摘要:Abstract:Through the shaking table test, studied the distance of different shape and size rock falls under different peak acceleration seismic, sine wave frequency and slope shape, and cleared the influence of rock falls on the buildings at the slope foot.The results showed that: the distance of rockfall increased with the increasing of seismic peak acceleration, where the maximum distance was spherical rockfall, cylindrical, the square rockfall was inferior to the spherical rockfall, and the minimum distance was rectangle rockfall. With the slope angle increased, the distance of rockfall first increased and then decreased, and the maximum distance of rockfall appeared when the angle of the slope reached 50°. The distance of rackfall increased with the slope height increased. With increasing of sine wave frequency, the distance of rock falls decreased first and then increased. And the maximum distance from largest to smallest were spherical, cylindrical, square and rectangle rockfall. Based on the probability statistics method, the maximum distance of rockfall distributed within the 1.6H, which provide reference for protection design of the structures at slope foot.
关键词:shaking table test;earthquake;distance of rock fall;shape and size of rock fall;slope shape
摘要:Abstract:Alternating actions of excavation and unload have a very important influence on deformation and instability of high and steep bedding rock slopes with upper steep and lower gentle style, which are widely distributed i in China's western regions. Taken the high and steep bedding rock slopes belonging to Jianshan phosphorite mine, Yunnan phosphate group Co., Ltd as engineering background, model experiment had been conducted on the basis of similarity theory, alternating actions of excavation and unload effects on their deformation and instability were investigated. Research results showed that: deformation displacements of high and steep bedding rock slopes with upper steep and lower gentle style progressively becomes larger with depth of excavation gradually increasing. Its integral deformation laws are that upper slope moves downward, while lower part uplift to the free surface due to compression. At the same time, when the excavation is to a certain depth, many strike cracks can be developed into the slope, then they continuously extended to the slope body in pace with increasing excavation depth; during the whole process of excavation, potentially most dangerous slip surface of composite-style with plane and circular arc surface have be developed gradually into the slope, and deformation and instability characteristics of slip—bending—shear style have be shown; although the deformation of the slope is reduced, the trend of deformation and instability of slope is not changed, so the shallow unloading effect is not ideal, while deep unloading effect is very obvious and cracks in the slope almost completely eliminated. Research achievements play a reference guiding role for early warning and forecast and control design of high and steep bedding rock slopes.
关键词:alternating actions of excavation and unload;upper steep and lower gentle style;high and steep bedding rock slopes;similarity model experiment
摘要:Abstract:Large-diameter steel pipe pile is a kind of single pile which is widely used in offshore wide farm. Based on the centrifugal model test results of large-diameter steel pipe pile under horizontal load, we conducted the centrifugal numerical test by 3D finite element method. Elastic modulus has the important influence on the deformation of the steel pipe pile under horizontal load. Through optimizing the elastic modulus, it is verified that the distributions of the deformation of pile top, the deflection and the bending moment of pile shaft by numerical method is consistent with those in model test. Then, we obtained the load-displacement relationship of steel pipe pile at different depths under horizontal load, and got the systematic knowledge for the horizontal resistance of foundation in different soil layers. So, the numerical test can overcome the deficiency that the monitoring data is limited and the arrangement of monitoring instrument is difficult in model test. Moreover, the load-displacement relationship by numerical method has common applicability, and the numerical test results can be references for actual projects.
摘要:Abstract:Ili loess, a typically large-thickness collapsible eolian loess, has parameters that uniformly changes along the depth direction. According to this characteristic and engineering requirements, centrifugal model tests of the self-weight collapsible loess, which has been conducted based on double-line method, is introduced in this paper. The results show that: 1) Settlement is highly affected by the boundary of the model box. 2) When the thickness of loess’ layer reaches 30m, for different saturation soil sample, boundary effect rate is much more than 25% and it is negative exponential attenuation with the increase of load. 3) Self-weight collapsible amount measured by laser displacement sensor, a more reliable measuring way than image acquisition system, is 296.2cm. By analyzing the three types of tests for collapsible loess, the paper gets that compared with that of indoor compression test, the value of centrifugal model test is closer to that of field soaking test. The correction coefficient of centrifugal model test for the Ili loess is 1.19. The centrifugal model test, which can save test cycle and cost, could be used for evaluating the collapsibility of loess owing to its similar stress state with natural foundation.
