Aiming at the problems of bispectral analysis when applied to machinery fault diagnosis, a machinery fault feature extraction method based on sparseness-controlled non-negative tensor factorization （SNTF） is propose...Aiming at the problems of bispectral analysis when applied to machinery fault diagnosis, a machinery fault feature extraction method based on sparseness-controlled non-negative tensor factorization （SNTF） is proposed. First, a non-negative tensor factorization（NTF） algorithm is improved by imposing sparseness constraints on it. Secondly, the bispectral images of mechanical signals are obtained and stacked to form a third-order tensor. Thirdly, the improved algorithm is used to extract features, which are represented by a series of basis images from this tensor. Finally, coefficients indicating these basis images＇ weights in constituting original bispectral images are calculated for fault classification. Experiments on fault diagnosis of gearboxes show that the extracted features can not only reveal some nonlinear characteristics of the system, but also have intuitive meanings with regard to fault characteristic frequencies. These features provide great convenience for the interpretation of the relationships between machinery faults and corresponding bispectra.展开更多
Aiming at the slow convergence and low accuracy problems of the traditional non-negative tensor factorization, a local hierarchical non-negative tensor factorization method is proposed by applying the local objective ...Aiming at the slow convergence and low accuracy problems of the traditional non-negative tensor factorization, a local hierarchical non-negative tensor factorization method is proposed by applying the local objective function theory to non- negative tensor factorization and combining the three semi-non- negative matrix factorization（NMF） model. The effectiveness of the method is verified by the facial feature extraction experiment. Through the decomposition of a series of an air compressor＇s vibration signals composed in the form of a bispectrum by this new method, the basis images representing the fault features and corresponding weight matrices are obtained. Then the relationships between characteristics and faults are analyzed and the fault types are classified by importing the weight matrices into the BP neural network. Experimental results show that the accuracy of fault diagnosis is improved by this new method compared with other feature extraction methods.展开更多
The structural features of fiber suspensions are dependent on the fiber alignment in the flows. In this work the orientation distribution function and orientation tensors for semi-concentrated fiber suspensions in ...The structural features of fiber suspensions are dependent on the fiber alignment in the flows. In this work the orientation distribution function and orientation tensors for semi-concentrated fiber suspensions in converging channel flow were calculated, and the evolutions of the fiber alignment and the bulk effective vis-cosity were analyzed. The results showed that the bulk stress and the effective viscosity were functions of therate-of-strain tensor and the fiber orientation state ; and that the fiber suspensions evolved to steady alignment and tended to concentrate to some preferred directions close to but not same as the directions of local stream-lines. The bulk effective viscosity depended on the product of Reynolds number and time. The decrease of ef-fective viscosity near the boundary benefited the increase of the rate of flow. Finally when the fiber alignment went into steady state, the structural features of fiber suspensions were not dependent on the Reynolds numberbut on the converging channel angle.展开更多
By the resultant theory, the E-characteristic polynomial of a real rectangular tensor is defined. It is proved that an E-singular value of a real rectangular tensor is always a root of the E-characteristic polynomial....By the resultant theory, the E-characteristic polynomial of a real rectangular tensor is defined. It is proved that an E-singular value of a real rectangular tensor is always a root of the E-characteristic polynomial. The definition of the regularity of square tensors is generalized to the rectangular tensors, and in the regular case, a root of the Echaracteristic polynomial of a special rectangular tensor is an E-singular value of the rectangular tensor. Moreover, the best rank-one approximation of a real partially symmetric rectangular tensor is investigated.展开更多
Edge detection of potential field interpretation is an important task. The traditional edge detection methods have poor ability in outlining weak amplitude anomalies clearly. The resolved edges position is blurred.We ...Edge detection of potential field interpretation is an important task. The traditional edge detection methods have poor ability in outlining weak amplitude anomalies clearly. The resolved edges position is blurred.We purposed new edge detection methods based on directional eigenvalues of potential field gradient tensor for the causative sources. In order to balance strong and weak amplitude anomalies simultaneously,we present one normalization method using different orders of vertical derivatives to improve the new filters. The presented filters were tested on synthetic and real potential field data to verify its feasibility. All of the results have shown that the new edge detection methods can not only display the sources edges precisely and clearly,but also bring out more geological subtle details.展开更多
基金The National Natural Science Foundation of China (No.50875048)the Natural Science Foundation of Jiangsu Province (No.BK2007115)the National High Technology Research and Development Program of China (863 Program)(No.2007AA04Z421)
文摘Aiming at the problems of bispectral analysis when applied to machinery fault diagnosis, a machinery fault feature extraction method based on sparseness-controlled non-negative tensor factorization （SNTF） is proposed. First, a non-negative tensor factorization（NTF） algorithm is improved by imposing sparseness constraints on it. Secondly, the bispectral images of mechanical signals are obtained and stacked to form a third-order tensor. Thirdly, the improved algorithm is used to extract features, which are represented by a series of basis images from this tensor. Finally, coefficients indicating these basis images＇ weights in constituting original bispectral images are calculated for fault classification. Experiments on fault diagnosis of gearboxes show that the extracted features can not only reveal some nonlinear characteristics of the system, but also have intuitive meanings with regard to fault characteristic frequencies. These features provide great convenience for the interpretation of the relationships between machinery faults and corresponding bispectra.
