Abstract:With the development of deep-space exploration, low-density ablators have received increased attention due to their distinguished features such as good ablation performance and low structure weight to fulfill the requirement of lightweight, low ablation rate, low thermal conductivity and long re-entry time of thermal protection for heatshield and backshell of planetary entry probes and re-entry capsules. The historical prospects on development of low-density ablative thermal protection systems and materials were introduced with emphasis on low-density honeycomb reinforced silicon rubber, phenolic and organic silicone resin ablators, as well as the typical representative of light weight ceramic ablators (LCAs) such as silicone impregnated reusable ceramic ablator (SIRCA) and phenolic impregnated carbon ablator (PICA). According to the task requirements of China's new-generation manned spacecraft, including low earth orbit flight mission, as well as manned flights to the Moon, the asteroid and the Mars, the progress of basic and applied research on lightweight thermal protection system for the spacecraft was presented. And then, it analyzes the technology gap between domestic and foreign, and provides several suggestions concerning the prospects of China's low-density ablators.

Abstract:To reduce the complexity of Massive multiple-input multiple-output (MIMO) signal detection, the iterative method is used for signal detection. Based on the implementation of the iterative method with matrix decomposition, the successive over relaxation algorithm is introduced and the range of relaxation factor is deduced by determinant calculation. Simultaneously, a three-dimensional spatially correlated channel model is built based on two-dimension with geometric method and the analytic form solution of spatially correlated channel is introduced by neglecting the high order. The simulation results show the three-dimensional spatially correlated channel model aggravates channel correlation and decreases detection performance. Under 8 times iteration and certain bit error rate, the signal-to-noise ratio of successive over relaxation iteration decreases with the optimal relaxation factor. In a definite signal-to-noise ratio, bit error rate decreases approximate two orders of magnitude, diversity gain promotes and detection algorithm performs equally 16 times iteration, which can also decrease computational complexity. The successive over relaxation detection algorithm can perform better in less iteration and achieve better detection performance with optimal relaxation factor by considering the signal-to-noise ratio and the complexity.

Abstract:To reduce the weight of antenna structures and improve the structural efficiency of the integrative antenna structures, a flexible and conformal membrane antenna is proposed. The microstrip quasi-yagi antenna is designed and prepared based on the antenna design theory and screen printing technology. The radiation layer is considered as a particle reinforced composite. The effective elastic modulus of the radiation layer is predicted using Mori-Tanaka method, which is in good agreement with that obtained by tensile method. To reveal the interface failure process during 90° peel test, the interface property between the substrate and radiation layer of the antenna is studied based on cohesive finite element model using ABAQUS software. The peel strength predicted by FEA is 64N/m and verified by experiment, which has the error of 8.11%. Finally, the reflection coefficient and radiation performance of the membrane antenna are discussed. The simulated resonance frequency is 3.02 GHz, which has the error of 2.37%. The results show that the flexible membrane antenna prepared based on screen printing technology has good mechanical and electric properties, which has extensive potential applications in aerospace lightweight structures.

Abstract:To solve the problem that the performance of failure detection is effected by mobility of vehicular cloud computing, a new failure detector (VC-FD) based on architecture of vehicular cloud computing was proposed. We found that vehicular cloud computing has a hierarchical structure and the vehicles can be grouped through Roadside Unit (RSU), according to the architecture analysis of vehicular cloud computing. On this basis, the sharing mechanism of detection results is introduced, and the new failure detector VC-FD for vehicular cloud computing was proposed. Then, the model of cooperative failure detection was proposed to evaluate the detection time, mistake rate and detection overhead. The performance of the VC-FD and non shared failure detector was compared in a simulation environment. The experimental results show that the VC-FD can ensure the accuracy, and significantly improve the node failure detection time by increasing limited detection overhead. Thus, the VC-FD can accurately and quickly find out the node failures in vehicular cloud computing, and effectively reduce the influence of vehicular mobility on the performance of failure detection.

