How to extract optimal composite attributes from a variety of conventional seismic attributes to detect reservoir features is a reservoir predication key,which is usually solved by reducing dimensionality.Principle co...How to extract optimal composite attributes from a variety of conventional seismic attributes to detect reservoir features is a reservoir predication key,which is usually solved by reducing dimensionality.Principle component analysis（PCA） is the most widely-used linear dimensionality reduction method at present.However,the relationships between seismic attributes and reservoir features are non-linear,so seismic attribute dimensionality reduction based on linear transforms can＇t solve non-linear problems well,reducing reservoir prediction precision.As a new non-linear learning method,manifold learning supplies a new method for seismic attribute analysis.It can discover the intrinsic features and rules hidden in the data by computing low-dimensional,neighborhood-preserving embeddings of high-dimensional inputs.In this paper,we try to extract seismic attributes using locally linear embedding（LLE）,realizing inter-horizon attributes dimensionality reduction of 3D seismic data first and discuss the optimization of its key parameters.Combining model analysis and case studies,we compare the dimensionality reduction and clustering effects of LLE and PCA,both of which indicate that LLE can retain the intrinsic structure of the inputs.The composite attributes and clustering results based on LLE better characterize the distribution of sedimentary facies,reservoir,and even reservoir fluids.展开更多
In this paper, the approximate expressions of the solitary wave solutions for a class of nonlinear disturbed long-wave system are constructed using the homotopie mapping method.
A clustering algorithm based on Sparse Projection (SP), called Sparse Projection Clus- tering (SPC), is proposed in this letter. The basic idea is applying SP to project the observed data onto a high-dimensional spars...A clustering algorithm based on Sparse Projection (SP), called Sparse Projection Clus- tering (SPC), is proposed in this letter. The basic idea is applying SP to project the observed data onto a high-dimensional sparse space, which is a nonlinear mapping with an explicit form and the K-means clustering algorithm can be therefore used to explore the inherent data patterns in the new space. The proposed algorithm is applied to cluster a complete artificial dataset and an incomplete real dataset. In comparison with the kernel K-means clustering algorithm, the proposed algorithm is more efficient.展开更多
Taking into account that fuzzy ontology mapping has wide application and cannot be dealt with in many fields at present,a Chinese fuzzy ontology model and a method for Chinese fuzzy ontology mapping are proposed.The m...Taking into account that fuzzy ontology mapping has wide application and cannot be dealt with in many fields at present,a Chinese fuzzy ontology model and a method for Chinese fuzzy ontology mapping are proposed.The mapping discovery between two ontologies is achieved by computing the similarity between the concepts of two ontologies.Every concept consists of four features of concept name,property,instance and structure.First,the algorithms of calculating four individual similarities corresponding to the four features are given.Secondly,the similarity vectors consisting of four weighted individual similarities are built,and the weights are the linear function of harmony and reliability.The similarity vector is used to represent the similarity relation between two concepts which belong to different fuzzy ontolgoies.Lastly,Support Vector Machine(SVM) is used to get the mapping concept pairs by the similarity vectors.Experiment results are satisfactory.展开更多
This paper is a research on the characteristics of power big data. According to the characteristics of ＂large volume＂, ＂species diversity＂, ＂sparse value density＂, ＂fast speed＂ of the power big data, a predict...This paper is a research on the characteristics of power big data. According to the characteristics of ＂large volume＂, ＂species diversity＂, ＂sparse value density＂, ＂fast speed＂ of the power big data, a prediction model of multi-source information fusion for large data is established, the fusion prediction of various parameters of the same object is realized. A combined algorithm of Map Reduce and neural network is used in this paper. Using clustering and nonlinear mapping ability of neural network, it can effectively solve the problem of nonlinear objective function approximation, and neural network is applied to the prediction of fusion. In this paper, neural network model using multi layer feed forward network--BP neural network. Simultaneously, to achieve large-scale data sets in parallel computing, the parallelism and real-time property of the algorithm should be considered, further combined with Reduce Map model, to realize the parallel processing of the algorithm, making it more suitable for the study of the fusion of large data. And finally, through simulation, it verifies the feasibility of the proposed model and algorithm.展开更多
Time-varying stiffness is one of the most important dynamic characteristics of rolling element bearings.The method of analyzing the elements in the bearing stiffness matrix is usually adopted to investigate the charac...Time-varying stiffness is one of the most important dynamic characteristics of rolling element bearings.The method of analyzing the elements in the bearing stiffness matrix is usually adopted to investigate the characteristics of bearing stiffness.Linear mapping structure of the bearing stiffness matrix is helpful to understand the varying compliance excitation and its influence on vibration transmission.In this study,a method to analyze the mapping structure of bearing stiffness matrix is proposed based on the singular value decomposition of block matrices in the stiffness matrix.Not only does this method have the advantages of coordinate transformation independence and unit independence,but also the analysis procedure involved is geometrically intuitive.The time-varying stiffness matrix of double-row tapered bearing is calculated and analyzed using the proposed method under two representative load cases.The principal stiffnesses and principal axes defined in the method together indicate the dominant and insignificant stiffness properties with the corresponding directions,and the vibration transmission properties are also revealed.Besides,the coupling behaviors between different shaft motions are found during the analysis of mapping structure.The mechanism of the generation of varying compliance excitation is also revealed.展开更多
基金National Key Science & Technology Special Projects(Grant No.2008ZX05000-004)CNPC Projects(Grant No.2008E-0610-10).