基金The National Natural Science Foundation of China(No.50875078)the Natural Science Foundation of Jiangsu Province(No.BK2007115)the National High Technology Research and Development Program of China(863 Program)(No.2007AA04Z421)
文摘Aiming at the slow convergence and low accuracy problems of the traditional non-negative tensor factorization, a local hierarchical non-negative tensor factorization method is proposed by applying the local objective function theory to non- negative tensor factorization and combining the three semi-non- negative matrix factorization（NMF） model. The effectiveness of the method is verified by the facial feature extraction experiment. Through the decomposition of a series of an air compressor＇s vibration signals composed in the form of a bispectrum by this new method, the basis images representing the fault features and corresponding weight matrices are obtained. Then the relationships between characteristics and faults are analyzed and the fault types are classified by importing the weight matrices into the BP neural network. Experimental results show that the accuracy of fault diagnosis is improved by this new method compared with other feature extraction methods.
文摘The structural features of fiber suspensions are dependent on the fiber alignment in the flows. In this work the orientation distribution function and orientation tensors for semi-concentrated fiber suspensions in converging channel flow were calculated, and the evolutions of the fiber alignment and the bulk effective vis-cosity were analyzed. The results showed that the bulk stress and the effective viscosity were functions of therate-of-strain tensor and the fiber orientation state ; and that the fiber suspensions evolved to steady alignment and tended to concentrate to some preferred directions close to but not same as the directions of local stream-lines. The bulk effective viscosity depended on the product of Reynolds number and time. The decrease of ef-fective viscosity near the boundary benefited the increase of the rate of flow. Finally when the fiber alignment went into steady state, the structural features of fiber suspensions were not dependent on the Reynolds numberbut on the converging channel angle.
文摘By the resultant theory, the E-characteristic polynomial of a real rectangular tensor is defined. It is proved that an E-singular value of a real rectangular tensor is always a root of the E-characteristic polynomial. The definition of the regularity of square tensors is generalized to the rectangular tensors, and in the regular case, a root of the Echaracteristic polynomial of a special rectangular tensor is an E-singular value of the rectangular tensor. Moreover, the best rank-one approximation of a real partially symmetric rectangular tensor is investigated.
基金financially supported by Sino Probe-09-01 Grant No. 201311192Project 2014100 Supported by Graduate Innovation Fund of Jilin University
文摘Edge detection of potential field interpretation is an important task. The traditional edge detection methods have poor ability in outlining weak amplitude anomalies clearly. The resolved edges position is blurred.We purposed new edge detection methods based on directional eigenvalues of potential field gradient tensor for the causative sources. In order to balance strong and weak amplitude anomalies simultaneously,we present one normalization method using different orders of vertical derivatives to improve the new filters. The presented filters were tested on synthetic and real potential field data to verify its feasibility. All of the results have shown that the new edge detection methods can not only display the sources edges precisely and clearly,but also bring out more geological subtle details.
