Hunan Institute of Technology

www.hnit.edu.cn
Hengyang, China
SEARCH FILTERS
Time filter
Source Type

Liu J.,Hunan Institute of Technology
International Journal of Modeling, Simulation, and Scientific Computing | Year: 2017

Online sequential extreme learning machine (OS-ELM) for single-hidden layer feedforward networks (SLFNs) is an effective machine learning algorithm. But OS-ELM has some underlying weaknesses of neglecting time series timeliness and being incapable to provide quantitative uncertainty for prediction. To overcome these shortcomings, a time series prediction method is presented based on the combination of OS-ELM with adaptive forgetting factor (AFF-OS-ELM) and bootstrap (B-AFF-OS-ELM). Firstly, adaptive forgetting factor is added into OS-ELM for adjusting the effective window length of training data during OS-ELM sequential learning phase. Secondly, the current bootstrap is developed to fit time series prediction online. Then associated with improved bootstrap, the proposed method can compute prediction interval as uncertainty information, meanwhile the improved bootstrap enhances prediction accuracy and stability of AFF-OS-ELM. Performances of B-AFF-OS-ELM are benchmarked with other traditional and improved OS-ELM on simulation and practical time series data. Results indicate the significant performances achieved by B-AFF-OS-ELM. © 2017 World Scientific Publishing Company


Weimin G.,Hunan Institute of Technology
International Journal of Future Generation Communication and Networking | Year: 2016

Choosing a reasonable Rescue route guarantees emergency rescue command center to start the rescue work timely and effectively. In this paper, use the optimal path choice of public emergencies as an entry point of research build a post-disaster Ad Hoc net through wireless sensor technology, use the limited sensor resources to meet the rescue vehicle path optimization problem. Ant Colony Algorithm absorb the behavior of insect kingdom ant, use internal search mechanism to introduce sensor networks useful information, Fully consider the impact of the road traffic rescue vehicle travel time because of changes the traffic capacity, from rescue vehicles transport resources’ two factors -the timeliness and economy, maximize meet the transport requirements of post-disaster emergency relief material to reduce the aid does not lead to timely the direct and indirect losses, maximize meet the transport requirements of post-disaster emergency relief material to reduce the aid does not lead to timely the direct and indirect losses. The simulation results show that the design of the optimal path to solve the disaster relief vehicles have obvious advantages and important practical significance. It Improved ant colony algorithm can quickly build a optimal routing, improve the efficiency of path optimization, effectively reduce the node energy consumption and ensure the whole network energy balance. © 2016 SERSC.


Guo C.,Hunan Institute of Technology
Zhongguo Tiedao Kexue/China Railway Science | Year: 2017

With minimizing the weighted comprehensive value of the total time operation for shunting locomotive's placing-in and taking-out of wagons, total wagons' preparation time, total wagons' travel kilometers as optimization objectives, regarding the capability of locomotive and the priority that the transferring operations requires between operating sites as constraints, a multi-objective optimization model for scheme of wagons' placing-in and taking-out in branch-shaped operating sites is formulated. Encoding the solutions with natural number, constructing a solution that satisfies the priority of the transferring as our initial solution, transforming the constraint of traction capability of shunting locomotive into penalty function and regarding it as evaluation function with the objective function, and executing a random local search with three methods of neighborhood structure operation successively, we solve the model with simulated annealing algorithm. A case of placing-in and taking-out wagons in a railway station is used to verify our model and algorithm. Results show that the model is consistent with the requirements of scheme for wagons' placing-in and taking-out and actual operation, and the computation efficiency and results can meet the requirements of railway operation. © 2017, Editorial Department of China Railway Science. All right reserved.


Xia L.,Central South University | Yao S.,Central South University | Yao S.,Hunan Institute of Technology
Frontiers in Behavioral Neuroscience | Year: 2015

Numerous studies have reported on the roles of genetic factors in the development of depression in adolescents and young adults. However, there are few systematic reviews that update our understanding of adolescent depression with the biological findings identifying the roles of gene expression and/or polymorphism(s). This review systematically summarized the findings that clearly identified the contribution of a gene to the risk of depression in adolescents between the ages of 10 and 19 years old and young adults between the ages of 20 and 25 years old. Data were obtained through searching PubMed, Embase, and Web of Science. A total of 47 studies on early adolescence and three studies on young adults were included in the current review. Most articles studied genes in the serotonergic system (n = 26), dopaminergic system (n = 3), and the Brain-derived neurotropic factor (BDNF) gene (n = 12). 92.3% of studies (24/26) identified positive associations of 5-HTTLPR polymorphism with depressive illness or depressive symptoms. 83.3% of studies (10/12) found positive association between BDNF Val66Met genotype and adolescent depressive symptoms. More studies should be conducted on the 18 genes reported in a few studies to clarify their roles in the risk for adolescent depression. © 2015 Xia and Yao.


Xiao K.,Hunan Institute of Technology | He X.-K.,Hunan First Normal University | Zhu J.-Y.,Beijing Normal University
Physics Letters, Section B: Nuclear, Elementary Particle and High-Energy Physics | Year: 2013

Phenomenological effect of the super-inflation in loop quantum cosmology (LQC) is discussed. We investigate the case that the Universe is filled with the interacting field between massive scalar field and radiation. Considering the damping coefficient Γ as a constant, the changes of the scale factor during super-inflation with four different initial conditions are discussed, and we find that the changes of the scale factor depends on the initial values of energy density of the scalar field and radiation at the bounce point. But no matter which initial condition is chosen, the radiation always dominated at the late time. Moreover, we investigate whether the super-inflation can provide enough e-folding number. For the super-inflation starts from the quantum bounce point, the initial value of Hubble parameter H(ti)~0, then it is possible to solve the flatness problem and horizon problem. As an example, following the method of [18] to calculate particle horizon on the condition that the radiation dominated at bounce point, and we find that the Universe has had enough time to be homogeneous and isotopic. © 2013 Elsevier B.V.