关键词:centrifugal model test;large-thickness collapsible loess;double-line method;Self-weight collapsible amount
摘要:Abstract:In order to study the seepage characteristics of rock mass containing periodic fractures without filling, an analysis model for groundwater flow was built. In this model, two assumption conditions were given for the interface between rock and facture. One condition was that the velocity was equal and the other was that the shear stress was continuous. According to the given conditions, the velocity distribution in fractures was deduced using the Navier−Stokes equation and that in rocks was deduced using the Brinkman-extended Darcy equation. The equivalent permeability coefficient of the theoretic model, i.e. the rock mass containing periodic fractures without filling, was derived from the linear Darcy's law. The derived explicit equation shows that the aperture of fracture plays an important role in the groundwater flow in rock mass. In addition, the laboratory seepage tests were performed to validate this model. During the testing process, concrete and fracture between two pieces of concretes were used to simulate rock and fracture respectively. The average permeability coefficient value measured by tests and the equivalent value calculated from the theoretical expression were in the same order of magnitude, and the error was very small. The test results validate the expression given in this study.
关键词:periodic fractures without filling;equivalent permeability coefficient;Navier−Stokes equation;Brinkman-extended Darcy equation;seepage tests
摘要:Abstract:According to the characteristics of salt rocks of Chinese energy underground storage which is high impurity uniaxial experiments were carried out on the pure salt rocks, argillaceous salt rock, halopelite, glauberite salt rock and glauberite rock . All of these specimens were taken from a proposed salt rocks reservoir, and then were divided by the impurity content and type . Through experimental study, the following results were obtained: (1) both the mudstone impurity and glauberite rock impurity can increase the compressive strength of pure rock salt and reduce its deformability, glauberite rock impurity has more influence on rock salt than the mudstone impurity;(2) with the increase of mudstone impurities, the active area of AE counts is postponed, high energy rate is over backwards, cumulative AE counts curve and the accumulative energy curve presented by the changing process from convex to concave;(3) with the increase of glauberite rock impurities , the time when the maximum AE count appeared was backwards , the AE counts and energy rate was relatively stable through the whole test , the accumulative energy - strain curve step growth more obviously.
关键词:rock salt with impurity;mudstone;glauberite;acoustic emission;uniaxial compression
摘要:Abstract:Rainstorm flood, landslide and debris flow frequently occur in mountainous area, and then lots of coarse particles such as pebble, gravel, boulder with wide size distribution have been entered into open channel. Boulders obviously affect characteristics of flow structure and bedload transport around these boulders. On the basis of the field investigation of boulder riverbed and series of flume experiments with limited water depth, the influence of boulder on the riverbed deformation and bedload transport rate have been analyzed and discussed. The results show that the increase of flow turbulence becomes stronger behind the boulder with the decrease of boulder particle relative exposure degree (D/h, ratio between boulder equivalent diameter and mean water depth), the scour ability has been enlarged, and the more bed material were transported. According to the bed surface coarse degree and boulder dissimilar state, several of experimental cases have been carried out in order to understand the variations of bedload transport rate, changes present that the influence of boulder on bedload transport rate may be several times to ten times with the premise of the flow-sediment conditions in this study.