Abstract:To track maneuvering target, cognitive radar could adjust its transmitted waveforms to obtain some sustainable or robust information of target. We propose a novel adaptive waveform design method based on matrix-weighted Interacting Multiple Model fusion (AMIMM) for tracking maneuvering target. The target state is modeled via the multi-model idea, and the tracking framework is formulated by using the matrix-weighted multi-model fusion in lieu of probability-weighted way. The fused covariance matrix of state estimation is selected as the ellipse metric, and its ellipse parameters could be obtained by using EigenValue Decomposition (EVD). Finally, according to the fused covariance matrix-related ellipse parameters, the fractional Fourier transform (FrFT) is utilized to rotate the measurement error ellipse to make them orthogonal to each other, and obtain the adaptive waveforms. Simulation results show that compared with several current algorithms, our algorithm could further improve the tracking accuracy and robustness.

Abstract:In order to investigate the influence of different porosity carbon fiber/epoxy resin laminates on the bending mechanical properties at the high and low temperature and hygrothermal aging environment and subjected external load, the supersonic state and the internal stress state were simulated. High and low temperature accelerated hygrothermal cyclical tests for T700CF / 3234EP were employed to simulate supersonic aircraft service environment. Three laminates with porosity of 0.04, 0.08 and 0.11 were prepared by controlling the molding pressure. The loading of the laminates was loaded with 0, 30%, 40% and 60% of the maximum bending load of the laminates, respectively, and then placed in the high and low temperature environment test chamber. The bending strength of composite laminates under different temperature cycles is measured respectively, which were simulated based on ABAQUS software. The experimental results show that the external load and porosity increase the bending strength reduction of the laminates under the alternating temperature and humidity field.During the high and low temperature and hygrothermal cycle 4~6 periods, the secondary curing phenomenon becomes more obvious as the load increases andthe changes of the bending strength of the laminate is relatively gentle. The calculated trend of bending strength based on finite element model was consistent with the experimental results. The effects of high and low temperature, hygrothermal aging environment and external loading on the bending strength of composites were explained. The bending property of carbon fiber / epoxy resin laminates with different porosities were predicted accurately.

Abstract:In order to reduce the influence of the selection of parameters in community detection algorithm on the results of community partitioning and detect adaptively, a community detection algorithm called KDED based on kernel density estimation for density peak clustering is proposed. Firstly, a distance measure based on trust is defined, and the user relationship in the social network is quantified as a distance matrix. Then the kernel density is estimated by the distance matrix to calculate the influence of each node on the network. The thermal diffusion model improves the computational flow so that it adapts to different sizes of data sets to improve computational accuracy. The clustering center is determined by density peak clustering principle and community property, and the node is allocated to the corresponding community which can obtain the hierarchical structure within the community and the natural structure among the community. The simulation results show that it can be observed by the visualization software that the community division result obtained by the KDED algorithm has a clear natural structure and internal hierarchical structure. The KDED algorithm has the best stability with the increase in the size of the community and the difficulty of detection compared with the classical algorithm. And it gets a community partition closer to the real partition in the real data set and the LFR benchmark network, which verifies the feasibility and effectiveness of the algorithm.

Abstract:Local information based level set methods are comparatively effective in segmention of images with intensity inhomogeneity which often occurs in real images. However, there are problems such as local minima in the active contour energy and the considerable computing-consuming in these models in practice. A novel level set method based on improved region-scalable fitting is proposed for such image segmentation. The global information containing local feature term, measured by Bregman divergence, is utilized to accelerate the contour evolution when the contour is far away from object boundaries, in order to improve the robustness of the algorithm to the initial position. The local feature item of the RSF model is used to promote the convergence of the contour curve to the boundary of the object, thereby increase the capability to cope with intensity inhomogeneity. Finally, comparative experiments on synthetic images, mediation images and the real images validate the performance of the algorithm.

