|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Transport network and flow mechanism of shallow ore-bearing magma in Tongling ore cluster area显示文摘Abundant studies revealed that shallow intrusions of the Yanshanian epoch resulted in the mass mineralization of the Tongling region. Various evidences showed there existed a concealed magma chamber at ?10 km depth in the middle part of this region during Yanshanian epoch, from which the ore-forming magma was generated and then transported to the superficial layer. Yet the transport network and flow mechanism of the shallow ore-bearing magma, the key problem associ- ated with ore-forming process, was relatively little focused on. Integrate analysis of structural me- chanics, statistical fractal and geological facts suggested that NE trending high-angle fold-related thrust faults and the tessellated basement ones served as the main pathways for the shallow magma’s transporting, moreover, the saddle void spaces among adjacent strata in the folds upon this fault system provided the place for magma’s emplacement. So the folds in the upper part and faults in the lower part of the upper crust constituted the fluid’s transport and emplacement network. During the deformation of geologic body with multi-layer structure, the layers in the upper part tended to fold when received the jacking stress from the lower part, while the lower one inclined to fault undergoing loads of the upper part. And the producing probability of this structure assemblage was highly in- creased in the condition, such as in the Tongling area, that the mechanic rigidity of the lower layers was stronger than that of the upper ones. For the pre-existence of fluid-conducting network, the top magma with high volatile in the magma chamber transported rapidly to the superficial layer in dyking pattern, located in the void spaces of folds, filled and reconstructed them. The sudden drop of pres- sure caused the fluid unmixing from the magma and mass ore-forming elements concentration. Pulse activity of the dyking may be the principal reason why magmatic bodies in the Tongling area were spatially-temporally concomitant and limited flux in chemical compositions. | DENG Jun1,2, WANG Qingfei1,2 , HUANG Dinghua3 , WAN Li4, YANG Liqiang1,2 & GAO Bangfei1,2 1. State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Beijing 100083, China 2. Key Laboratory of Lithosphere Tectonics and Lithoprobing Technology of Ministry of Education, China University of Geo- sciences, Beijing 100083, China 3. Faculty of Earth Sciences, China University of Geosciences, Wuhan 430074, China 4. School of Mathematics and Information Science, Guangzhou University, Guangzhou 510405, China | 2006 | Science China Earth Sciences2006，49，4： | 21 |
| 2 | Study on the Overfitting of the Artificial Neural Network Forecasting Model显示文摘Because of overfitting and the improvement of generalization capability (GC) available in the construction of forecasting models using artificial neural network (ANN), a new method is proposed for model establishment by means of making a low-dimension ANN learning matrix through principal component analysis (PCA). The results show that the PCA is able to construct an ANN model without the need of finding an optimal structure with the appropriate number of hidden-layer nodes, thus avoids overfitting by condensing forecasting information, reducing dimension and removing noise, and GC is greatly raised compared to the traditional ANN and stepwise regression techniques for model establishment. | 金龙 况雪源 黄海洪 覃志年 王业宏 | 2005 | Acta meteorologica Sinica2005，19，2： | 9 |
| 3 | Neural Network Based on GA-BP Algorithm and its Application in the Protein Secondary Structure Prediction显示文摘The advantages and disadvantages of genetic algorithm and BP algorithm are introduced. A neural network based on GA-BP algorithm is proposed and applied in the prediction of protein secondary structure, which combines the advantages of BP and GA. The prediction and training on the neural network are made respectively based on 4 structure classifications of protein so as to get higher rate of predication---the highest prediction rate 75.65%,the average prediction rate 65.04%. | YANG Yang LI Kai-yang | 2006 | Chinese Journal of Biomedical Engineering(English Edition)2006，15，1： | 8 |
| 4 | Study on Missile Intelligent Fault Diagnosis System Based on Fuzzy NN Expert System显示文摘In order to study intelligent fault diagnosis methods based on fuzzy neural network (NN) expert system and build up intelligent fault diagnosis for a type of missile weapon system, the concrete implementation of a fuzzy NN fault diagnosis expert system is given in this paper. Based on thorough research of knowledge presentation, the intelligent fault diagnosis system is implemented with artificial intelligence for a large-scale missile weapon equipment. The method is an effective way to perform fuzzy fault diagnosis. Moreover, it provides a new way of the fault diagnosis for large-scale missile weapon equipment. | Yang Jun Feng Zhensheng Zhang Xien & Liu Pengyuan Dept. of Missile Engineering, Ordnance Engineering College, Shijiazhuang 050003, P. R. China | 2001 | Journal of Systems Engineering and Electronics2001，12，1： | 7 |
