Table of Content

15 October 2017, Volume 34 Issue 5 Previous Issue    Next Issue

Select

Information Flow Characteristics Analysis of Vehicular Ad-hoc Network ZHANG Hong, LV Yue-jing 2017, 34 (5):  449-457.  doi: 10.3969/j.issn.1005-3085.2017.05.001 Abstract ( 107 )   PDF (330KB) ( 265 )   Save In order to reveal the dynamic topological characteristics of Vehicular Ad-hoc Network (VANET), predict the behavior of VANET and alleviate the traffic congestion, extensive simulations of information propagation under normal circumstances from the microscopic point of view, we investigate the information flow distribution characteristics of VANET and the relationship between its information flow and degree of the node under different traffic demands in VANET. Firstly, the relationships among the degree, the degree distribution index and the information flow are proposed, and then scale-free network model is established with static and dynamic methods based on complex network theory, and when the parameters are varied, the change law of information flow is discussed by numerical simulation. Furthermore, the characteristics of information flow in capacity constraints are studied by means of nonlinear dynamics theory. Numerical results show that the information flow index is a general characteristic parameter of the scale-free network, and those larger degree nodes have a greater effect on the network and which also have faster propagation speed. We also find that the relationship of information flow and nodes degree obeys the power law distribution. Hub nodes may be overburdened when the information flow demand increases, and many information flows may choose other nodes to avoid communicating with the Hubs nodes, then the smaller degree of nodes will undertake this part of the information flow. Related Articles | Metrics

Select

Numerical Simulation of Unsteady Heat Transfer Process for Large Floating Roof Oil Tanks SUN Wei, CHENG Qing-lin, LI Yu-chun, SUN Zhe, SUN Hai-ying, LIU Yang 2017, 34 (5):  458-468.  doi: 10.3969/j.issn.1005-3085.2017.05.002 Abstract ( 100 )   PDF (3716KB) ( 232 )   Save With the rapid development of oil storage construction, the tank size is developing towards large size and being able to adapt to extreme conditions. In order to avoid oil solidification in the tanks and other safety accidents caused by too low oil temperature, the law of oil temperature field in tanks needs to be calculated accurately. Based on the law of conservation of energy, the heat transfer coefficient of tanks can be solved by fractional steps numerical algorithm, then the unsteady heat transfer discrete equation is solved by numerical method which is set by the Taylor series expansion method. The application analysis of the $10\times 10^4$m$^{3}$ floating roof tank in Daqing Oilfield shows that, the oil temperature drop rate increases gradually with the decrease of ambient temperature, and the oil temperature is higher and the temperature drop rate is smaller in the tank with higher tank level and larger volume. The research and analysis provide the important data for optimizing the storage design of large floating tanks and ensuring the safety and economic operation of oil depot. Related Articles | Metrics

Select

Improved Hidden Markov Model and Its Application in Financial Forecasting XU Zhu-jia, XIE Rui, LIU Jia, MEI Yu 2017, 34 (5):  469-478.  doi: 10.3969/j.issn.1005-3085.2017.05.003 Abstract ( 433 )   PDF (400KB) ( 409 )   Save Hidden Markov model (HMM) has been widely applied to many fields. This paper tries to improve current HMMs from different aspects and then applies the improved HMM to financial forecasting. Firstly, by fixing the initial points for the K-means clustering algorithm so that its clustering results are more stable, we use the resulting K-means clustering algorithm to seek better initial values for the Baum-Welch algorithm. To improve the forecasting accuracy, we apply the following new techniques: we choose the model parameters obtained from the Baum-Welch algorithm as the inputs for the Vertibi algorithm to determine the optimal sequence of the hidden states, and we repartition the observing vector. Then we determine the sets of observing vectors corresponding to the different hidden states. Based on the outputs of the Vertibi algorithm, we recompute the means and variances of different classes of observing vectors. The resulting mean vector and variance-covariance matrix are taken as the initial values for the Baum-Welch algorithm, which finally finds the optimal model parameters for HMM. Last but not least, instead of the existing methods seeking similar movements of the practice in the historical interval, we not only obtain the fine expression step of conditional probability, and by maximizing the conditional probability values, we derive better predictive value. Numerical results based on the real trading data of Chinese stock markets indicate the superiority of the improved HMM.  Related Articles | Metrics

Select

Research on Traffic Anomaly Detection Method Based on the Logistic Regression Model HOU Ai-hua, GAO Wei, WANG Lin 2017, 34 (5):  479-489.  doi: 10.3969/j.issn.1005-3085.2017.05.004 Abstract ( 315 )   PDF (366KB) ( 374 )   Save Network traffic is a basic data source of anomaly detection, and the accurate description of its behavioral characteristics plays an important role in real-time network abnormal behavior detection. To solve the problem of traffic anomaly detection, a logistic regression model-based network traffic anomaly detection method is proposed in this paper. By analyzing several basic characteristics of network traffic such as source IP and destination IP, the training machine of network abnormal and normal behaviors is constructed. Then, the mining model of anomaly network traffic is established by using logical regression. To valid the effectiveness of the proposed mining model, real network traffic collected by our lab is applied to test the model. Experimental results show that the proposed mining model of the network abnormal traffic is able to yield high accuracy, and achieve real-time performance as well. Related Articles | Metrics