关键词:Mountain river, wide size distribution sediment, boulder, bedload transport rate, sediment supply
摘要:Abstract:Based on the heating method included in the temperature tracer method, a new monitoring system characterized by combinations of FBG sensors and the water cycling heating system is presented to apply on the leakage monitoring of rock-fill dam, aiming to overcome the disadvantages of conventional means such as spatial discontinuity and execution inefficiency. The main structure of this novel system is designed by embedding the FBG sensors in the pipes of the water cycling heating system. Several major components of this system including boiler, sub-water catcher, heating pipeline, and FBG temperature sensors can play the functions of heating, allocation, transporting and monitoring, respectively. The temperature of heated water along the pipeline can be measured by embedded FBG. Based on the operational principle of this monitoring system, a series of experiments that take the influence factor of leakage passage (location and amount) and leakage intensity into account are performed for simulating the concentrated leakage monitoring. A cooling curve can be derived from the measured temperature data using Newton’s Law of cooling. The testing results in temperature decreasing phase are basically accorded with fitting results determined by Newton's law of cooling. The coefficient ζv fitted from cooling curve can be used as the key factor to identify the location and quantity of leakage, because the magnitude of ζv has a good agreement with the leakage state. Therefore, the seepage monitoring system and its corresponding leakage identifying method are valid.
关键词:safety monitoring of rock-fill dam;temperature tracer method;optical fiber sensing;water cycling heating;Newton's law of cooling
摘要:Abstract:This paper simulated the process of field loading test for laterally loaded rock-socketed piles by finite element method. The load displacement curves obtained were consistent with the field test results within the loading range, and the rationality of the numerical model is verified. Continues load increment was done until the pile failed using the numerical model to generate a full load - displacement curve. Based on the existing foundation reaction modulus Ki solution method, regression analysis was carried out with the field test data and the empirical formula of the initial slope Kh of the load displacement curves was obtained. Hence, a hyperbolic function theory for load displacement curve of laterally loaded rock-socketed piles was proposed. The accuracy of the proposed method was verified by contrasting with the field test curves. Using the conventional four laterally loaded rock socketed pile capacity calculation method, the numerical test was extended through laterally loaded pile top displacement curves analysis to carry out analysis of the applicability of the different methods. The method of hyperbolic function was suggested for calculation of the bearing capacity of rock-socketed laterally loaded piles after comparing the various methods.
摘要:Abstract:This paper presents a study on the pull-out behavior of the PBL shear connector and the PZ shaped composite dowel, a total of 15 specimens, by using a self-designed monotonic loading equipment, aimed to study the ultimate uplift capacity, initial crack load, failure pattern and strain states of the steel web during the loading progress, the influence of the hole size, depth of embedment and form of the shear connector were considered. Based on the results of the PBL shear connectors, the calculation formula for ultimate pull-out bearing capacity was established. The results showed that the enlargement of the hole size or embedment depth could lead to direct increase on the ultimate pull-out bearing capacity of these two shear connectors. Ceteris paribus, the PZ shaped dowel had bigger bearing capacity than the PBL one, also better fatigue property based on other researchers’ work, which could lead to further study on the mechanical property of the PZ shaped dowel with various geometric dimensions.
摘要:Abstract:This paper presents a finite element analysis on flexural behaviors of box girders with corrugated steel web (hereafter called box girders). Displacement, stress and strain under static load is used to confirm the failure mechanism and stress weak spots. Meanwhile, fatigue life of box girders in different conditions is predicted by analyzing local stress and strain action, which is acquired by finite element analysis. Based on the fatigue damage and fatigue fracture theory, the predicted results of cracks initiation location and fatigue life are approximately agree with the experimental. Results show that the standard C of America code AASHTO2004 is suitable for designing the fatigue life of box girders, What is more, the Palmgren-Miner learner accumulated damage criterion can be an efficient method to analysis the fatigue damage process, which provides a reference for box girders fatigue design in practical engineering.
SHEN XIA,YAN ZHONGLUAN,HUANG JUPING,FENG JINGJIE,LIANG RUIFENG,Li Ran
Vol. 48, Issue 6, Pages: 83-89(2016)
摘要:Abstract:In order to evaluate the influence of the deep water discharged from the sewage on the marine environment accurately, numerical simulation of the deep water discharge is of great significance. Considering the effect of the buoyant jet caused by different salinity and sediment, three-dimensional RNG κ-ε turbulence model was developed, including sediment scour equation and salinity advection equation. The mechanism experiment was carried out to verify the flow field and salinity field of the mathematical model. It showed that the simulation results are in good agreement with the experiment data. The simulation for a specific discharge case was conducted to predict the hydraulic characteristics, the distribution of the salinity and suspended sediment after the sandy water discharge with low salinity was discharged into the sea at a certain depth. It demonstrated that the sewage discharged into the sea shows buoyant jet feature due to the difference of the density between the inflow and the sea water. The axis of the buoyant jet exhibits a parabolic form in the vertical plane. After the buoyant jet having reached the surface of the sea water, the salinity in the butterfly-shaped diffusion zone decreased gradually while the sediment concentration increased gradually. With the time went on, the impact of the sediment discharge on the sea water was more and more serious. In the beginning, the diffusion rate of the pollution zone was relatively faster, and it was gradually slowing down with time.