| 5 | A new sequential learning algorithm for RBF neural networks显示文摘Due to their inherent imperfections, it is hard to use the static neural networks for nonlinear time-varying process modeling and prediction, and the minimal resource allocation network (MRAN) is difficult to be realized for its too many regulation parameters. A new sequential learning algorithm for radial basis function (RBF) neural networks based on local projection named Local Projection Network (LPN) is proposed in this paper. The results of validation for several benchmark problems with the new algorithm show that the presented LPN not only has the same level as M-RAN in network size and precision of the outputs, but also has fewer regulation parameters and is more predictable. | YANG Ge1, LV Jianhong1 & LIU Zhiyuan2 1. Department of Power Engineering, Southeast University, Nanjing 210096, China 2. Department of Power Engineering, Nanjing Institute of Technology, Nanjing 210013, China | 2004 | Science China(Technological Sciences)2004，47，4： | 5 |
| 6 | Modeling of CVI process in fabrication of carbon/carbon composites by an artificial neural network显示文摘The chemical vapor infiltration(CVI) process in fabrication of carbon-carbon composites is very complex and highly inefficient, which adds considerably to the cost of fabrication and limits the application of the material. This paper tries to use a supervised artificial neural network(ANN) to model the nonlinear relationship between parameters of isothermal CVI(ICVI) processes and physical properties of C/C composites. A model for preprocessing dataset and selecting its topology is developed using the Levenberg-Marquardt training algorithm and trained with comprehensive dataset of tubal C/C components collected from experimental data and abundant simulated data obtained by the finite element method. A basic repository on the domain knowledge of CVI processes is established via sufficient data mining by the network. With the help of the repository stored in the trained network, not only the time-dependent effects of parameters in CVI processes but also their coupling effects can be analyzed and predicted. The results show that the ANN system is effective and successful for optimizing CVI processes in fabrication of C/C composites. | 李爱军 李贺军 李克智 顾正彬 | 2003 | Science China(Technological Sciences)2003，46，2： | 5 |
| 7 | Recent Advances on Neural Headline Generation显示文摘Recently, neural models have been proposed for headline generation by learning to map documents to headlineswith recurrent neural network. In this work, we give a detailed introduction and comparison of existing work and recentimprovements in neural headline generation, with particular attention on how encoders, decoders and neural model trainingstrategies alter the overall performance of the headline generation system. Furthermore, we perform quantitative analysis ofmost existing neural headline generation systems and summarize several key factors that impact the performance of headlinegeneration systems. Meanwhile, we carry on detailed error analysis to typical neural headline generation systems in orderto gain more comprehension. Our results and conclusions are hoped to benefit future research studies. | Ayana Shi-Qi Shen Yan-Kai Lin Cun-Chao Tu Yu Zhao Zhi-Yuan Liu Mao-Song Sun | 2017 | Journal of Computer Science & Technology2017，32，4： | 5 |
| 8 | Utility-based bandwidth allocation algorithm for heterogeneous wireless networks显示文摘In next generation wireless network (NGWN), mobile users are capable of connecting to the core network through various heterogeneous wireless access networks, such as cellular network, wireless metropolitan area network (WMAN), wireless local area network (WLAN), and ad hoc network. NGWN is expected to provide high-bandwidth connectivity with guaranteed quality-of-service to mobile users in a seamless manner; however, this desired function demands seamless coordination of the heterogeneous radio access network (RAN) technologies. In recent years, some researches have been conducted to design radio resource management (RRM) architectures and algorithms for NGWN; however, few studies stress the problem of joint network performance optimization, which is an essential goal for a cooperative service providing scenario. Furthermore, while some authors consider the competition among the service providers, the QoS requirements of users and the resource competition within access networks are not fully considered. In this paper, we present an interworking integrated network architecture, which is responsible for monitoring the status information of different radio access technologies (RATs) and