Entity

Time filter

Source Type

Nanjing, China

Nanjing University of Aeronautics and Astronautics is a university located in Nanjing, Jiangsu province, China. It was established in October 1952. In Chinese, the university name is sometimes shortened to Nanhang . The university is operated by Ministry of Industry and Information Technology and is one of China's leading universities on research and education. It is regarded as one of the top engineering universities in China and also has a great influence on China's aerospace industry. Wikipedia.


Dai Q.,Nanjing University of Aeronautics and Astronautics
Knowledge-Based Systems | Year: 2013

Ensemble pruning is crucial for the consideration of both efficiency and predictive accuracy of an ensemble system. This paper proposes a new Competitive technique for Ensemble Pruning based on Cross-Validation (CEPCV). The data to be learnt by neural computing models are mostly drifting with time and environment, therefore a dynamic ensemble pruning method is indispensable for practical applications, while the proposed CEPCV method is just the kind of dynamic ensemble pruning method, which can realize on-line ensemble pruning and take full advantage of potentially valuable information. The algorithm naturally inherits the predominance of cross-validation technique, which implies that those networks regarded as winners in selective competitions and chosen into the pruned ensemble have the "strongest" generalization capability. It is essentially based on the strategy of "divide and rule, collect the wisdom", and might alleviate the local minima problem of many conventional ensemble pruning approaches only at the cost of a little greater computational cost, which is acceptable to most applications of ensemble learning. The comparative experiments among the four ensemble pruning algorithms, including: CEPCV and the state-of-the-art Directed Hill Climbing Ensemble Pruning (DHCEP) algorithm and two baseline methods, i.e. BSM, which chooses the Best Single Model in the initial ensemble based on their performances on the pruning set, and ALL, which reserves all network members of the initial ensemble, on ten benchmark classification tasks, demonstrate the effectiveness and validity of CEPCV. © 2012 Elsevier B.V. All rights reserved. Source


Chen K.,Nanjing University of Aeronautics and Astronautics
International Journal of Production Economics | Year: 2012

This paper investigates the coordination mechanism for supply chain with one manufacturer and multiple competing suppliers in the electronic market. We first study two conventional price-only policies, including wholesale price policy and catalog policy, based on the reverse Vickrey auction, and show that both the buyer and the powerful suppliers (with production cost less than a special threshold value) prefer catalog policy to wholesale price policy, and the powerless suppliers prefer wholesale price policy to catalog policy. Simultaneously, neither policy can coordinate the channel composed of the manufacturer and the winning supplier. We also show that a quantity discount policy cannot coordinate the supply chain with competing suppliers unless a kind of restriction is imposed. The aim of the paper is to explore a coordination mechanism, i.e., the price-restricted quantity-discount policy. Pareto analysis indicates that the manufacturer and the winning supplier will realize the 'win-win' situation, and the channel can also be coordinated. A key managerial implication of our study is that additional restrictive condition may be necessary to eliminate system inefficiency. Some numerical examples are also given to illustrate management insights. © 2012 Elsevier B.V. All rights reserved. Source


Tan X.,Nanjing University of Aeronautics and Astronautics | Triggs B.,French National Center for Scientific Research
IEEE Transactions on Image Processing | Year: 2010

Making recognition more reliable under uncontrolled lighting conditions is one of the most important challenges for practical face recognition systems. We tackle this by combining the strengths of robust illumination normalization, local texture-based face representations, distance transform based matching, kernel-based feature extraction and multiple feature fusion. Specifically, we make three main contributions: 1) We present a simple and efficient preprocessing chain that eliminates most of the effects of changing illumination while still preserving the essential appearance details that are needed for recognition; 2) We introduce local ternary patterns (LTP), a generalization of the local binary pattern (LBP) local texture descriptor that is more discriminant and less sensitive to noise in uniform regions, and we show that replacing comparisons based on local spatial histograms with a distance transform based similarity metric further improves the performance of LBP/LTP based face recognition; and 3) We further improve robustness by adding Kernel principal component analysis (PCA) feature extraction and incorporating rich local appearance cues from two complementary sourcesGabor wavelets and LBPshowing that the combination is considerably more accurate than either feature set alone. The resulting method provides state-of-the-art performance on three data sets that are widely used for testing recognition under difficult illumination conditions: Extended Yale-B, CAS-PEAL-R1, and Face Recognition Grand Challenge version 2 experiment 4 (FRGC-204). For example, on the challenging FRGC-204 data set it halves the error rate relative to previously published methods, achieving a face verification rate of 88.1% at 0.1% false accept rate. Further experiments show that our preprocessing method outperforms several existing preprocessors for a range of feature sets, data sets and lighting conditions. © 2010 IEEE. Source


Qing H.,Nanjing University of Aeronautics and Astronautics
Computational Materials Science | Year: 2014

The influence of particle arrangements and interface strengths on the mechanical behavior of the particle reinforced metal-matrix composite (MMC) is investigated under different loading conditions in this work. During the loading process, three different failure mechanisms are distinguished in MMC: ductile failure in metal matrix, brittle failure in SiC particles and interface debonding between matrix and particles. The damage models based on the stress triaxial indicator and maximum principal stress criterion are developed to simulate the failure process of metal matrix and SiC particles. Meanwhile, 2D cohesive element is utilized to describe the debonding behavior of interface. Series of numerical experiments are performed to study the macroscopic stress-strain relationships and microscale damage evolution in MMCs under different loading conditions. An agreement between the simulation results and the experimental data is obtained. © 2014 Elsevier B.V. All rights reserved. Source


Xiao H.,Nanjing Southeast University | Xie S.,Nanjing University of Aeronautics and Astronautics
IEEE Transactions on Power Electronics | Year: 2012

Characterized by low leakage current and low voltage stress of the power device, a neutral point clamped three-level inverter (NPCTLI) is suitable for a transformerless photovoltaic (PV) grid-connected system. Unfortunately, the shoot-through problem of bridge legs still exists in an NPCTLI, so its operation reliability is degraded. An improved three-level grid-connected inverter is proposed based on the NPCTLI and the dual-buck half-bridge inverter (DBHBI), and which avoids the shoot-through problem. The proposed topology guarantees no switching-frequency common-mode voltage and no shoot-through risk. Furthermore, the freewheeling diode of bridge legs of the DBHBI can be removed taking into consideration the unity power factor of grid current, and a straightforward topology is thus derived. The new topology is referred to as split-inductor NPCTLI (SI-NPCTLI). The operation mode, common-mode characteristic, and control strategy are analyzed. Finally, both the simulation and the experimental results of a 1-kW SI-NPCTLI prototype verify the analysis. © 2011 IEEE. Source

Discover hidden collaborations