摘要:Abstract:Under the influence of GDOP(Geometric Dilution of Precision),the localization noise of MLAT system relate directly to target's geometric position, which will degrade the performance of traditional test method. This paper proposes a test method of target stationary/motion state in the airport surface for MLAT surveillance system. The proposed method is able to eliminate GDOP effects and its carry on motion states test by constructed the normalized statistic that works on the principle that when target are moving, the normalized statistic has a noncentral distribution, versus a central Chi-Square distribution when there is static. Then a test method using sliding window is put forward in order to meet different performance requirements. The proposed methods in this paper are rigorously derived and analyzed in detail. Simulation results prove the effectiveness and rationality of proposed method. The proposed method has been applied into an airport project.
关键词:Multilateration(MLAT);statistical tests;motion state analysis;sliding window;positioning algorithm
摘要:Abstract:As the excavation of auxiliary opening L3 at the right bank of underground cavern in Wudongde hydropower station was carried out, the excavation disturbance led to rapid growth of surrounding rock deformation and surface rupture along interlayers of rock mass. The characteristics of deformation and failure of surrounding rock was preliminarily described based on the geological conditions, monitoring and geophysical prospecting data, and construction process. On the basis, deformation and cracking mechanism of surrounding rock mass was discussed by combining the surrounding rock mass structure characteristics, geostress and construction process as well as the three dimensional distinct element method. The results were as follows: (1) The steeply dipped rock strata was basic condition which caused rapid and continuous growth of surrounding rock deformation at upstream side wall of 7#~8# unit since the excavation of auxiliary opening L3. (2) The upstream side wall was exposed to considerable excavation disturbance effect due to the rapid construction of auxiliary opening L3. Then, the upstream side wall surrounding rock became a weak link and caused surface either open or slip. (3) The characteristics of deformation and failure of steeply dipped rock strata was very sensitive to excavation disturbance, which could be effectively reduced through reasonable control of the scale excavation and unloading disturbances. The results above could provide important reference for stability analysis and design support of similar steeply dipped rock strata of large span underground projects.
关键词:underground project;Wudongde hydropower station;steeply dipped rock strata;deformation and cracking mechanism;three dimensional distinct element method
摘要:Abstract:To study the influence of temperature on the modes of two-tooth difference swing-rod movable teeth transmission structure, the temperature field and the thermal stress of the central wheel of the transmission were analyzed.On the basis of thermal structure coupling analysis of two-tooth difference swing-rod movable teeth transmission, modal analysis of the cental wheel was carried out.
The results of modal analysis under different working conditions were compared,
and the influence of temperature field on the modes of the central wheel of the transmission was investigated. The results showed that the natural frequencies of the central wheel are influenced in different degrees by the change of mechanical properties of the materials caused by temperature change, and that the natural frequency variation caused by thermal stress is less than that caused by changes in mechanical properties of materials. The analysis results are in accord with the actual working conditions,provide a theoretical basis for vibration reliability design and dynamic characteristics´research of two-tooth difference swing-rod movable teeth transmission.
摘要:Abstract:Abstract: When traditional A-B-C method is used for sided balancing machine, the measurement error is too large to meet the requirement as ignoring the influence of the vibration system’s torsion. Assessing application scope of ABC method is crucial to improve dynamic balance accuracy. So this paper established the vibration model of swing frame for sided horizontal balancing machine, calculated the location of vibration center of swing frame, deduced the torsional stiffness formula of a rectangular plate spring twisting around the vibration center, analyzed the relationship between torsional stiffness and location dimensions of the plate spring; Verified the measurement error caused by using traditional statics methods by experiments, compared the torsional stiffness deduced by experiment with that calculated by theory, analyzed how the torsional situation of plate spring effects the separation ratio of sided balancing machines; estimated the influence of torsional vibration on measurement error and assessed the application scope of ABC method.