executing the resource allocation algorithm. Within this architecture, the problem of joint bandwidth allocation for heterogeneous integrated networks is formulated based on utility function theory and bankruptcy game theory. The proposed bandwidth allocation scheme comprises two successive stages, i.e., service bandwidth allocation and user bandwidth allocation. At the service bandwidth allocation stage, the optimal amount of bandwidth for different types of services in each network is allocated based on the criterion of joint utility maximization. At the user bandwidth allocation stage, the service bandwidth in each network is optimally allocated among users in the network according to bankruptcy game theory. Numerical results demonstrate the efficiency of the proposed algorithm. | CHAI Rong WANG XiuJuan CHEN QianBin SVENSSON Tommy | 2013 | Science China(Information Sciences)2013，56，2： | 4 |
| 9 | Remote sensing data classification using tolerant rough set and neural networks显示文摘BP algorithm of neural net is used more in remote sensing data classification. One of drawbacks of BP algorithm is the overall low function when the net is training. To avoid this kind of problem, the paper introduces the tolerant rough set for classification-preprocessing the training data to reduce the influence elements of the training convergence in order to improve the net training successful rate. ETM+ data of Beijing in May 2003 is selected in the study. ETM+ data before and after classification preprocessing, respectively, are used for BP (Back propaga-tion) training. The result shows that such a preprocessing not only compensates the drawback of BP algorithm when processing ETM+ data but also improves classification accuracy. | MA Jianwen HASI Bagan | 2005 | Science China Earth Sciences2005，48，12： | 3 |
| 10 | Optimal locations of monitoring stations in water distribution systems under multiple demand patterns: a flaw of demand coverage method and modification显示文摘A flaw of demand coverage method in solving optimal monitoring stations problem under multiple demand patterns was identified in this paper. In the demand coverage method, the demand coverage of each set of monitoring stations is calculated by accumulating their demand coverage under each demand pattern, and the impact of temporal distribution between different time periods or demand patterns is ignored. This could lead to miscalculation of the optimal locations of the monitoring stations. To overcome this flaw, this paper presents a Demand Coverage Index (DCI) based method. The optimization considers extended period unsteady hydrau- lics due to the change of nodal demands with time. The method is cast in a genetic algorithm framework for integration with Environmental Protection Agency Net (EPANET) and is demonstrated through example applica- tions. Results show that the set of optimal locations of monitoring stations obtained using the DCI method can represent the water quality of water distribution systems under multiple demand patterns better than the one obtained using previous methods. | Shuming LIU Wenjun LIU Jinduan CHEN Qi WANG | 2012 | Frontiers of Environmental Science & Engineering2012，6，2： | 3 |
| 11 | GOMA:functional enrichment analysis tool based on GO modules显示文摘Analyzing the function of gene sets is a critical step in interpreting the results of high-throughput experiments in systems biology. A variety of enrichment analysis tools have been developed in recent years, but most output a long list of significantly enriched terms that are often redundant, making it difficult to extract the most meaningful functions. In this paper, we present GOMA, a novel enrichment analysis method based on the new concept of enriched functional Gene Ontology (GO) modules. With this method, we systematically revealed functional GO modules, i.e., groups of functionally similar GO terms, via an optimization model and then ranked them by enrichment scores. Our new method simplifies enrichment analysis results by reducing redundancy, thereby preventing inconsistent enrichment results among functionally similar terms and providing more biologically meaningful results. | Qiang Huang Ling-Yun Wu Yong Wang Xiang-Sun Zhang | 2013 | Chinese Journal of Cancer2013，32，4： | 3 |
| 12 | Learning Bayesian network structure with immune algorithm显示文摘Finding out reasonable structures from bulky data is one of the difficulties in modeling of Bayesian network(BN), which is also necessary in promoting the application of BN. This paper proposes an immune algorithm based method(BN-IA) for the learning of the BN structure with the idea of vaccination. Furthermore, the methods on how to extract the effective vaccines from local optimal structure and root nodes are also described in details.Finally, the simulation studies are implemented with the helicopter convertor BN model and the car start BN model. The comparison results show that the proposed vaccines and the BN-IA can learn the BN structure effectively and efficiently. | Zhiqiang Cai Shubin Si Shudong Sun Hongyan Dui | 2015 | Journal of Systems Engineering and Electronics2015，26，2： | 3 |