文摘How to extract optimal composite attributes from a variety of conventional seismic attributes to detect reservoir features is a reservoir predication key,which is usually solved by reducing dimensionality.Principle component analysis（PCA） is the most widely-used linear dimensionality reduction method at present.However,the relationships between seismic attributes and reservoir features are non-linear,so seismic attribute dimensionality reduction based on linear transforms can＇t solve non-linear problems well,reducing reservoir prediction precision.As a new non-linear learning method,manifold learning supplies a new method for seismic attribute analysis.It can discover the intrinsic features and rules hidden in the data by computing low-dimensional,neighborhood-preserving embeddings of high-dimensional inputs.In this paper,we try to extract seismic attributes using locally linear embedding（LLE）,realizing inter-horizon attributes dimensionality reduction of 3D seismic data first and discuss the optimization of its key parameters.Combining model analysis and case studies,we compare the dimensionality reduction and clustering effects of LLE and PCA,both of which indicate that LLE can retain the intrinsic structure of the inputs.The composite attributes and clustering results based on LLE better characterize the distribution of sedimentary facies,reservoir,and even reservoir fluids.
基金Supported by the National Natural Science Foundation of China under Grant No.40876010the Main Direction Program of the Knowledge Innovation Project of Chinese Academy of Sciences under Grant No.KZCX2-YW-Q03-08+2 种基金the LASG State Key Laboratory Special Fundthe Foundation of Shanghai Municipal Education Commission under Grant No.E03004the Natural Science Foundation of Zhejiang Province under Grant No.Y6090164
文摘In this paper, the approximate expressions of the solitary wave solutions for a class of nonlinear disturbed long-wave system are constructed using the homotopie mapping method.
基金Supported by the National Natural Science Foundation of China (No.60872123)the Joint Fund of the National Natural Science Foundation and the Guangdong Provin-cial Natural Science Foundation (No.U0835001)
文摘A clustering algorithm based on Sparse Projection (SP), called Sparse Projection Clus- tering (SPC), is proposed in this letter. The basic idea is applying SP to project the observed data onto a high-dimensional sparse space, which is a nonlinear mapping with an explicit form and the K-means clustering algorithm can be therefore used to explore the inherent data patterns in the new space. The proposed algorithm is applied to cluster a complete artificial dataset and an incomplete real dataset. In comparison with the kernel K-means clustering algorithm, the proposed algorithm is more efficient.
基金supported by the Natural Science Foundation of Beijing City under Grant No.4123094the Science and Technology Project of Beijing Municipal Commission of Education under Grants No.KM201110028020,No.KM201010028019+1 种基金the National Nature Science Foundation under Grants No.61100205,No.60873001,No.60863011,No.61175068the Fundamental Research Funds for the Central Universities under Grant No.2009RC0212
文摘Taking into account that fuzzy ontology mapping has wide application and cannot be dealt with in many fields at present,a Chinese fuzzy ontology model and a method for Chinese fuzzy ontology mapping are proposed.The mapping discovery between two ontologies is achieved by computing the similarity between the concepts of two ontologies.Every concept consists of four features of concept name,property,instance and structure.First,the algorithms of calculating four individual similarities corresponding to the four features are given.Secondly,the similarity vectors consisting of four weighted individual similarities are built,and the weights are the linear function of harmony and reliability.The similarity vector is used to represent the similarity relation between two concepts which belong to different fuzzy ontolgoies.Lastly,Support Vector Machine(SVM) is used to get the mapping concept pairs by the similarity vectors.Experiment results are satisfactory.
文摘This paper is a research on the characteristics of power big data. According to the characteristics of ＂large volume＂, ＂species diversity＂, ＂sparse value density＂, ＂fast speed＂ of the power big data, a prediction model of multi-source information fusion for large data is established, the fusion prediction of various parameters of the same object is realized. A combined algorithm of Map Reduce and neural network is used in this paper. Using clustering and nonlinear mapping ability of neural network, it can effectively solve the problem of nonlinear objective function approximation, and neural network is applied to the prediction of fusion. In this paper, neural network model using multi layer feed forward network--BP neural network. Simultaneously, to achieve large-scale data sets in parallel computing, the parallelism and real-time property of the algorithm should be considered, further combined with Reduce Map model, to realize the parallel processing of the algorithm, making it more suitable for the study of the fusion of large data. And finally, through simulation, it verifies the feasibility of the proposed model and algorithm.
基金the Joint Funds of the National Natural Science Foundation of China(Grant No.U1834202).
文摘Time-varying stiffness is one of the most important dynamic characteristics of rolling element bearings.The method of analyzing the elements in the bearing stiffness matrix is usually adopted to investigate the characteristics of bearing stiffness.Linear mapping structure of the bearing stiffness matrix is helpful to understand the varying compliance excitation and its influence on vibration transmission.In this study,a method to analyze the mapping structure of bearing stiffness matrix is proposed based on the singular value decomposition of block matrices in the stiffness matrix.Not only does this method have the advantages of coordinate transformation independence and unit independence,but also the analysis procedure involved is geometrically intuitive.The time-varying stiffness matrix of double-row tapered bearing is calculated and analyzed using the proposed method under two representative load cases.The principal stiffnesses and principal axes defined in the method together indicate the dominant and insignificant stiffness properties with the corresponding directions,and the vibration transmission properties are also revealed.Besides,the coupling behaviors between different shaft motions are found during the analysis of mapping structure.The mechanism of the generation of varying compliance excitation is also revealed.