关键词:dynamic-balancing;plate spring;torsion center;torsional stiffness;separation ratio
摘要:Abstract:Ultra-precision noncircular section components, such as hyperbolic quadrupole in mass spectrometer, can be machined by diamond turning assisted by slow tool servo (STS). However, the bandwidth of STS is usually small, which limits the STS's capability in following the required tool path, leading to a large form error. To reduce the form error, this paper proposes an approach to apply variable spindle speed (VSS) to STS-based turning. Design of the VSS trajectory based on the noncircular profile of the optical component was presented. To validate the proposed approach, experiment on the application of VSS in the STS-based turning process was conducted to machine a typical noncircular component which has a sinusoidal-wave along peripheral direction. The input signals which actuate the tool servo were also analyzed by using FFT to obtain their frequency spectrum.Experiments results show that the form error of the noncircular component in VSS way is less than that machined in CSS way.Therefore,the proposed approach is effective increasing a higher form accuracy of the machined noncircular optical components.
摘要:Abstract:To study the forming principle of circular arc gear shaped by rotary cutter ,get a reliable circular arc gear 3D model to guide the manufacturing processing and simulation calculation, according to the development of the cutter and the gear workpiece, establish a new reference coordinate system, use the basic knowledge of spatial meshing theory and differential geometry, deduced the tooth surface equations of rotating cutter circular arc gear, derived involute tooth profile of the middle section , analyzed the forming principle of rotating cutter circular arc gear, discussed the range of the parameters of the equation, gave the Calculation formulas under given parameters of the gear workpiece rotation angle, got the digital and three-dimensional model of circular arc gear shaped by rotating cutter. It provides a reference for the parameter selection and design of the circular arc gear, and provides the experimental basis for the following simulation and experiment analysis.
摘要:Abstract:In order to treat the problem of parameter reconfiguration of the cognitive radio system, an improved hybrid cross-layer decision engine based on the Cellular Quantum Artificial Bee Colony Algorithm (CQABC) and the channel case base was proposed. In the decision engine, the parameters at different layers of a wireless communication network were considered and the overall performance of the network was the optimization goal. A fast strategy with quantum rotation angle adjustment based on cellular automata and social cognitive strategy and two kinds of chaos initialization methods were used in the proposed CQABC. Furthermore, the historical experience and expertise was referred to build up the case library of the cognitive radio parameters based on the channel gain for a quick decision-making process. The results of simulation showed that the cross-layer decision engine is capable of dynamic re-configuration of parameters according to changes in the wireless communication environment and user requirements, on the meantime the proposed decision engine has better convergence, precision and stability than the traditional decision engine based on Binary artificial bee colony algorithm and quantum genetic algorithm.
摘要:Abstract:Abstract:The flow filed model of cutter-suction was established according to the structure and working principle of cutting head.The characteristic of flow filed was researched by utilizing Computational fluid dynamics theory and Fluent simulation software.The law of flow field and velocity field and The change law of solid volume fraction of outlet was analysed. The main working parameters(speed of cutter and velocity of sucker mouth) which effected cutter-suction capacity was simulated in detail. The relationship between the parameters and the cutter-suction capacity was obtained. Increasing the cutter-suction flow is beneficial to the improvement of the cutter-suction capacity,when the cutter-suction flow over 2000m3/h,continue to increase the cutter-suction flow leads to high velocity of solid phase in the pipeline,and increase wear between solid particles and pipes;When the speed of cutter below 20r/min,increasing the speed of cutter is beneficial to the improvement of the cutter-suction capacity; When the speed of cutter over 20r/min, continue to increase speed of cutter will reduce the cutter-suction capacity significantly.