Select

Research on Stability of Large-scale Adversarial Service Networks under Priority LI Quan-lin, ZHANG Yu 2017, 34 (5):  490-506.  doi: 10.3969/j.issn.1005-3085.2017.05.005 Abstract ( 130 )   PDF (347KB) ( 224 )   Save Recently, internet of things, big data and cloud computing provide new information technologies for a variety of large-scale service systems and for their resource management and task scheduling, and they also lead to many crucial characteristics, such as, large scale, loose coupling, cross organizations, heterogeneity and so on. To study the large-scale service systems, the adversarial networks becomes a hot and interesting research direction through a simply deterministic hypothesis and an upper bound estimation for system stability. In this paper, we firstly analyze a directed ring of service adversarial network with priority, and develop an upper bound estimation method by means of a mathematical modeling and analysis, hence also propose a new discussion of system stability. Then, we study the performance measures of this network, and also provide an upper bound of the running time that each customer is served in the adversarial network. Finally, we give some numerical examples to demonstrate how the upper bounds of the total number of customers and of the running time of each customer depend on some key parameters of this adversarial network. Related Articles | Metrics

Select

A Non-monotone Hybrid Newton Method for Solving the Variational Inequality Problems GONG En-long, WANG Xuan-zhan, GAO Miao-miao, DU Xiao-yu, SUN Qing-ying 2017, 34 (5):  507-516.  doi: 10.3969/j.issn.1005-3085.2017.05.006 Abstract ( 109 )   PDF (168KB) ( 239 )   Save In this paper, the variational inequality problem is transformed as an unconstrained optimization problem through the generalized $D$-gap function. A non-monotone hybrid Newton method based on Zhang H.C.'s non-monotone line search technique is proposed for minimizing the general form of the generalized $D$-gap function. Then, the global convergence property of the algorithm is analyzed. Under some proper conditions, we prove that the algorithm is globally quadratically convergent. Moreover, we obtain a global error bound of the algorithm when the mapping $F$ is strongly monotone without Lipschitz continuous. Numerical results indicate that the new algorithm is efficient. Related Articles | Metrics

Select

Penalty Finite Element Approximation for the Brinkman-Forchheimer Equations LIU De-min 2017, 34 (5):  517-533.  doi: 10.3969/j.issn.1005-3085.2017.05.007 Abstract ( 135 )   PDF (233KB) ( 254 )   Save Brinkman-Forchheimer equations (BF equations) describe the motion of the incompressible fluid under the strong nonlinearities. The accurate treatment of the incompressibility condition is critical for the numerical treatment of the BF equations. The penalty treatment is introduced to relax the incompressibility condition. In order to obtain the well-posedness  of the penalty problem, the pressure term is eliminated by using the penalty term, and an equivalence between the monotonous nonlinear elliptical problem and a minimization problem of corresponding energy functional is proposed. From the LBB condition, the existence and uniqueness of the variational problem are obtained. The convergence with respected to the penalty parameter is proved. Finally, the existence and uniqueness of the finite dimensional approximating problem are derived, and the error estimate based on the conforming finite element discretization is obtained. Numerical results show that the penalty finite element approximation is effective. Related Articles | Metrics

Select

The Effects of Spatial Variation in Lotka-Volterra Competition-diffusion Model YUAN Hai-long, LI Yan-ling 2017, 34 (5):  534-550.  doi: 10.3969/j.issn.1005-3085.2017.05.008 Abstract ( 103 )   PDF (154KB) ( 389 )   Save In this paper, a two-species Lotka-Volterra competition-diffusion model with homo-geneous Neumann boundary conditions is considered. The effect of spatial heterogeneity and spatial homogeneity of environment on two competing species and their different competition abilities are studied. In particular, we consider the distribution of resources is heterogeneous for one species but homogeneous for another species with the different total resources in the weak competition. It turns out to be that some parameters play very important roles in this model. The existence, the stability of coexistence state of system is considered, and hence the unique coexistence state and the globally asymptotically stable of the coexistence state of system, and any semi-trivial solution of system can be established under some suitable conditions. Moreover, some limiting behaviors of coexistence state as the dispersal rates are also studied. Our results show that the dynamics of system is very complicated for some general parameters. The proposed method of analysis is based on spectral analysis and monotone dynamical systems theory. Related Articles | Metrics

Select

Conjugate Direction-optimized Iterative Learning Control for Discrete Linear Time-invariant Systems YANG Xuan, RUAN Xiao-e 2017, 34 (5):  551-562.  doi: 10.3969/j.issn.1005-3085.2017.05.009 Abstract ( 143 )   PDF (137KB) ( 524 )   Save In this paper, an iterative learning control profile for a class of discrete linear time-invariant systems is designed by employing a conjugate direction optimization scheme. The control design is formulated in the trial/iteration domain on basis of the lifted system representation, where both input and output sampled during the trials are grouped into super vectors. In this form, the control input signal of the next trial is generated by compensating for the current one with a searching direction described as the current tracking error vector minus its projections on historical searching directions. Both monotonic convergence and quadratic termination of the proposed iterative learning control algorithm are analyzed by the technique of the mathematical induction according to the properties of the conjugate direction. Effectiveness and validity of the control algorithm are illustrated in numerical simulations. Moreover, compared with existing proportional-type and norm-optimal iterative learning control methods, the superiority of the proposed method is also demonstrated. Related Articles | Metrics

Select

Best Simultaneous Approximations in Bochner-Lebesgue Spaces WEI Hai-hua, XU Jing-shi 2017, 34 (5):  563-570.  doi: 10.3969/j.issn.1005-3085.2017.05.010 Abstract ( 94 )   PDF (111KB) ( 261 )   Save In this paper, we consider the best simultaneous approximations in Bochner-Lebesgue spaces with respective to Minkowski' norms in Euclidean spaces. Firstly, we give a characterization of best simultaneous approximations by the distance functions. Then, by applying this characterization and a measurable selection theorem we show the simultaneous proximinality of a Bochner-Lebesgue space whose functions take values in a closed separable subspace is equivalent to the simultaneous proximinality of the closed separable subspace. Finally, we conclude that for their equivalence, the separability of the subspace is necessary. Related Articles | Metrics