Abstract: The recognition of acoustic signal in oil pipe was researched,and there was uncertain attenuation property in acoustic communication channel for oil field pipe,and the property was different with the difference of length of pipeline and environment. Traditionally,the recognition model is established based on the stable signal feature. And it is difficult to establish the model with the uncertain attenuation property which resulted in difficult to recognition. In the paper,an acoustic communication channel model establishment method was proposed based on the ray theory. According to the acoustic ray theory and the real collected data from the oil fields,the acoustic communication property of the pipes was researched,the underwater communication channel characteristics of the injection well were analyzed,and the physical environment of the acoustic communication channel was analyzed. The propagation regular of acoustic in this condition was computed for the precise recognition. Simulation result shows that the underwater acoustic signal can reduce to zero on the mouth of the well,the underwater acoustic signal can be received on the well mouth and there is coupling signal 50m to the well. The research result shows good performance in signal recognition.

Abstract: With the rapid development of new Information Technology(IT), big data platform technology has been one of the important supporting technologies of modern modeling simulation field. On the basis of the current situation and research of big data applications, the architecture of big data supporting platform was proposed, and its application mode, big data application standard, big data storage, big data analysis, and some related technologies were researched and summarized. The applications and development of big data platform were discussed.

Abstract: In order to achieve reliable attitude control of the four- rotor aircraft,we used the 6- DOF motion equation module to construct the non- linear model of the four- rotor aircraft under the environment of Matlab in this paper. We chose the attitude of the four- rotor as control parameter,designed the fuzzy PID controller with the help of Matlab fuzzy toolbox and also edited the fuzzy rules rely on expert's experience. Meanwhile,we designed the convention PID controller with the best parameters,and employed Matlab simulation to realize the attitude control simulation with this two kinds of controllers of the four- rotor aircraft. The result of the simulation shows that,the fuzzy PID controller has more excellent dynamic property and robustness compared with the convention PID controller.

Abstract: For the cooperative attack of multiple missiles against the targets with close-in weapon systems, cooperative guidance law was designed with impact time and angle constraints considered. Mathematical models for the guidance problem were constructed, and the design problem of the cooperative guidance law was formulated as a nonlinear optimal control problem with fixed terminal horizon and terminal state constraints. The Gauss pseudospectral method was employed to transform the optimal control problem to a nonlinear programming(NLP) problem, and solved by sequential quadratic programming(SQP) method. The predictive control method was used to implement the designed optimal guidance law. The simulation results verify the feasibility of this method.

Abstract: Based on the theory of function approximation and polynomial interpolation, polynomial neuronets are constructed by using linearly independent or orthogonal polynomials as activation functions of hidden-layer neurons. After using Legendre polynomials, Hermite polynomials, Chebyshev polynomials of class I, Chebyshev polynomials of class II, Bernoulli polynomials and power functions as activation functions to construct single-input neuronets, a weights-and-structure-determination algorithm of growing type is proposed, which can be applied to these six aforementioned neuronets to determine the optimal structure and weights. With this algorithm, the abilities of learning and prediction of these six neuronets activated by different polynomials are further investigated. Simulation results show that, though the Hermite polynomial neuronet and the Bernoulli polynomial neuronet perform relatively ordinarily, other four neuronets possess superior abilities of learning and prediction. Finally, the neuronet using Chebyshev polynomials of class I is used to simulate the trend of the world population.

Abstract: Threats modeling is an important prerequisite for UAV trajectory planning and task deducting. It is difficult to quantify the UAV battlefield threats,and how describe multi-threats interaction is another problem. Firstly, in order to quantize the UAV battlefield threats,models based on artificial potential field were brought forward,which constructed potential functions for different threats. And then,defining the superposition principle of potential field solved the problem of multi-threats interaction. Finally,considering UAV characters of stealth,jam and flexibility as attenuation indexes of threats,the threats situation of UAV in the battlefield was better performed. Simulation results for two different UAVs show that the computation is simple and it is able to describe the characters of UAV battlefield threats.