关键词:Marine gas hydrate;cutting head;solid liquid two phase flow;cutter-suction characteristic;numerical simulation
摘要:Abstract:The tooth surface coordinates are difficult to be got about HFT hypoid gear. For the problem, a mathematical modeling method of tooth surface based on MATLAB is presented in this paper, by which the tooth surface coordinates can be calculated automatically. Firstly, according to the Gleason machine, the coordinate transformation matrix is established from the machine tool to the workpiece. Secondly, the equation of tooth surface is established based on the gear machining parameters and the tooth germ parameters. Then, the estimated value of the gear meshing mean points is obtained based on the gear machining theory, and other points’ coordinates of the tooth surface can be solved by taking this point as the initial value of the function. Finally, SURFACE, the solution programmer of tooth surface coordinates of HFT hypoid gear, is established. To validate the method presented, taken the hypoid gear as an example, 3D models of the gear sets is established through importing tooth surface coordinates into CATIA Software. There is a little difference between the tooth surface of the established model and the actual gear, so this method can be applied to solve this kind of gears.
关键词:hypoid gear;meshing mean points;tooth surface coordinates;3D models
摘要:Abstract:It is theoretically analyzed that the upper bound capacity of a gray image block with n pixels is not more than ⌊log2 n2+ 2 n+2⌋ binary bits, when not more than 2 pixels are modified and every pixel is modified not more than
1. It is proved that such a weighted matrix can’t be constructed, but a way to construct a suboptimum weighted matrix is proposed in this paper , the matrix can get better balance between the hiding capacity and the imperceptibility compared with the optimum maxtrix, the suboptimum can hide one bit less than the optimum matrix at the worst condition when n is not more than 25 million pixels.
摘要:Abstract:In order to construct a Perfect and Ideal image threshold secret sharing scheme (Complete Scheme) and improve sharing efficiency, a novel secret image sharing complete scheme based on array codes with a (t, n) threshold structure was proposed. The scheme takes image pixel values, in binary form, as the data in disk array, encodes image pixel array using the method of array code. The operation of the scheme is XOR only, and the process is in a parallel way, while the threshold structure is Perfect and Ideal. The results show the proposed scheme is more efficient and has higher security.
摘要:Abstract: In order to improve the grid-connected efficiency of the large scale offshore wind farm, HVDC was used widely. However, when the wind turbines and wind generators are two-mass or multi-mass shaft model, the control mode of power electronic equipment at rectifier may introduce negative damping to nearby generators and cause torsional vibration named subsynchronous torsional interaction. This phenomenon may lead to system instability. This paper gets electrical damping coefficient curve base on complex torque coefficient approach firstly, and works out the phase lag between the electromagnetic torque and the rotate speed at nature frequency of the shaft. Then a supplementary subsynchronous damping controller is designed to damping the oscillation between shaft masses. The effectiveness of the controller can be verified through time domain simulation and eigenvalue analysis. The simulation results showed that the oscillation amplitude could decreased quickly when the controller was added to the system, which provides a theoretical basis when the offshore wind farms are connected to the onshore grid through a HVDC transmission line.
摘要:Abstract:In order to overcome the defects of high complexity and probability of falling into the local minimum occurred in the present neural network (NN) based predistortion methods, a new predistortion method based on real-valued focused time-delay fully connected cascade neural network (RVFTDFCCNN) trained by the joint algorithm of orthogonal differential evolution krill herd (ODEKH) and Neuron-by-Neuron (NBN) was proposed. Both the predistorter and the inverse estimator in the predistortion system were modeled by RVFTDFCCNN. The initial parameters of RVFTDFCCNN were first obtained by the ODEKH algorithm with strong global search ability. Then the RVFTDFCCNN was further trained by the Neuron-by-Neuron (NBN) algorithm. According to the chain rule of complex function partial derivative, the calculation of the Jacobian matrix in the NBN algorithm was optimized at two levels. In the simulation, the DTMB wideband signal was used as the input signal of the predistortion system. The simulation results showed that, with the same requirement of the training error and the generalization error, the computation amount of the NBN algorithm in training RVFTDFCCNN was considerably reduced than that in SHL neural network. The ODEKH algorithm converged faster than the traditional krill herd algorithm. The joint ODEKH-NBN training algorithm had higher training accuracy than the Levenberg-Marquardt(LM) training algorithm by one amplitude order, and outperformed the latter algorithm by 2dB in terms of adjacent channel power ratio (ACPR) of the HPA output signal after predistortion. Thus it was concluded that the proposed predistortion method owned low complexity and good predistortion performance.