Abstract:To solve the less comprehensive and objective problem of the traditional microblog user influence evaluation algorithms, through the analysis of the definition and influencing factors of microblog user influence, this paper proposes an improved user influence ranking algorithm based on PageRank algorithm, named as Self and Followers User Influence Rank (SF-UIR). The user's own factors are quantified by using the three indicators, the number of followers, the situation of certification, and the microblog dissemination ability, and the poor objectivity situation of PageRank values for user influence ranking is improved. The disadvantage of influence equivalent transfer of the followers' influence is overcame by adopting weighting factor to distribute the influence contribution value of different followers scientifically and quantitatively. Compared with the four mainstream algorithms, the results show that the proposed algorithm is more comprehensive, more objective, and can reflect the influence of microblog users better because of considering the influencing factors based on the user's behavior and followers factors based on the topology, which can effectively solve the interference problem of "zombie fan" in a number of followers ranking algorithm. It can reflect the user's influence level more realistically than average forwarding number algorithm, and can availably avoid the serious defects of not taking the microblog user's behavior into account and giving equal treatment to all followers in K-coverage algorithm. The proposed algorithm can greatly improve the shortage of relying solely on the quantity and quality of followers in PageRank algorithm.

Abstract:Using 3DS MAX to obtain elemental image (EI) array in the virtual integral imaging system need to put large-scale camera array which is difficult to be applied to practice. To solve this problem we establish a sparse acquisition integral imaging system. In order to improve the accuracy of disparity calculation, a method is proposed using color segmentation and integral projection to calculate the average disparity value of each color object between two adjacent render output pictures. First, a virtual scene and micro lens array model in 3DS MAX is established. According to the mapping relationship between EI and sub image (SI), we can obtain the SI first, and then calculate the EI. Based on color image segmentation method and integral projection method, we acquire the average value of the disparity from different color objects between adjacent images; and then translate a rectangular window of fixed size in accordance with the average disparity to intercept rendering images to get the sub images. Finally, after stitching and mapping of the sub images we obtain the elemental images, which can be used to display a 3-dimentioanal (3D) scene. The experimental results show that we can use only 12×12 cameras instead of 59×41 cameras to obtain elemental images, and the stereoscopic display effect is obvious. The error rate of disparity computation is 0.433% in both horizontal and vertical directions, which is obviously better than other methods of disparity error rate of 2.597% and 4.762%. The sparse acquisition integral imaging system is more accurate and more convenient which can be used for EI content acquisition for large screen stereoscopic display.

Abstract:To obtain the effect of different parameters on the tensile performance of stepped-lap repaired composite laminates, the repaired specimens were investigated experimentally. In terms of the laminates with 8 plies, stepped-lap repaired specimens with 2, 4 and 8 steps were tested, respectively. In addition, the stepped-lap joint with 4 steps was studied with different number of external plies. As a comparison, a tapered scarf repair with the same bonding area was also tested. Results showed that the tensile strength of stepped-lap repairs increased with increasing the number of steps. When the number of steps increased from 2 to 8, the strength recovery rate of the repaired joint increased from 36% to 67%. The introduction of external plies could also improve the efficiency of repairs. However, the dispersion of the repair strength increased with the increase of external plies. As the stress distribution in the adhesive film is more uniform for the tapered scarf joint, its repair strength is higher. Compared the stepped-lap repaired joint with the tapered scarf joint with the same bonding area, the repair efficiency of the latter is higher. Take the 4 stepped-lap repaired joint for example, its repair strength is increased by 25% compared with that of the tapered scarf joint. Finally, according to the stress/strain distribution on the surface of the repaired joints, the failure mechanism and failure process of the repaired joints under tensile load were obtained.

Abstract:Aiming at the shortcomings of traditional graphite negative electrode, such as low specific capacity, large volume expansion of charging, resulting in instability of electrode structure and poor rate performance, it is proposed to use nickel graphene nanosheet (Ni-Gns) composite coating as a cathode electrode material for lithium-ion battery. To verify the performance of this material, a new ultrasonic assisted jet electrodeposition experiment device is developed, and the Ni-Gns composite coating is prepared under different process conditions. Then, the battery negative electrode is fabricated based on such composite materials.. An orthogonal experimental method is employed to measure the discharge capacity of the electrode over 50 charge/discharge cycles, and the electrodeposition process parameters are optimized. The results show that the dominant electrodeposition parameters sequences: added amount of Gns m > coating thickness h > current density D>bath temperature T; optimized process parameters: m = 0.75 g·L^-1, h = 30 μm, D = 33 A·dm^-2, T= 35 ℃. The surface morphology and microstructure of the optimized coating are observed, which show the Ni -Gns composite coating has a compact structure and a good interface bonding. For transition metal-carbon composite materials, the mechanism model for the electrodeposition process of carbon nanofibers (whiskers) as reinforcement in this composite is proposed. The study suggests that the ultrasonic wave can improve the efficiency of electrodeposition and the structure of the coating. Under the synergistic action of power ultrasonic and electrodeposition, Ni (matrix) and Gns (reinforcing phase) connected by "nickel bridge" between the overlapping form dense and uniform distribution of "percolation conductive network". Combined with Gns's multilayer thin honeycomb lamella structure and its excellent mechanical and electrochemical properties, the material has excellent conductivity and stable structure, and is suitable for lithium ion battery cathode material.