| 13 | Type-Aware Question AnsweringAttention-Based Tree-Structuredover Knowledge Base withNeural Networks显示文摘Question answering (QA) over knowledge base (KB) aims to provide a structured answer from a knowledgebase to a natural language question. In this task, a key step is how to represent and understand the natural languagequery. In this paper, we propose to use tree-structured neural networks constructed based on the constituency tree tomodel natural language queries. We identify an interesting observation in the constituency tree: different constituentshave their own semantic characteristics and might be suitable to solve different subtasks in a QA system. Based on thispoint, we incorporate the type information as an auxiliary supervision signal to improve the QA performance. We call ourapproach type-aware QA. We jointly characterize both the answer and its answer type in a unified neural network modelwith the attention mechanism. Instead of simply using the root representation, we represent the query by combining therepresentations of different constituents using task-specific attention weights. Extensive experiments on public datasets havedemonstrated the effectiveness of our proposed model. More specially, the learned attention weights are quite useful inunderstanding the query. The produced representations for intermediate nodes can be used for analyzing the effectivenessof components in a QA system. | Jun Yin Wayne Xin Zhao Xiao-Ming Li | 2017 | Journal of Computer Science & Technology2017，32，4： | 3 |
| 14 | A new method for constructing infinite families of k-tight optimal double loop networks显示文摘The double loop network (DLN) is a circulant digraph with n nodes and outdegree 2. DLN has been widely used in the designing of local area networks and distributed systems. In this paper, a new method for constructing infinite families of k-tight optimal DLN is presented.method, where the number nk(t,a) of their nodes is a polynomial of degree 2 in t and contains a parameter a. And a conjecture is proposed. | CHEN Xiebin | 2006 | Science China Mathematics2006，49，4： | 3 |
| 15 | Study on Polyurethane/(vinyl ester resin) IPN Damping Materials显示文摘Polyurethane/(vinyl ester resin) interpenetrating polymer network (PU/VER IPN) materials with broad temperatureranges and excellent damping properties from Iow temperature to room temperature were prepared. The influenceof comonomers and component ratios on the compatibility and damping properties of IPN materials was studied byDMA which indicates that such properties are improved by introducing acrylic esters instead of polystyrene (PSt)into VER comonomer system. The detected results of microstructure by AFM show that the phase ranges of thedual-phase continuous IPN materials obtained are both in nanometer scale. The results of mechanical propertiesshow that IPN materials show the regulation from elastic deformation to brittle deformation with the increase of VERproportion. | Chuanli QIN Dongyan TANG Jun CAI Jusheng.ZHANG Weimin CAI Xiaodong SUN | 2003 | Journal of Materials Science & Technology2003，19，z1： | 3 |
| 16 | Seabed Classification Using BP Neural Network Based on GA显示文摘Side scan sonar imaging is one of the advanced methods for seabed study. In order to be utilized in other projects, such as ocean engineering, the image needs to be classified according to the distributions of different classes of seabed materials.In this paper, seabed image is classified according to BP neural network, and Genetic Algorithm is adopted in train network in this paper. The feature vectors are average intensity, six statistics of texture and two dimensions of fractal. It considers not only the spatial correlation between different pixels, but also the terrain coarseness. The texture is denoted by the statistics of the co-occurrence matrix. Double Blanket algorithm is used to calculate dimension.Because a uniform fractal may not be sufficient to describe a seafloor, two dimensions are calculated respectively by the upper blanket and the lower blanket. However, in sonar image, fractal has directivity, i.e. there are different dimensions in different direction. Dimensions are different in acrosstrack and alongtrack, so the average of four directions is used to solve this problem. Finally, the real data verify the algorithm. In this paper, one hidden layer including six nodes is adopted. The BP network is rapidly and accurately convergent through GA. Correct classification rate is 92.5% in the result. | Yang Fanlin1, Liu Jingnan2 1. GPS Engineering Research Center, Wuhan University, Wuhan 430079, China. 2. Presidential Secretariat, Wuhan University, Wuhan 430079, China | 2003 | Acta Oceanologica Sinica2003，22，4： | 3 |