关键词:High power amplifier;Predistortion;Krill herd algorithm;Neural network;Neuron-by-neuron algorithm
摘要:Abstract:In order to improve operators’ situation awareness of dealing with loop closing scenarios, a fast simulation method of distribution network loop closing based on compensation method and nodal equation was proposed to simulate the influence of loop closing on distribution network rapidly. The change of loop closing electrical quantities was depicted by compensation method, and loop closing simulation formulas based on incidence matrix and admittance matrix were derived with nodal equation to quantitatively describe the influence of loop closing on distribution network. Only running state and network parameters before loop closing were needed to calculate the state after loop closing, and there was no need to recalculate power flow. The PG&E 69-node system and a practical distribution network were simulated, and the results proved the validity and practicality of the proposed method.
摘要:Abstract:To solve the incompleteness of single data for facial expression recognition, we concatenated images and mark points as inputs. For the complexity of hand-crafted feature in convolutional pattern recognition methods, we adopted neural network to extract feature automatically. The method concatenated images and mark points as basis and framed by neural network. Sparse Auto-encoder was used to pre-train the network and make the network sparse. In addition, structured regularization was added to restrict the connection between different inputs and neuron in hidden layer. The experimental results showed that the concatenation of images and mark points could present the facial expression more thoroughly. The application of Sparse Auto-encoder and structured regularization could help the network extract the key feature effectively and learn the importance of different data to facial expression recognition automatically.
关键词:Image Mark point Data fusion Sparse Auto-encoder Facial expression recognition
摘要:Abstract:the effective representation of the video is the key and difficulty in human action recognition. In this paper, we proposed an improved string of feature graphs method to describe an video, which combines submodular optimization method and graphic matching technique in the framework of dynamic programming. Firstly, space-time feature points in a video are obtained by utilizing spatio-time interest point detector, and leveraging the submodular means considering the time order divides the video into many small time intervals. Then the representation of the video can be transformed into a string graphs which are constructed based on these feature points falling in the corresponding time interval. Finally, measuring the similarity of pair of videos is implemented through using the techniques of the Reweighted Random Walks for Graph Matching (RRWM) and Dynamic Time Warping (DTW) between string graphs from two videos respectively. Here we provide comparisons against other methods on the two published datasets (KTH and UT-interaction) and the results demonstrate that this algorithm is effective and feasible.
关键词:Submodular Optimization;String of Feature Graphs;RRWM;DTW
摘要:Abstract:Person tracking problem under the indoor environment often face light, occlusion and similar color interference, in order to enhance the tracking stability and accuracy in the situation, this paper proposes a root mean square (RMS) unscented H-infinity filtering method in Considering interference noise perturbation. First of all, Unscented transformation is used to replace the complex Jacobi matrix calculation under the framework of extended H-infinity filter. The method, which using Gaussian density approximating filter distribution, reduce the linear approximation error, and reduce the interference of system model noise to the estimation values. Second, the Cauchy decomposition to the state covariance matrix RMS, which uses diagonal elements to calculate, reduce the disturbance of observation noise error and filter calculation consumption. Video tracking experiments showed that compared with the EKF、UKF and PF algorithm, the proposed algorithm can effectively improve the pedestrian tracking accuracy in mutation, and reduce the time consumption.
摘要:Abstract:In order to make up the deficient of model verification in enterprise application, a new method aiming at application system is proposed based on the state of art in the research filed of business process modeling and verification By use of analysis principle of petri-net, the idea for business process verification is deduced based on petri-net characters such as incidence matrix,home state, transition sequence, and a problem-solving algorithm through the invariants of petri-net is well formed. Experiments are designed, and the proposed algorithm is tested in both theory analysis and application simulation perspective, which proves the effectiveness of the algorithm by comparison of two perspective output. Moreover, efficiency performance tests are carried out to confirm that, the method of this paper is better than coverage graph and causality methods. Finally, a tool package is developed according to the algorithm, the application capability of which is displayed in practice. Therefore, it can be concluded that, the verification algorithm presented is reasonable and efficiency, which is suit for practice using.