Abstract: Due to the geological conditions,the drilling construction technics and other reasons,the heterogeneity of load on the casing will get increased,and be under complex stress state,which leads it prone to failure and damage. As an example with N80 casing wear,based on the theory of rock mechanics and elastic- plastic mechanics,we used a software ABAQUS to construct the model of formation concluding cement and cased for finite element analysis in view of the problems on the performance of cement sheath,degree of centered and deficiency of material. The results of research demonstrate that the cementing quality with higher strength,larger stiffness,no deficiency and better centered degree is the best to protect casing from failure. The method and achievements in the paper provide the theoretical supports for casing damage by poor cementing quality.

Abstract: Robot 6 degree of freedom vision positioning is a popular and difficult topic in the field of robot visual servoing for its property of strong coupling and nonlinearity A robot 6-degree of freedom uncalibrated vision positioning method was proposed using an adaptive kalman filter Firstly, a set of image moments was designed to represent the relative translation and rotation motion between camera and object Then, an adaptive kalman filter was designed to estimate the image Jacobian matrix on-line in the uncalibrated hand-eye coordination systems and visual control law was designed to calculate the motion control quantity of the robot Finally, the Simulink model for a robot 6DOF uncalibrated vision positioning system with eye-to-hand configuration was built using Matlab After this, the 6-degree of freedom vision positioning of robot was achieved The experiment results of 6- degree of freedom vision positioning show that the adaptive kalman filter can guide the robot to the desired position with high accuracy under the partial known noises

Abstract: In the cognitive radio system,spectrum allocation technology is the key to decide whether the limited frequency spectrum can be used fully and efficiently,which depends on the optimal solution of maximizing system benefit. The hybrid adaptability was introduced firstly to solve the problems that the traditional adaptive genetic algorithm fall into local optimal solution and has a large amount of calculation,in which fixed crossover and mutation probability or adaptive crossover and mutation probability was used according to the population evolution algebra. Then the idea of golden ratio used to calculate adaptive crossover and mutation probability was presented to solve the problem of long operation time. In the end,the performance of system benefit and time cost were simulated. Simulation results show that the improved algorithm can find the optimal adaptive point quickly and accurately and improve the search efficiency,and has better system benefit.

Abstract: For the virtual machine resource scheduling problem in the cloud computing, scheduling and management of virtual machine resources effectively and to achieve a good balance in terms of resource utilization, the cost of overhead, and time constraints, multi-objective comprehensive evaluation of the virtual machine resource scheduling model was proposed, and the typical multi-objective optimization algorithm was applied to the problem, the virtual machine resource scheduling and task allocation merged into a process in order to reduce the complexity of the problem. Analysis through simulation comparison experiments, the method has good practicality and expansion, and the ability to take into account the resources allocated user preferences.

Abstract: In order to solve the problems of bat algorithm,such as low convergence accuracy,slow convergence velocity and easily falling into local optimization,this paper presented an improved bat algorithm based on differential evolution algorithm.The mutation,crossover and selection mechanism of differential evolution algorithm were introduced into the bat algorithm,so that the bat algorithm lack of mutation mechanism has the variation mechanism,which can enhance the diversity of bat algorithm,the avoid the population falling into local optimum,and enhance the ability of global optimization for bat algorithm.The simulation results of six standard benchmark functions show that the improved algorithm can greatly improve the convergence precision,convergence speed and robustness,and effectively discourage the premature convergence.Meanwhile,the improved algorithm was applied to solve nonlinear equations and the numerical examples were proposed,which proves the feasibility and effectiveness of the improved algorithm.