| 17 | Implementing of the JPWSPC method in RIV1H for unsteady flow modeling in general river networks显示文摘RIV1H is the stand-alone hydraulic program of CE-QUAL-RIV1, a longitudinal hydraulic and water quality model developed by U.S. Army Corps of Engineers Waterways Experiment Station. RIV1H solves the Saint-Venant equations using the widely accepted four-point implicit Preissmann scheme, and the resulting nonlinear equations are solved using the Newton-Raphson method. RIV1H is capable of simulating multiple branches, and in-stream hydraulic control structures. It treats tributary networks using a double sweep algorithm based on upstream ordering of the branches. It treats the control structures following a downstream solution order, which also is based on the upstream ordering of the branches. Since an upstream ordering cannot be achieved for looped networks, RIV1H is only applicable to non-looped tributary networks. In the current study, the junction-point water stage prediction and correction (JPWSPC) method is extended to take into account the control structures and the method is used to improve the RIV1H model, enabling it to be applied to both non-looped and looped networks with in-stream hydraulic control structures. The JPWSPC method makes the linear equation system for each segment complete while maintaining the banded property, thus the system can be independently and efficiently solved. It has the advantages to be efficient, robust, and very suitable for parallel computing. The improved RIV1H model was tested using two idealized networks and the results demonstrated the success of the improvement. | Dejun Zhu Yongcan Chen | 2019 | International Journal of Sediment Research2019，34，4： | 2 |
| 18 | Optimization of thermomechanical processes in Cu-Cr-Zr lead frame alloy using neural networks and genetic algorithms显示文摘The thermomechanical treatment process is effective in enhancing the properties of the lead frame copper alloy. In this study, an optimal pattern of the thermomechanical processes for Cu-Cr-Zr was investegated using an intelligent control technique consisting of neural networks and genetic algorithms. The input parameters of the artificial neural network (ANN) are the reduction ratio of cold rolling, aging temperature and aging time. The outputs of the ANN model are the two most important properties of hardness and conductivity. Based on the successfully trained ANN model, genetic algorithms (GA) are used to optimize the input parameters of the model and select perfect combinations of thermomechanical processing parameters and properties. The good generalization performance and optimized results of the integrated model are achieved. | SU Juanhua1,2, LIU Ping2, DONG Qiming2 & LI Hejun1 1. College of Materials Science and Engineering, Northwestern Polytechnical University, Xi’an 710072, China 2. College of Materials Science and Engineering, Henan University of Science and Technology, Luoyang 471003, China | 2005 | Science China(Technological Sciences)2005，48，5： | 2 |
| 19 | Prediction Method of Vessel Maintenance Outlay Based on the BP Neural Network显示文摘With the development of technology, the performance of vessel equipment is improved, the structure is more complicated, the automation level is enhanced, the source needed by maintenance is increased and the outlay is rising day by day. For these questions, this paper analyzes the factors that affect the outlay of equipment maintenance, and describes the computational principle of the BP (back propagation) artificial neural network and its applications in the maintenance of naval ship and craft. Finally, a dynamic investment prediction model of outlay for the military equipment maintenance is designed. It is important for decreasing the entire ilfe period outlay and drawing up the maintenance plan and programming to analyze the position and action of maintenance outlay in entire life period outlay. | 郭冰冰 黎放 王威 | 2002 | Journal of Systems Engineering and Electronics2002，13，3： | 2 |
| 20 | Error resilient concurrent video streaming over wireless mesh networks显示文摘In this paper, we propose a multi-source multi-path video streaming system for supporting high quality concurrent video-on-demand (VoD) services over wireless mesh networks (WMNs), and leverage forward error correction to enhance the error resilience of the system. By taking wireless interference into consideration, we present a more realistic networking model to capture the characteristics of WMNs and then design a route selection scheme using a joint rate/interference-distortion optimiza- tion framework to help the system optimally select concurrent streaming paths. We mathematically formulate such a route selec- tion problem, and solve it heuristically using genetic algorithm. Simulation results demonstrate the effectiveness of our proposed scheme. | CHUAH Chen-nee YOO Ben S.J. | 2006 | Journal of Zhejiang University-Science A(Applied Physics & Engineering)2006，7，5： | 2 |
您所在的机构暂未开通此部分的功能，请联系管理员