关键词:Business Process Management;PetriNet;Workflow;Model Verification
摘要:Abstract:In order to quantitatively evaluate the ability of the pilot to perform the flight tasks strictly in accordance with the specified flight procedure, the model of flight procedure representation of terminal area is studied. Typically, a true flight track of the terminal area has about 150 track points, but the standard flight procedure has only 4 to 9 points and is not sufficient to describe a full track. Aiming at this problem, a method to measure the similarity of the flight tracks based on the weighted track area is proposed and the CURE algorithm is improved to realize the flight track clustering. Based on the analysis of the relationship between the average track and the representative tracks of the clustering set, the representation model of flight procedure track is established. In the end, the proposed method is verified by using a standard departure flight procedure and the actual flight tracks in terminal area of a large international airport .The experiments show that the method proposed in this paper can not only achieve better flight track clustering performance, but also can accurately represent the flight track of the flight procedure. Therefore, the research in this paper provides the theoretical basis and potential practical application for the analysis and assessment of pilot's flight quality.
摘要:Abstract:In order to solve the problem that the multi-particle joint measurements are difficult to implement with the existing technology in the quantum key agreement protocols based on multi-particle entangled states , a new two-party quantum key agreement protocol was proposed based on three-particle GHZ states. Two participants in the protocol could fairly establish a shared secret key by using the measurement correlation property of three-particle GHZ states and the delayed measurement technique. Since the new protocol only used Bell measurements and single particle measurements, it was easy to achieve on the basis of the existing technology. The security analysis showed that the protocol could not only resist against participant attacks and outsider attacks, but also resist against two kinds of Trojan horse attacks. The protocol was secure not only in the noise-free quantum channel but also in the quantum noise channel. Finally, the comparison with the existed secure two-party protocols showed that the protocol also achieved higher qubit efficiency.
摘要:Abstract:To learn the context semantic information of word for paraphrase identification, the model for representing sentence semantic distances based on word embeddings was proposed for paraphrase detection tasks. Firstly, a large-scale word vectors was trained with word2vec model, which embedded the semantic information in word distributional representation. Then, the travel cost between words in sentences computed with Euclidean distance in the word2vec embedding space. Finally, the model from word embeddings to sentence distances was built based on EMD, and sentence transportation matrix was presented for distance metric between sentences. The sentence semantic distances were used for paraphrase recognition. Experiments based on SemEval-2015 PIT Task showed that the proposed model approximates to the baseline in supervised method and gives an improvement of 5.8% in unsupervised methods, compared to the weighted matrix factorization.
摘要:Abstract:In view of the problems of the delay of harmonic detection link and bad robustness, tracking accuracy was not high of harmonic current tracking link for Shunt Active Power Filter, the mathematical model of decoupled system was established and global integral sliding mode control for Shunt Active Power Filter based on RLS algorithm was proposed. In harmonic detection link, improved id-iq method of instantaneous reactive power theory was used and the conventional Butterworth low-pass filter was been replaced by RLS adaptive filter to solve the problem of blind-spot detection in a fundamental cycle (20ms) due to the delay of conventional Butterworth low-pass filter. In harmonic current tracking link, global integral sliding mode control was used and the global integral sliding surface was introduced to export the control law taking the rapidity of global sliding and the accuracy of integral sliding into account based on Lyapunov stability theory. In conditions of solving the delay of harmonic detection link, contrasting global integral sliding mode control strategy with traditional PI control and hysteresis control. Simulation experimental results show that the global integral sliding mode control for the command current has higher tracking accuracy and lower current total harmonic distortion of the grid side.
关键词:Recursive Least Squares algorithm;Shunt Active Power Filter;global integral sliding mode;Low Pass Filter