Abstract:To solve the problems that the traditional fruit fly algorithms fall into convergence too early and the results are not stable, an improved fruit fly optimization algorithm based on global-local bidirectional driving is proposed. Firstly, in order to comprehensively consider the global driving information of fruit fly population and the local driving information of a fruit fly individual, the conceptions of advanced group and memory space are introduced. In each iteration, the fruit flies which have good performances are defined as the advanced group, and the historical best positions of a fruit fly are defined as the memory space of this fruit fly. Secondly, in order to avoid the premature convergence problem, the global driving effect of the fruit flies in the advanced group is considered, and the dimensional components of the fruit fly position vectors are updated sequentially in the position updating processes. Finally, in order to avoid the blind global searching when the population approaches convergence, each fruit fly will consider the local driving effect of its own cognitive experience, and the roulette strategy is used to select the positions in the memory space for jumping out the local optimum. The experimental results of typical test functions and the web anomaly detection simulation show that, the proposed fruit fly optimization algorithm based on global-local bidirectional driving has high searching accuracy, good stability, and high convergence speed, and is suitable for dealing with the complex problems with high dimensions in web anomaly detection.

Abstract:Aiming at the inaccuracy of superpixel segmentation and the value calculation of saliency map in image saliency detection, this paper proposes an alternative form of saliency detection and updated algorithm based on the level set superpixels and Bayesian framework. A division or merger of the superpixels will achieve a level set method to form superpixels which are adaptive to the different regions of different size. The model's original map will be constructed by using color and space contrast between boundary and inner superpixels. The saliency region can be indicated with the level set superpixels and then three updated algorithms are effectively put forward via the Bayesian framework and based on K-means clustering algorithm. As a result of this update to the original mapping system, the final saliency result can be discovered in a manner that is a vast improvement of the existing method while still achieving a better level. Using our algorithm can improve the accuracy, recall rate, F value and reduce the mean absolute error value. Finally, this saliency detection algorithm for the face recognition can be effectively utilized to deal with the pictures containing people. The experimental results on three open databases show that our proposed algorithm is superior to FT, CA, XL, MR, wCO, BSCA and other saliency detection algorithms in accuracy, recall rate, F value and mean absolute error.

Abstract:Wave-transmitting Si₃N₄ ceramics is one of the main materials for making the antenna window. The tool-workpiece friction characteristics in milling are closely related to the machined surface morphology. Through the tool-workpiece contact relationship analysis, the friction contact area is mainly focused on the cutting edge flank surface and the workpiece surface. The friction mechanism of the contact area under different processing mechanisms has been analyzed. It comes to conclusions that the friction mechanism is adhesive friction, plough friction of tool surface roughness crest and rolling friction of powder for the plastic domain processing, while it also includes the plough friction of the broken ceramics particles for the brittle domain processing. The friction coefficient calculation formula has been proposed with the existing friction theory and the analysis results have been verified with the experimental data. It shows that the critical cutting depth of the transition mechanism for the wave-transmitting Si₃N₄ ceramics is in the range of 0.3 to 0.4 mm. The maximum error rate of the friction coefficient formula is 20.46%, which can reflect the evolution of the friction characteristic to some extent. In the initial machining mechanism transition stage, the friction coefficient decreases, and the influence of friction coefficient on surface roughness follows the third order polynomial distribution. This study provides the reference for improving the surface quality of wave-transmitting Si₃N₄ ceramics milling.