Li L.,Hunan Institute of Technology
Journal of Alloys and Compounds | Year: 2013

Microstructure and texture evolution during super plastic deformation of Mg-Gd-Y-Zr extruded rods were investigated in the study. For the purposes, the tensile tests under super plastic deformation condition were conducted at different strain levels. The static annealing tests were performed to reveal the micro structural evolution only influenced by thermal load. Grain structure analysis proves that initial grain refinement during the preheating stage is driven by store energy inherited from extrusion deformation. Microstructure and texture analysis suggests that deformation mechanisms are initial grain boundary sliding and subsequent slip creep. Moreover, the β phases (Mg5 (Gd, Y)) on (sub) grain boundaries hinder grain boundary sliding and continuous dynamic recrystallization. Macro-texture evolution shows that texture components varied according to the transition of deformation mechanisms. Micro-texture measurement using EBSD proves that β phase has undergone severe deformation before elongation-to-failure. © 2012 Elsevier B.V. All rights reserved.


Li L.,Hunan Institute of Technology
Materials Science and Engineering A | Year: 2011

Microstructural evolution and texture of a cast Mg-9Gd-4Y-0.6Zr ingot under hot compression were studied in this paper. Post-deforming microstructures were characterized by optical microscopy, scanning electron microscopy and transmission electron microscopy, while crystallographic orientation information was obtained from X-Ray macro-texture measurement and EBSD micro-texture analysis. Dynamic recrystallization (DRX) initiated from the deformation bands (DB) forming on original grain boundaries; the DB became widen with continuously conversion of low-angle-boundary grains into high-angle-boundary grains. The tendency of strain localization increased with Z parameter. The macro-texture analysis indicates that uniaxial compression yielded out the randomized basal texture component. This texture component was found to be strengthened with increasing Z parameter. The micro-texture analysis shows that the deviation from the ideal basal texture arose from orientated growth within DBs. Moreover, the localization deformation promoted dynamic precipitation within DBs, which inhibited the development of DRX. © 2011 Elsevier B.V.


Liu K.,Hunan Institute of Technology
Gaodeng Xuexiao Huaxue Xuebao/Chemical Journal of Chinese Universities | Year: 2014

α-Halomethylketones are building blocks for the synthesis of various compounds due to their selective transformations with different reagents. Moreover, a wide range of α-halomethylketone derivatives attract increasing interests in pharmaceutical chemistry. α-Halomethylketones are also widespread used as intermediates in organic synthesis, such as in Favorskii rearrangement and Reformatsky reaction. However, their direct synthesis from the corresponding methylktones by N-halosuccinimides, molecular halogen and metal halides has been plagued by poor regioselectivity and over-halogenation. For the above reasons, a simple, convenient and environmentally benign protocol to synthesize the α-halomethylketones was developed by catalytic hydration of haloalkynes, which can be easily prepared in one step from commercial terminal alkynes. In the presence of 3%(molar fraction) BrettPhosAuNTf2 and 3 mmol H2O in dichloroethane at room temperature, a broad range of haloalkynes were converted into α-halomethylketones in excellent yield through gold-catalyzed hydration reaction. The presented methodology will provide new strategies for α-halomethylketones-type drug design and synthesis, which has important academic significance and application value.


Kuang F.,Nanjing University of Science and Technology | Kuang F.,Hunan Institute of Technology | Xu W.,Nanjing University of Science and Technology | Xu W.,Changsha University of Science and Technology | Zhang S.,Hunan Institute of Technology
Applied Soft Computing Journal | Year: 2014

A novel support vector machine (SVM) model combining kernel principal component analysis (KPCA) with genetic algorithm (GA) is proposed for intrusion detection. In the proposed model, a multi-layer SVM classifier is adopted to estimate whether the action is an attack, KPCA is used as a preprocessor of SVM to reduce the dimension of feature vectors and shorten training time. In order to reduce the noise caused by feature differences and improve the performance of SVM, an improved kernel function (N-RBF) is proposed by embedding the mean value and the mean square difference values of feature attributes in RBF kernel function. GA is employed to optimize the punishment factor C, kernel parameters σ and the tube size É of SVM. By comparison with other detection algorithms, the experimental results show that the proposed model performs higher predictive accuracy, faster convergence speed and better generalization. © 2014 Elsevier B.V.


Xin J.,Hunan Institute of Technology | Huang B.,Hunan Institute of Technology
Journal of Agricultural and Food Chemistry | Year: 2014

A greenhouse experiment was conducted to compare the subcellular distribution and chemical forms of cadmium (Cd) in roots, stems, leaves, and fruits between a low-Cd cultivar (Yeshengchaotianjiao, YCT) and a high-Cd cultivar (Jinfuzaohuangjiao, JFZ) of hot pepper (Capsicum annuum L.). The Cd concentrations in the root's subcellular fractions, and in all chemical forms in roots, were 1.85-4.88- and 1.84-4.90-fold higher, respectively, in YCT than in JFZ. Compared with JFZ, YCT had significantly lower Cd concentrations in the subcellular fractions (1.10-2.42-fold) of stems and leaves and in almost all chemical forms (1.17-2.97-fold) in the stems and leaves. Also, in fruits, the concentrations of Cd in the cell wall and soluble fractions were 1.18-2.24-fold significantly lower in YCT than in JFZ, and there were lower Cd concentrations (1.36-2.08-fold) in the chemical forms in YCT than in JFZ. © 2013 American Chemical Society.

Loading Hunan Institute of Technology collaborators
Loading Hunan Institute of Technology collaborators