluation and the running instance of the system. Through tryout and validation, we attain satisfactory results.展开更多
The high accurate classification ability of an intelligent diagnosis method often needs a large amount of training samples with high-dimensional eigenvectors, however the characteristics of the signal need to be extra...The high accurate classification ability of an intelligent diagnosis method often needs a large amount of training samples with high-dimensional eigenvectors, however the characteristics of the signal need to be extracted accurately. Although the existing EMD(empirical mode decomposition) and EEMD(ensemble empirical mode decomposition) are suitable for processing non-stationary and non-linear signals, but when a short signal, such as a hydraulic impact signal, is concerned, their decomposition accuracy become very poor. An improve EEMD is proposed specifically for short hydraulic impact signals. The improvements of this new EEMD are mainly reflected in four aspects, including self-adaptive de-noising based on EEMD, signal extension based on SVM(support vector machine), extreme center fitting based on cubic spline interpolation, and pseudo component exclusion based on cross-correlation analysis. After the energy eigenvector is extracted from the result of the improved EEMD, the fault pattern recognition based on SVM with small amount of low-dimensional training samples is studied. At last, the diagnosis ability of improved EEMD+SVM method is compared with the EEMD+SVM and EMD+SVM methods, and its diagnosis accuracy is distinctly higher than the other two methods no matter the dimension of the eigenvectors are low or high. The improved EEMD is very propitious for the decomposition of short signal, such as hydraulic impact signal, and its combination with SVM has high ability for the diagnosis of hydraulic impact faults.展开更多
Fault diagnosis of rotating machinery has always drawn wide attention.In this paper,Intrinsic Component Filtering(ICF),which achieves population sparsity and lifetime consistency using two constraints:l1=2 norm of col...Fault diagnosis of rotating machinery has always drawn wide attention.In this paper,Intrinsic Component Filtering(ICF),which achieves population sparsity and lifetime consistency using two constraints:l1=2 norm of column features and l3=2-norm of row features,is proposed for the machinery fault diagnosis.ICF can be used as a feature learning algorithm,and the learned features can be fed into the classification to achieve the automatic fault classification.ICF can also be used as a filter training method to extract and separate weak fault components from the noise signals without any prior experience.Simulated and experimental signals of bearing fault are used to validate the performance of ICF.The results confirm that ICF performs superior in three fault diagnosis fields including intelligent fault diagnosis,weak signature detection and compound fault separation.展开更多
The planetary gearbox is a critical part of wind turbines,and has great significance for their safety and reliability.Intelligent fault diagnosis methods for these gearboxes have made some achievements based on the av...The planetary gearbox is a critical part of wind turbines,and has great significance for their safety and reliability.Intelligent fault diagnosis methods for these gearboxes have made some achievements based on the availability of large quantities of labeled data.However,the data collected from the diagnosed devices are always unlabeled,and the acquisition of fault data from real gearboxes is time-consuming and laborious.As some gearbox faults can be conveniently simulated by a relatively precise dynamic model,the data from dynamic simulation containing some features are related to those from the actual machines.As a potential tool,transfer learning adapts a network trained in a source domain to its application in a target domain.Therefore,a novel fault diagnosis method combining transfer learning with dynamic model is proposed to identify the health conditions of planetary gearboxes.In the method,a modified lumped-parameter dynamic model of a planetary gear train is established to simulate the resultant vibration signal,while an optimized deep transfer learning network based on a one-dimensional convolutional neural network is built to extract domain-invariant features from different domains to achieve fault classification.Various groups of transfer diagnosis experiments of planetary gearboxes are carried out,and the experimental results demonstrate the effectiveness and the reliability of both the dynamic model and the proposed method.展开更多
Purpose-Intelligent diagnosis of equipment faults can effectively avoid the shutdown caused by equipment faults and improve the safety of the equipment.At present,the diagnosis of various kinds of bearing fault inform...Purpose-Intelligent diagnosis of equipment faults can effectively avoid the shutdown caused by equipment faults and improve the safety of the equipment.At present,the diagnosis of various kinds of bearing fault information,such as the occurrence,location and degree of fault,can be carried out by machine learning and deep learning and realized through the multiclassification method.However,the multiclassification method is not perfect in distinguishing similar fault categories and visual representation of fault information.To improve the above shortcomings,an end-to-end fault multilabel classification model is proposed for bearing fault diagnosis.Design/methodology/approach-In this model,the labels of each bearing are binarized by using the binary relevance method.Then,the integrated convolutional neural network and gated recurrent unit(CNN-GRU)is employed to classify faults.Different from the general CNN networks,the CNN-GRU network adds multiple GRU layers after the convolutional layers and the pool layers.Findings-The Paderborn University bearing dataset is utilized to demonstrate the practicability of the model.The experimental results show that the average accuracy in test set is 99.7%,and the proposed network is better than multilayer perceptron and CNN in fault diagnosis of bearing,and the multilabel classification method is superior to the multiclassification method.Consequently,the model can intuitively classify faults with higher accuracy.Originality/value-The fault labels of each bearing are labeled according to the failure or not,the fault location,the damage mode and the damage degree,and then the binary value is obtained.The multilabel problem is transformed into a binary classification problem of each fault label by the binary relevance method,and the predicted probability value of each fault label is directly output in the output layer,which visually distinguishes different fault conditions.展开更多