Abstract:Aiming at the real-time detection and weak echo signal characteristics of push-broom airborne laser detection and ranging (LADAR) system, one algorithm for echo time abstracting is researched based on Gauss normal echo signal. The algorithm is fast, precise and easy to design in hardware, which achieves high precision distance detection as a result. The existing problems and accuracy of the traditional waveform centroid algorithm are analyzed. In order to reduce the influence of the data precision and noise interference on the hardware, a corrected approach based on the median method is proposed, which can effectively weaken the layered phenomenon in hardware implementation of algorithm. A mathematical model is established by MATLAB and the simulation experiment is carried out. The simulation results show that the corrected centroid algorithm has better robustness compared with the traditional centroid algorithm and Gaussian fitting algorithm. The accuracy can reach 0.5 ns, which is 45% higher than that of the traditional centroid when the SNR is 5 dB. The proposed algorithm can achieve higher accuracy than Gaussian's with the SNR becoming greater, which reaches 0.01 ns theoretically. The corrected algorithm is simple and efficient. In addition, real-time judgments and correction on the results of the centroid are made without increasing the complexity of the system. Owing to the simplicity of the algorithm, it is suitable for transplanting to FPGA platform. The algorithm has been utilized for push-broom airborne LADAR system information processing circuit. The results of experiment verify the validity of the theoretical analyses and show that the algorithm can achieve the ranging accuracy of ±7.5 cm, which meets the real-time requirements in practical application.

Abstract:Image quality assessment is a basic problem in the multimedia field. A new image quality metric is constructed based on Structural Similarity (SSIM) index, by exploiting Human Visual System (HVS) characteristics. First, considering the masking effects of HVS, the distorted image is preprocessed by a distortion model with the input of the error between it and the original one and just noticeable distortions (JNDs) derived from a human visual model, where Sigmoid function is explored. The process makes the visible errors in the modified distorted image to be more notable. Second, considering the visual attention characteristics of HVS, the image area weight model is designed to quantify the importance of image local regions for the visual quality. The interesting content of an image can be represented by the saliency image, from which the weights of different regions are obtained. Finally, the local SSIM between the modified distorted image and original image is calculated, and the global image quality metric can be expressed by weighting all local quality with the normalized regional weights. Compared with the state-of-the-art image metrics, the proposed metric fits subjective visual quality better in evaluating the local image quality, has better performance in terms of the mean square error, correlation coefficient and other indicators for predicting the subjective image quality, and has a moderate computational complexity, well below the run time of the superior performance metrics.

Abstract:Thermal conductivity of a material is one of the important parameters of its thermal properties and efficiency in measuring the thermal conductivity of a material is of great significance in an engineering sense. In the engineering field, quite a number of materials are transversely isotropic. A common method to measure the thermal conductivities of transversely isotropic materials is the guarded hot-plate (GHP) method, which applies heat flux to the specimen respectively in different directions inducing steady-state temperature field in the material and obtains the conductivity of each direction via the relationship between the heat flux and the temperature gradient. This method as a steady-state method is theoretically simple but time-consuming, limited to those scenarios where there are many specimens to be tested. Transient line heat source (LHS) method, based on the transient heat conduction theory, is a simple and effective method in measuring thermal conductivities of materials. The function of LHS method was extended based on the heat conduction theory in transversely isotropic media. A new method of measuring thermal conductivities of transversely isotropic materials with LHS method was proposed. The newly proposed method was validated by comparing with guarded hot-plate method based on the results from a test on specimens made of an artificial transversely isotropic material. It turns out that reliable and reproducible results can be obtained as long as the actual needle-specimen system under each single measurement is close enough to its corresponding idealized model.

Abstract:Track quality is essential to track safety. A reasonable forecast of track quality is a good instructor for the department of railway maintenance when arranging track maintenance schedule. Track quality index (TQI) can evaluate track quality in a unique track interval. With research on the changing tendency of TQI, this paper proposes a forecasting method that combines unequal-interval Grey Model and Elman Neural Network. The Grey model GM (1, 1) is previously exploited to obtain an approximate forecast of original TQI series and then the residual error correction is corrected by using Elman Neural Network optimized by Genetic algorithm. The new method takes random fluctuation of changing tendency of TQI into consideration, thus the historical data can be treated more efficiently. The proposed method is demonstrated with practically measured data of Shanghai-Kunming Railway Line. The forecasting results show that, comparing to other forecasting methods, the method which uses Elman Network to correct residual error correction reaches higher forecasting accuracy at root mean square error, correlation coefficient and determination coefficient.