Abstract: A three-dimensional marine engine room visual simulation system with strong sense of reality and interactivity was designed Equipment common material was developed by use of shader Illumination models and special effects were realized in shader, thus improved real-time rendering performance Human-machine interactive functions including roaming, navigation and picking up were designed Camera control method under different roaming modes and perspective was given Automatic roaming was realized by path planning algorithm and real-time follow up function under third person perspective was achieved by virtual force method Three navigation modes including global map, local map and portal were designed To improve three-dimensional pick up efficiency in virtual engine room, the pick up result's proportion was introduced to aid judgment Based on binocular stereo vision, stereoscopic imaging was realized Comparative analysis indicates the characteristics of binocular convergence projection and binocular parallel projection in marine engine room visual simulation The designed marine engine room three-dimensional visual simulation system has been successfully applied in controllable pitch propeller marine engine room simulator

Abstract: The operational capability of flight deck is the critical factor to affect the sortie generation of carrier-based aircraft, including launch operation, recovery and respot operations, serving and so on The definition of optimized flight deck operation plan was given A method to calculate the number of sorties generation in optimized flight deck operation plan was proposed Some factors including aircraft number, launch time, respotted time, recovery time how to affect the sortie generation were analyzed Finally, a type example was given The results show that promoting the capability of flight deck is the key factor to increase the number of sorties The results can provide reference and quantization basis for improving the capacity of sortie generation and operation efficiency for carrier-borne aircraft

Abstract: In order to improve the efficiency of assembly simulation, the method of Assembly Sequence Planning(ASP) based on ant colony genetic hybrid algorithm and the best hybrid strategy of controlling two algorithms dispatch were proposed. In the process of solving the assembly sequences, the concept of assembly costs was proposed, which include the change of assembly tools, the alteration of assembly direction, and custom assembly criteria. The interactive verification method of assembly sequence was adopted, which the optimal assembly sequence obtained from the intelligent algorithm was simulated to assembly in virtual assembly system in order to influence the subsequent calculation. The simulation experiment indicates that the efficiency of hybrid algorithm proposed is better than ant colony algorithm and genetic one, respectively, and the interactive method of assembly verification could examine the rationality of assembly sequence to make the assembly sequence obtained conform to the requirements of the real assembly.

Abstract: In order to improve vehicle performance and safety,the accurate estimation of power lithium battery state of charge( SOC) is needed,and the key technical challenge is to establish a reasonable and effective battery model because of nonlinear characteristic of power lithium-ion battery In this paper,we analyzed the works of power lithium-ion battery,presented a battery model based on RC equivalent circuit,deduced and established the equations of RC equivalent circuit,and identified parameters of the mathematical model by experimental data and laplace At last,we simulated the estimation for the SOC of power lithium-ion battery based on EKF The simulation results show better battery model accuracy and less computation

Abstract: This theis aims to avoid big errors caused by the NLOS and multipath transmission in indoor environment when determining the position of an object or person. In order to improve positional accuracy,the author proposes that firstly use Wavelet analysis to denoise,which is conducted by measuring the distance between target nodes( object or person with positioning terminal) and three anchor nodes( lower computer) obtained by TOA model of UWB Indoor Positioning System. Then determine the final location of target nodes by adopting Full Centroid Position scheme,namely finding the coordinate values of the target nodes in two-dimensional coordinate system. Simulation results indicate that the above method has the advantage of possessing higher positional accuracy and less data testing compared with traditional particle filter,Bayesian filter method and Taylor Series Approach,thus having better real-time. This method can avoid the defects of weak non-convergence robustness caused by improper initial positional results when using Taylor Series Approach. It is proved that the advanced TOA model has more feasibility.