基金Supported by Hebei Provincial Natural Science Foundation of China(Grant No.F2016203421)
文摘The performance of traditional vibration based fault diagnosis methods greatly depends on those hand- crafted features extracted using signal processing algo- rithms, which require significant amounts of domain knowledge and human labor, and do not generalize well to new diagnosis domains. Recently, unsupervised represen- tation learning provides an alternative promising solution to feature extraction in traditional fault diagnosis due to its superior learning ability from unlabeled data. Given that vibration signals usually contain multiple temporal struc- tures, this paper proposes a multiscale representation learning (MSRL) framework to learn useful features directly from raw vibration signals, with the aim to capture rich and complementary fault pattern information at dif- ferent scales. In our proposed approach, a coarse-grained procedure is first employed to obtain multiple scale signals from an original vibration signal. Then, sparse filtering, a newly developed unsupervised learning algorithm, is applied to automatically learn useful features from each scale signal, respectively, and then the learned features at each scale to be concatenated one by one to obtain multi- scale representations. Finally, the multiscale representa- tions are fed into a supervised classifier to achieve diagnosis results. Our proposed approach is evaluated using two different case studies: motor bearing and wind turbine gearbox fault diagnosis. Experimental results show that the proposed MSRL approach can take full advantages of the availability of unlabeled data to learn discriminative features and achieved better performance with higher accuracy and stability compared to the traditional approaches.
基金the National Key R&D Program of China(2022YFB3402100)the National Science Fund for Distinguished Young Scholars of China(52025056)+4 种基金the National Natural Science Foundation of China(52305129)the China Postdoctoral Science Foundation(2023M732789)the China Postdoctoral Innovative Talents Support Program(BX20230290)the Open Foundation of Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment(2022JXKF JJ01)the Fundamental Research Funds for Central Universities。
文摘The success of deep transfer learning in fault diagnosis is attributed to the collection of high-quality labeled data from the source domain.However,in engineering scenarios,achieving such high-quality label annotation is difficult and expensive.The incorrect label annotation produces two negative effects:1)the complex decision boundary of diagnosis models lowers the generalization performance on the target domain,and2)the distribution of target domain samples becomes misaligned with the false-labeled samples.To overcome these negative effects,this article proposes a solution called the label recovery and trajectory designable network(LRTDN).LRTDN consists of three parts.First,a residual network with dual classifiers is to learn features from cross-domain samples.Second,an annotation check module is constructed to generate a label anomaly indicator that could modify the abnormal labels of false-labeled samples in the source domain.With the training of relabeled samples,the complexity of diagnosis model is reduced via semi-supervised learning.Third,the adaptation trajectories are designed for sample distributions across domains.This ensures that the target domain samples are only adapted with the pure-labeled samples.The LRTDN is verified by two case studies,in which the diagnosis knowledge of bearings is transferred across different working conditions as well as different yet related machines.The results show that LRTDN offers a high diagnosis accuracy even in the presence of incorrect annotation.
基金This study was financially supported by the National Key R&D Program of China(Grant No.2017YFD0400405).
文摘Deep learning has achieved much success in mechanical intelligent fault diagnosis in recent years. However, many deep learning methods cannot fully extract fault information to recognize mechanical health states when processing high-dimensional samples. Therefore, a multi-model ensemble deep learning method based on deep convolutional neural network (DCNN) is proposed in this study to accomplish fault recognition of high-dimensional samples. First, several 1D DCNN models with different activation functions are trained through dimension reduction learning to obtain different fault features from high-dimensional samples. Second, the obtained features are constructed into 2D images with multiple channels through a conversion method. The integrated 2D feature images can effectively represent the fault characteristic contained in raw high-dimension vibration signals. Lastly, a 2D DCNN model with multi-layer convolution and pooling is used to automatically learn features from the 2D images and identify the fault mode of the mechanical equipment by adopting a softmax classifier. The proposed method, which is validated using the bearing public dataset of Case Western Reserve University, USA and a one-stage reduction gearbox dataset, has high recognition accuracy. Compared with other classical deep learning methods, the proposed fault diagnosis method has considerable improvements.
文摘After analyzing the structure and characteristics of the hybrid intelligent diagnosis system of CNC machine toolsCNC-HIDS), we describe the intelligent hybrid mechanism of the CNC-HIDS, and present the evaluation and the running instance of the system. Through tryout and validation, we attain satisfactory results.