Abstract:According to the rim of dual mass flywheel (DMFW) drive plate is 3~5 times thicker than the original thin plate, such formation of multi-steps spinning thickening forming process was investigated based on the SIMUFACT software. According to the analysis of spinning thickening deformation, the mould and the finite element model of spinning thickening forming were established. Taking the 1st step spinning thickening forming as an example, the stress field distribution and the deformation of the sheet metal during the forming process were analyzed. Meanwhile, 4 measuring points on the surface of the plate were used to analyze the metal flowing in the 1st step spinning thickening forming process, and such forming rules of the steps ranging from 2 to 5 were summarized. Taking the roller feed speed, friction factor and rotation rate as independent variables, Three factors and three levels orthogonal experiments were carried out to obtain the maximum forming load and thickening rate data. The grey system theory was used to optimize the process parameters of the spinning thickening forming of DMFW drive plate, Such experiments were conducting on the CDC-4S80 spinning machine. The results show that the radial force is the largest, the tangential force is the second, and the axial force is the smallest, in the process of spinning thickening forming, ; The smaller friction and the roller feed rate can improve the quality of forming parts. The rim of the plate was thickened from 3 mm to 10.5 mm by physical test, which verifies the feasibility of the forming process and the grey system theory.

Abstract:To overcome premature convergence and local minimum of Ant Lion Optimizer (ALO) algorithm, an innovative ALO algorithm based on self-adaptive chaos search which is combined with Tent chaos algorithm is proposed. Tent mapping is applied to initialize individuals in the search space. Self-adaptive adjustment of chaos search scopes is presented to get the better solution, perfect the fitness of the worse individuals and enhance the whole individuals' fitness and efficiency. At the same time, Tournament selection strategy is used to screen out ant lion individuals. At the end, ants' random walking behavior is optimized by chaos operator to form a parallel search mode with ant lion foraging behavior, which is beneficial to the overall optimization performance to obtain the optima. Algorithm's performance is tested through complex benchmark functions with high-dimension and path planning problem. Experimental results of the former further confirm that, as for benchmark functions with 30 dimensions, it costs about 0.5 s to converge to the optimum, and for 50 dimensions, it is about 2 s. Compared with ALO and other optimization algorithms, the improved algorithm not only accelerates the convergence rate and improves the precision, but it is also fit for complex high-dimensional functions optimization. Path planning test manifests that for airspace with 7 menaces, while the search dimension is 10, after 0.939 s, and 30 iterations, the global optimum of the path cost function can be got. Compared with ALO, it has advantage in speed and accuracy to get the specific path, and it is of great value in actual problems.

Abstract:The successive cancellation list (SCL) decoding algorithm of polar code is of bitwise, which is the main source of the decoding complexity and latency. In view of the analysis of the advantages and disadvantages of the existing SCL and multi-bit decision decoding, each set of multi-bit code words (multiple independent channels) is treated as a whole to construct a multi-bit virtual channel, which can transmit multi-bit code words and decode synchronously. Therefore a construction of multi-bit virtual channel based on the SCL algorithm under the AWGN channel is proposed to against the problem of high complexity and large delay for decoding system caused by SCL decoding. It is also enhanced by combining with the establishment threshold of the decoder tree nodes to reduce the number of tree nodes' split. Under the AWGN channel, the bit error rate and the frame error rate performance of the virtual 2, 4 and 8 bit channels are simulated. The simulation results show that the decoding performance is close to that of the traditional SCL algorithm when the default threshold S=30 in the case of a virtual 8-bit channel. The total number of nodes is reduced by 63.7% and the total number of additions is 17% of the 8-bit decision decoding algorithm. The proposed algorithm can effectively reduce the calculation complexity of decoding and the hardware storage. It is more suitable for hardware implementation and has a certain practical value.