Abstract: Accuracy of power load forecasting was researched.Power load is related with weather,economy,and holiday factors.The variation rule is cyclical and random,and traditional method cannot describe the variation rule,which leads to low accuracy of prediction.In order to improve the accuracy of power load forecasting,an extreme learning machine(ELM) was proposed for predicting the power load,by integrating similar days selecting method based on the time sequence encoding.The method that selects similar days based on the time sequence encoding integrates the information of the entire sequence into each encoded point,which can not only describe the trend of a sequence but also can describe the relative position of the point in the sequence.Then,the ELM was used for prediction,because it only need to set the number of hidden layer nodes and can generate a unique optimal solution,while the input weights of the network and the bias of the hidden layer nodes do not need to adjust during the execution of the algorithm.The power load data of one building were used for simulating.The proposed method was compared with support vector machines(SVM) and back propagation(BP) neural network,and the experimental results show that the proposed method has high prediction precision,high adaptability and shorter running time.

Abstract: A method for automatic extraction of landmarks for vision- based navigation is proposed in this paper to solve the short- comings of low matching probability and slow speed in scene matching area's selection methods.According to different size of sample images,this algorithm uses image gray features,edge features and correlation plane statistic features to training parameters via Support Vector Machine( SVM). Then we use SVM to classify testing images,and finally we take unsupervised clustering algorithm to optimize landmarks selected by SVM. Simulation experiments and real flight data show that our method can automatic extract landmarks,and the extracted landmarks have characters of stability,high matching accuracy and short matching time.

Abstract: The safety features of nuclear power plants require digital control system software must have high reliability.Because of the special nature of computer software,to discover some design flaws and logic errors must rely on special conditions or man-made destructive testing,the cost is very high.This article provided a validation and verification method based on nuclear power plant full scope simulator,design various verification tests by using high-precision simulation model and simulating unexpected conditions in nuclear power plant,to validate the safety and reliability of IC software in such a contingency conditions.Under the assumptions "digital control system is unsafe",we performed some unconventional operations to verify the logic of the instrument control software and procedures unsafe factors in the digital control system.This article also provided a test of unplanned reactor accident caused by trip to identify possible defects in the design and improvements.The results show that the full-scope simulator-based IC software design verification is feasible.

Abstract: 시뮬레이션 모델은 사이버 물리 시스템(Cyber-Physical System, CPS)의 개발을 위한 중요한 토대로서 CPS 분야에서 광범위하게 응용되고 있다. CPS와 전통적인 임베디드 시스템(embedded system)은 아주 큰 차이가 있다. 즉 시스템의 규모가 더 크고 시스템의 이질성(heterogeneity)이 더 뚜렷하다. 또한 네트워크의 개방성...

Abstract: Data Distribution Service(DDS) is a specification about data distribution service for real-time systems published by OMG, and defines a mechanism for Data-Centric Publish-Subscribe(DCPS) It can support the requirement for real-time and effective data-interactive Depending on the study of DDS, a group of APIs and a distributed simulation system were designed, and the system is more flexible, real-time, scalable and reliable It is shown that the DDS is available for real-time system communications according to its some practical applications

Abstract: A fast 3D cloud simulation algorithm based on particle system was proposed. It used the given several spheres to simulate the outline of clouds and determined the bounding box of cloud particles. Then the particles were judged whether they were part of cloud particles by comparing the position relationship between particles and spheres. Finally, rendering and drawing cloud particles to realize 3D simulation of clouds. As particles kept moving, so it was need to update attributes of particles at internals. If the particles moved out of the spheres, they should be deleted. By simulating the movement of particles, the method implemented the simulation of the motion of clouds. At last, it proves that the method can produce shapes of 3D clouds quickly and really by several experiments, at the same time the algorithm can implement the dynamic simulation of clouds.

