Abstract:

Abstract:

Abstract: Bacterial optimization was a kind of swarm intelligence approach proposed recent year. In this paper, troduced several typical bacterial optimization algorithms and analyzed the keywords of relevant literatures were analyzed. Based on the frequency of these keywords, the algorithm improvement researches were reviewed. It manly contains four main aspects: parameter and structure improvement、algorithm hybrid、operator improvement and multi-objective reconstruction. Applications of bacterial optimization on engineering problems were reviewed as well. At last, the research direction in future was prospected.

Abstract: Efficient scheduling algorithm plays a key role in improving the efficacy of tracking and data relay satellite system (TDRS). Scheduling of TDRS aims to scientifically allocate TDRS resources according to the task application information from the users, such that maximal task requirements are met and the optimal activity schedule is made for the TDRS system. The mathematical model is constructed for the TDRS scheduling problem with the consideration of multiple slide windows in real-world requirements. The objective of the model is to maximize the task completion rate and the expectation satisfaction degree of users. The involved constraints include task requirement constraints and resource using constraints. A heuristic algorithm based on time freedom degree is proposed to solve the model, which includes four operators, i.e., evaluation of the time freedom degree of each task, matching between tasks and resources, task insertion and resource update. At last, extensive experimental simulation demonstrates the effectiveness of the proposed algorithm.

Abstract: Traditional appearance transfer methods automatically took advantage of dense patch matching method and requried the source image and exemplar images had higher visual similarity. However, such pairs were rare. Furthormore, existing methods used a single source image, which were unable to deal with complex image appearance transfer. To solve the above problems, a multi-source image appearance transfer was proposed by editing propagation. Firstly user intuitively built regions, which were likely to have similar color statistics by strokes and general precise stokes by a new automatically stroke correction method. Then edit propagation method was employed to find all the pixels with the similarappearance. To deal with the complex image, a multi-source image based automatically transfer method was introduced. Finally, our method performed color transfer or texture transfer between designated regions and generated the result of appearance transfer. Various experimental results verified the effectiveness of the proposed method.

Abstract: Pharmscophore-based virtual screening technology had higher accuracy rate, faster calculation speed and lower research cost than traditional screening technology. However, the microscopic and struectural complexity of the molecule posed a challenge to the user’s spatial perception and data analysis capabilities. In this paper, the virtual reality and pharmacophore computing technologywere combined design to interactive visual virtual screening system based on Unity 3D engine and HTC Vive device. The main functions of the system included molecular analysis and rendering, pharmacophore calculation, visual interaction.based on HTC Vice. Experiments showed that the interactive visual virtual screening system proposed in this paper could bring a new perspective for the design of pharmacophor and enhance the interaction experience of drug design researchers. 

Abstract: In this paper, a method of positioning the human joints based on RGB-D was proposed. The human body skeleton line was obtained by the deep image filtering, dilating and thinning processing of the kinect acquisition，the end and turning points of the human skeleton line were obtained according to the corner detection algorithm. Then tracked and positioning the human’s face in the color image; at last the 15 joints of human body were positioned combining with the proportion of the rigid parts of the human body. The results showed that the accuracy of positioning joints of this method was improved by 7% compared with the method of positioning the joints directly according to the proportion of the rigid parts of the human body, and the accuracy of positioning the joint of head is improved by 42%

Abstract: The research of public opinion on education had important values in the field of crisis management, public opinion guidance and so on. At present, Internet has become the main channel for people to express their opinions. However, distilling the emotion trend of public opinion and propagation mechanisms of the emotion, and analyzing the characteristics of emotion on multiple media platforms were still challenges. Therefore, a visualization system of emotion on education public opinion was designed. Firetly, the requirements of emotion analysis of public opinion were defined after the discussions with the users. Secondly, the emotion recognition algorithm was used to recognize emotion of text, and the emotion propagation algorithm was proposed based on the emotion propagation model. Thirdly, multiple interactive views were designed, which allows users to do comparative analysis on different online media platforms, and analyzed propagation mechanisms of emotion. Lastly, case studies on network data of education in 2015 showed that the system could meet the requirements of users and effectively support users’ selection of media platforms in public opinion guidance and crisis managment.

Abstract: In this paper, the system function analysis was carried out based on the capability demand of the joint operation for Cyberspace War Situation Visualization System(CWSVS). Then, a distributed system architecture based on HLA was constructed based on the capability perspective, service perspective and operational perspective, and the scenario generating architecture and real-time running architecture were designed. The component-based rapid and customizable situation driven technology and the map-based multi-layered dynamic fusion visualization technology were taken as the key technology examples. At last, the cyberspace Offensive and defensive operations against the C4ISR system was given as an typical example, and the utility of the system was demonstrated

Abstract: Image debluring technology played an important part in the image processing field. Total variables regularization with overlapping sparisity was gradually applied to the image deblurring problem. It could preserve image edge characteristics and suppress the generation of the staicase effect. When using the alternate direcction multiplier(ADMM) method to solve the overlapping group sparsity total variables model, the penalty factor could greatly influence the deblurring process and it was not easy to adjust. Therefore, a method was proposed to adaptively adjust the penalty factor according to the recovered image when the model was being optimized. This method adaptively restored the best picture and ensured the robustness of the algorithm while guaranteening the speed of calculation. Experimental results showed that the proposed method outperformed other recovery models in terms of PSNR, SNR, relative error and other evaluation indices.   