Abstract:There are direct relations between machining deformation and stress field of aerospace Al-based alloy thin-walled component. With stresses profile characteristics of the component after milling processing, building a mechanical model of stress-deformation analyses that effect mechanism of stress on deformation and then get the corresponding relationship between the both of deformation and stress. The process is as follows: the different degree stress-deformation distributions of the thin-walled component after milling is obtained by shot-peening treatment, and then the experimental values of the deformation and the calculated values of analytic functions under different technological conditions are compared and analyzed using layer removed method, XRD stress and three-coordinate shape measurement technology. The experimental results show that the deformation deflection is 3~12 μm between the experimental value and the calculation, and the deviation is within the deformation uncertainty. By this work mentioned above, the deformation function can achieve the predictive calculation and analysis of the deformation degree under the premise of the geometrical shape, the surface stress and the initial stress of the component. The research satisfies the demand for engineering application.

Abstract:Computing the bisimulation partition of a graph plays a key role in various application areas. Two nodes are bisimular if and only if they obtain the same signature. Faced with a big graph, traditional in-memory algorithms can hardly meet practical need. Recently, researchers have proposed two kinds of MapReduce based distributed algorithms to handle bisimulation partition on the big graph, both calculating localized bisimulation. MapReduce based algorithms suffer from huge network communication burden, at the meantime, we only need local information to calculate the signature of a node. Motivated by these observations, we propose an algorithm based on a graph processing system that only uses local information to calculate the signature of a node. We argue that our algorithm can make considerable reduction on the network transportation during computation. In the experiment calculated on the big graph which contains billions of edges, our algorithm can be seven to sixteen times faster even without graph pre-partition.

Abstract:To study the hydrodynamic performance characteristics of tandem turbines, put forward the reasonable layout scheme and increase power generation efficiency of tidal power station, the transient numerical simulations are performed to investigate a three-blade vertical axis tidal turbine, and the upstream turbine's hydrodynamic performance and the wake effect on the downstream turbine are analyzed. Considering that the turbine's blade has the same profile on the span, the turbine model is simplified to 2D model and the open source fluid mechanics software OpenFOAM is used to simulate the flow field. Firstly, changing law of turbine's power coefficient with tip speed ratio is obtained through the simulation of a single turbine. By comparing with the experimental data in reference, it is verified that the OpenFOAM numerical model can predict the hydrodynamic performance of the 2D vertical axis turbine accurately under the reasonable setting by using the k-ω SST turbulence model and the PIMPLE algorithm. Then two tandem turbines cases are simulated, the distance between two turbines is from 5D to 50D, where D stands for turbine's diameter. The wake characteristics of the sections behind turbines and the influence characteristics of turbines' distance on the downstream turbine's hydrodynamic performance are obtained. Finally the results show that when the distance between two tandem turbines is more than 40D, the effect of upstream turbine on the downstream turbine could be ignored.

Abstract:The prior information of channel can't be induced in the process of channel estimation in MIMO relay communication system. To solve the problem, a novel semi-blind channel estimation based on variational inference is proposed. In this algorithm, some latent hyper-parameters such as factor precision, noise precision are introduced into the algorithm, and channel estimation probability model is built based on nested PARAFAC tensor decomposition. Since the posterior probability distribution of the channel parameters is complex, some point estimation methods, such as traditional maximum likelihood and maximum posteriori algorithm, are difficult to implement. The iteration formulas of factor matrix, factor precision and noise precision are deduced by the idea of variational inference principle, making the q distribution, which has the factor decomposition form, approach the unknown parameter posterior distribution. In addition, low bound of model evidence, model initiation and algorithm complexity are also analyzed. The algorithm can utilize the prior information of channel to improve channel estimation performance. The parameters can be tuned automatically, and complexity is linear with the dimension of observed data. Simulations show that the proposed algorithm has better estimation performance under flat Rayleigh channel condition, compared with No-blind algorithm, Alternating Least Square (ALS) based algorithm and No-linear Least Square (NLS) based algorithm, and has lower complexity and faster convergence speed, compared with alternating least square algorithm.