Abstract: In order to satisfy the requirement of visual analysis for multi-type of hierarchical datasets and their relationships in some fields, such as food safety, a hybrid layout algorithm was proposed for Double Interrelated Tree(DIT) based on node-link and sunburst A node-link tree based on force-directed algorithm and a space-filling tree based on sunburst were used to visualize two types of hierarchical dataset respectively, and tree nodes having association relationships together were connected by a line to form a double interrelated tree In order to reduce visual clutter and edge crossings, a layout optimization algorithm DRA(Detour Route Algorithm) based on path around-on was proposed The optimized double interrelated tree was applied into the pesticide residue detection dataset The layout result can clearly show regional hierarchical information, pesticides' classification information, and relationships between them

Abstract: In this paper,a method of engine accurate fault diagnosis was studied. When the engine is failure,the use of traditional algorithm for fault diagnosis needs to compare the engine fault parameters with the running status parameters of every component one by one. Thus,it has strong lag in component fault diagnosis. In order to avoid the disadvantages of the traditional algorithm,this paper proposed an engine fault diagnosis method based on PSO-SVM. Using the method of particle swarm,all of the engine fault signals were searched within a specified space to obtain the optimal particle and provide the basis for engine fault diagnosis. Then using support vector machine( SVM) method,the classification of engine fault signal was achieved to complete the engine fault diagnosis. The experimental results show that the algorithm presented in this paper for engine fault diagnosis can greatly improve the accuracy,so as to meet the practical requirements of producing and living and get the favorable results.

Abstract: A delay model of at traffic junctions and an algorithm of the Traffic Light Based Shortest Paths(TLBSP)were proposed,and Dijkstra's algorithm was improved also.The TLBSP algorithm can fit most traffic networks and traffic rules at present and can be calculated for each vehicle.The experimental results show that,the paths based on the TLBSP algorithm are more optimal than the path based on Distance Shortest Paths and TDSP with fixed delay.In addition,the algorithm can be applied to urban traffic navigation.It is beneficial to reasonable distribution of urban vehicle and alleviate urban traffic congestion.

Abstract: Flow visualization is a very active subfield of scientific visualization, and the study of 2D flow visualization can afford foundation for 3D flow or higher dimensionality flow. There have been several classifications by different authors. The classifications were integrated into four types: direct flow visualization, texture-based flow visualization, geometric flow visualization and feature-based flow visualization, after consulting papers published in recent years, a new type of 2D flow visualization technique was added based on information theory, then several typical algorithms were introduced and compared. On this basis, charts and forms were used to state the results of comparison.

Abstract: The rapid and efficient tourist evacuation when stampede accidents occurre in the holiday travel area was studied. A data mining method based on agent decision algorithm for tourist rapid evacuation in the holidays crowded travel area was proposed. Through data mining,the flow direction and mutation parameters of density were analyzed to obtain the constraints of rapid tourist evacuation and establish the agent decision-making model. Taking the advantage of rapid decision-making of data mining method,the effects of the mutation and non-linear can be overcome,to achieve the rapid tourist evacuation in holiday crowded travel area. The simulation results show that the data mining algorithms for tourist rapid evacuation can improve the speed of evacuation.

Abstract: 클라우드 추론(cloud reasoning) 기술을 정보 융합(information fusion) 분야에 도입하여, MIN-MAX 클라우드 무게중심 추론(cloudy MIN-MAX center of gravity reasoning)에 기초한 새로운 위협성 평가(threat assessment)방법을 제안하였다. 실제 응용 배경에서 표적의 특성을 근거로 하여 계층적 위협성 평가를 위한 속...

Abstract: As a result of the non-linear characteristics and the uncertain disturbances in high-power AC servo system, it is difficult to construct an accurate mathematical model. In order to solve this problem, a system identification method based on particle swarm optimization wavelet neural network was proposed. Due to the advantages of the particle swarm optimization including avoidance of the local minimum of the unstable system and fast convergence rate, the particle swarm optimization was used to optimize the parameter of wavelet neural network. In the particle's positions vector, the connection weights and the threshold value as a particle swarm optimization algorithm. Moreover, according to the optimal value in the particle swarm algorithm, the network weights and threshold value was replaced the traditional gradient descent method. Compared with traditional wavelet neural network, the identification result shows that the particle group of wave neural network algorithm has reduced function approximation error and improved ability of the network performance significantly. And to a certain extent, it also solves the local minimum value problem.