Abstract: Due to the low monitoring intensities on key tracking persons（suspects）, the police suffered from the very small amounts of suspect social location data, which was hard to effectively reveal the social mobility patterns of suspects, and restrict the police action validity for suspects filtering and crime blockading etc. Facing this data sparsity challenge, a novel Tensor Collective Decomposition Location Prediction (TCDLP) MODEL WAS PROPOSED, to estimate the latent visiting intensity at an arbitrary spatiotemporal node. Specifically, it modeled the visiting intensities of suspects with 3D tensor, where the three dimensions stood for suspects, locations, and time slots respectively. Then, the missing entries in the tensor would be filled through a multi-data fusion tensor decomposition approach, which integrates the correlations of locations and suspects relying on multiple social environment data. So by supplementing the visiting intensities in this tensor, the social spatiotemporal distribution pattern for each suspect could uncovered.TCDLP was evalvated by using a real-world suspect dataset collected form 241 suspects over 6 months with about 19 thousands location records, showing our model outperformed ststed-of -the -art approaches to the problem..

Abstract: The language model based on neural network LSTM structure, the LSTM structure used in the hidden layer unit, the structure unit comprises a memory unit which can store the information for a long time, which has a good memory function for the historical information. But the LSTM in the current input information state9 does not affect the final output information of the output gate, get less historical information. To solve the above problems, this paper puts forward based on improved LSTM  (long short-term memory) modeling method of network model. The model increases the connection from the current input gate to the output gate, and simultaneously combines the oblivious gate and the input gate into a single update. The door keeper input and forgotten past and present memory consolidation, can choose to forget before the accumulation of information, the improved LSTM model can learn the long history of information, solve the drawback of the LSTM method is morerobust. This paper uses the neural network languag LSTM model based on the inproved model on TIMIT data sets show that the axxuracy of test. The results illustrate that the improved LSTM identification error rate is 5
% lower than the standard LSTM identification error rate. 

Abstract: The fault feature should be extracted before the SVM was applied to the bearing fault diagnosis. In the previous feature signal extraction, it was often based on the existing knowledge model. With the application and promotion of DNN in recent years, AE had a special advantage in feature extraction. As an unsupervised learning method, AE could extract the features of the signal based on data driven, making the feature extraction no longer depends on prior knowledge, and the whole fault diagnosis processed more intelligent. In this paper, the improved AE、DAE，were used to extract the features of the bearing signals, and the fault diagnosis was carried out by SVM. Finally, by compared with the SVM based on EMD energy entropy feature extraction, the superiority of DAE-SVM with unsupervised learning method was reflected in beraing fault diagnosis, and its diagnostic accuracy was neraly 100% 

Abstract: Commercial automayic guided vehicles (AGV) usually used chain transmission mechanism power transmission, and the fixed structure of the wheel could be considered as cantilever structure. Therefore, the problem of wheels "tilting" and start-stop "shocking" easily occurs, which limited the accurate movement of the AGV during frequent and rapid acceleration or deceleration. In this paper, AGV designed by a company was taken as an example. Though repeated tests and numerical simulations, the structure and force analysis were used to find out the reasons for this phenomeno. The larger stress was caused by the "L"-shaped suspension mechanism, which magnified the contact gaps of each component; the uses of the chain transmission mechanism could make it easy for the AGV to form gaps between the sprocket and the chain when the AGV started, stopped, moved forward, backward frequently. Then a new drive unit structure was put forward from the engineering point of view, which could solves the above problems, at the same time-greatly could reduced the stress in the mechanism, could improve the transmission precision, and could provide a more practical and optimized driving structure for the design of AGV.

Abstract: By means of EDEM-FLUENT simulation and VOF(Volume of Fluid) method and Euler-Lagrangian model, a mixture model of discrete solid, continuous liquid and gas phase was constructed to simulate the three-phase flow with solid-liquid-gas in a stirring tank. The effect of the moving state of solid particles in stirring tank and free liquid level were explored. The gas-liquid continuous phase modeling based on VOF method using FLUENT software could capture gas-liquid interface well and the model was closer to the actual working condition. Based on the Discrete Element Method(DEM), the discrete element modeling of solid particles was established and its position information in the tank was simulated intuitively by the joint simulation of the two software. The dispersion of solid particles was consistent with the results obtained by Euler method.

Abstract: In order to study the influence of coarse sand lime with different contents on the compressive strength of the mud cube and the correlation between them, the conpressive strength test of the mud cube werecarried out, and the optimal mixing ratio of different admixtures was found to provide theoretical data for the improvement and reinforcement of the adobe wall. The experimental results showed that the compressive strength of the specimen increased first and then decreases with the increase sand content. The compressive strength of the test block was slightly improved by adding lime. with the increase of lime contment, the conpressive strength of the test block had no obvious change. There was a good polynomial relationship be tween the compressive strength of the slurry and the admixture. 

Abstract: According to the early partial deformation failure, which was caused by great difference of the traveling state in different flighting areas, based on ABAQUS softwareplatrorm, this paper analyzed the change law of shear stress、shear strain and permanent  deformation when aircraft load, temperature and driving speed were changed. It showed that the force and deformation of asphalt overlay would be in the limit under the coupling effect of three kinds of heavy load, high temperature and low speed. By reasonable design of material type combination and thickness, the early rutting of parallel taxiway and liaison road could be significantly reduced, and the service level of pavement could be improved. Thus, based on the existing design, this paper presented a design method for different regions of asphalt overlay in airport pavement, that the allowable shear strss and allowable rut for checking index.