基金Supported by National Natural Science Foundation of China(Grant Nos.51175511,61472444)Jiangsu Provincial Natural Science Foundation of China(Grant No.BK20150724)Pre-study Foundation of PLA University of Science and Technology,China(Grant No.KYGYZL139)
文摘The high accurate classification ability of an intelligent diagnosis method often needs a large amount of training samples with high-dimensional eigenvectors, however the characteristics of the signal need to be extracted accurately. Although the existing EMD(empirical mode decomposition) and EEMD(ensemble empirical mode decomposition) are suitable for processing non-stationary and non-linear signals, but when a short signal, such as a hydraulic impact signal, is concerned, their decomposition accuracy become very poor. An improve EEMD is proposed specifically for short hydraulic impact signals. The improvements of this new EEMD are mainly reflected in four aspects, including self-adaptive de-noising based on EEMD, signal extension based on SVM(support vector machine), extreme center fitting based on cubic spline interpolation, and pseudo component exclusion based on cross-correlation analysis. After the energy eigenvector is extracted from the result of the improved EEMD, the fault pattern recognition based on SVM with small amount of low-dimensional training samples is studied. At last, the diagnosis ability of improved EEMD+SVM method is compared with the EEMD+SVM and EMD+SVM methods, and its diagnosis accuracy is distinctly higher than the other two methods no matter the dimension of the eigenvectors are low or high. The improved EEMD is very propitious for the decomposition of short signal, such as hydraulic impact signal, and its combination with SVM has high ability for the diagnosis of hydraulic impact faults.
基金supported by the Major National Science and Technology Projects(No.2017-IV-0008-0045)the National Natural Science Foundation of China(Nos.51675262 and 51975276)+1 种基金the Advance Research Field Fund Project of China(No.61400040304)the National Key Research and Development Program of China(No.2018YFB2003300)。
文摘Fault diagnosis of rotating machinery has always drawn wide attention.In this paper,Intrinsic Component Filtering(ICF),which achieves population sparsity and lifetime consistency using two constraints:l1=2 norm of column features and l3=2-norm of row features,is proposed for the machinery fault diagnosis.ICF can be used as a feature learning algorithm,and the learned features can be fed into the classification to achieve the automatic fault classification.ICF can also be used as a filter training method to extract and separate weak fault components from the noise signals without any prior experience.Simulated and experimental signals of bearing fault are used to validate the performance of ICF.The results confirm that ICF performs superior in three fault diagnosis fields including intelligent fault diagnosis,weak signature detection and compound fault separation.
基金Natural Science Foundation of Shanghai (21ZR1425400)Shanghai Rising-Star Program (21QC1400200)+1 种基金National Natural Science Foundation of China (51977128)Shanghai Science and Technology Project (20142202600).
文摘The planetary gearbox is a critical part of wind turbines,and has great significance for their safety and reliability.Intelligent fault diagnosis methods for these gearboxes have made some achievements based on the availability of large quantities of labeled data.However,the data collected from the diagnosed devices are always unlabeled,and the acquisition of fault data from real gearboxes is time-consuming and laborious.As some gearbox faults can be conveniently simulated by a relatively precise dynamic model,the data from dynamic simulation containing some features are related to those from the actual machines.As a potential tool,transfer learning adapts a network trained in a source domain to its application in a target domain.Therefore,a novel fault diagnosis method combining transfer learning with dynamic model is proposed to identify the health conditions of planetary gearboxes.In the method,a modified lumped-parameter dynamic model of a planetary gear train is established to simulate the resultant vibration signal,while an optimized deep transfer learning network based on a one-dimensional convolutional neural network is built to extract domain-invariant features from different domains to achieve fault classification.Various groups of transfer diagnosis experiments of planetary gearboxes are carried out,and the experimental results demonstrate the effectiveness and the reliability of both the dynamic model and the proposed method.
基金supported by the Mountain and Sea Talents Project of Shandong University of Science and Technology(grant numbers:01040055230)the National Nature Science Foundation of Shandong Province of China(Grant No.ZR2018MEE024).
文摘Purpose-Intelligent diagnosis of equipment faults can effectively avoid the shutdown caused by equipment faults and improve the safety of the equipment.At present,the diagnosis of various kinds of bearing fault information,such as the occurrence,location and degree of fault,can be carried out by machine learning and deep learning and realized through the multiclassification method.However,the multiclassification method is not perfect in distinguishing similar fault categories and visual representation of fault information.To improve the above shortcomings,an end-to-end fault multilabel classification model is proposed for bearing fault diagnosis.Design/methodology/approach-In this model,the labels of each bearing are binarized by using the binary relevance method.Then,the integrated convolutional neural network and gated recurrent unit(CNN-GRU)is employed to classify faults.Different from the general CNN networks,the CNN-GRU network adds multiple GRU layers after the convolutional layers and the pool layers.Findings-The Paderborn University bearing dataset is utilized to demonstrate the practicability of the model.The experimental results show that the average accuracy in test set is 99.7%,and the proposed network is better than multilayer perceptron and CNN in fault diagnosis of bearing,and the multilabel classification method is superior to the multiclassification method.Consequently,the model can intuitively classify faults with higher accuracy.Originality/value-The fault labels of each bearing are labeled according to the failure or not,the fault location,the damage mode and the damage degree,and then the binary value is obtained.The multilabel problem is transformed into a binary classification problem of each fault label by the binary relevance method,and the predicted probability value of each fault label is directly output in the output layer,which visually distinguishes different fault conditions.