Abstract:

Abstract: In order to explore the influencing factors and their heterogenous impacts on vehicle groups accident under the multi-factor coupling effect, a random parameter logit model was developed based on 21 types of alternative factors in 4 aspects (road characteristics, driver characteristics, vehicle characteristics, environmental factors) from the road crash data of Texas in 2016. And the marginal effect formula was used to measure the impact of each factor on the level of collision damage. The model estimation results showed that 15 types of single factors such as driver gender, age, the use of seat belts, the release of airbags, and 4 coupling factors such as all seat belts-airbag initiation, elderly drivers-large trucks were significantly related to the degree of road vehicle accident injury; road speed limited >100 km/h, and the release of airbags had differences in the degree of injury from different accidents. The research results could have a certain guiding role in reducing the severity of highway multi-vehicle accidents.

Abstract: The accuracy and praclicability of life loss assessment can not be realized synergistically due to many factors affecting the life loss of dam break and relatively few historical statistics.A model suitable for the evaluation og life loss of dam brach is prosposd. Eight indexes,i.e.,severity of dam break flood,alarm time,occurrence time of dam break,risk population,understanding degree of dam break,proportion of young and middle-aged people,building quality and rescue capacity,are selected to build a dam break lif loss index system.The index grade is divided according to the relevant standards and criteria.AHP-BN method is used to comprehensively consider the influence of weight and probability,calculate the risk value of life loss,and build a dam break life loss assessment model.The application of the model to an example shows that risk values of life loss of the three dams are ranked in the model to an example shows that the risk values of life loss of the three dams are ranked in the order of Lijiazui Reservoir (0.626),Dongkoumiao Reservoir (0.594) and Shijiagou Reservoir (0.522).The evaluation results are consistent with the actual situation,which verifies the validity and practical value in engineering of this model.

Abstract: In strong electromagnetic interference environment, traditional acoustic magnetic EAS system has some shortcomes, such as short detection distance, high rate of false alarm and miss alarm. In order to solve these problems, an acoustic magnetic EAS system based on support vector machine (SVM) is proposed. The algorithm is implemented on FPGA chip. The fast Fourier transform (FFT) method is used to extract the frequency characteristics of tag and noise signals, which are used as the training samples of SVM algorithm. SVM algorithm classifies data frames into labeled and unlabeled ones, and the training part is implemented in MATLAB. The classification calculation part is implemented in FPGA, which adopts three-level parallel computing structure, and the algorithm time is less than 0.5 ms. The test results showed that: Under the interference environment, the longest detection distance of hard tag can reach 1.45 m, and the recognition rate can reach 99.4%. The longest detection distance of soft tag can reach 0.75 m, and the recognition rate can reach 98.9%. Compared with the traditional acoustic magnetic EAS system, this system can significantly increase the detection distance, improve the recognition rate, and provide a solution for the stable operation of EAS system in complex environment.

Abstract: To address the time delay and load balancing problems faced by multiple controllers deployed in the software definition network (SDN), in this paper, a multi-controller placement algorithm is proposed to reduce the time delay between controllers, and improve the network performance on the basis of load balancing. Aiming at the slow convergence speed of traditional particle swarm optimization algorithm, this paper proposes an improved particle swarm optimization algorithm to deploy the SDN controller. The improved particle swarm optimization algorithm is used to deploy the SDN controller to minimize the propagation delay between the switch and the controller while considering the load balance of the controller.The simulation results show that the improved particle swarm optimization algorithm for controller deployment can guarantee high load balancing performance and the better overall network performance by acquire fitness about 0.05. And compared with the traditional particle swarm optimization algorithm, the improved particle swarm optimization algorithm can improve the convergence speed of the whole network about 6.3% with lower time delay.

Abstract: The 3D human pose estimation is a challenging task in computer vision. Due to the difficulty of annotation, only some disperse key-point data form limited scenes are available, which makes 3D prediction a big challenge. In this paper, the human body is deemed as a flexible structure, but a specific limb can be viewed as a rigid-body. Given depths of two points on both ends, the depths of the whole limb can be estimated by dense interpretation. Therefore, this paper proposes a method that can take the dense depth interpretation feature map as middle supervision. It provides a denser and more structured target, instead of regression for disperse key-points directly. The MPJPG on Human3.6M reaches 50.9 mm with only a simple network structure. The cross-domain experiments on dataset MPI-INF-3DHP further show the generalization ability of the proposed method.

Abstract: Depth cameras such as TOF can obtain low-resolution depth images only, and cannot meet the needs of 3D vision systems. High-resolution depth images can be obtained by the super-resolution algorithm of depth images, but with problems such as artifacts in local areas of the image and unclear edge structures due to texture replication. Based on the HR depth pseudo-matrix, an algorithm of depth value reconstruction guided by adaptive parameters and edge points is proposed. In this paper, the pixels in the edge area of the low-resolution depth image are used to find the pixels with the wrong depth value in the correction image: and clustering is performed to obtain the reconstructed depth image. By introducing an autoregressive model guided by adaptive parameters, the depth value of pixels in the edge area is predicted. The experimental results prove that the algorithm can effectively reduce the ambiguity of the depth discontinuous area, and obtain higher quality high-resolution depth images. Compared with the existing algorithms, when the upsampling factors are 2, 4 and 8, the average dead pixel rate of the output results of this algorithm is lower than 0.1, which can effectively verify the advantages of the algorithm in this paper.

Abstract: Path planning is one of the key technologies of UAV autonomous flight. The typical path planning can be divided into three steps: firstly, the preliminary planning of flight path should be carried out by fully considering various threat environments; secondly, the optimal path should be found by using the optimization search algorithm; finally, the path should be smoothed. This paper systematically summarizes the studies of UAV path planning in recent years; analyzes the flight airspace, dynamic constraints and environmental constraints in the process of path planning; expounds the key technologies involved in path planning, including terrain acquisition, threat and cost modeling, path planning algorithm and path smoothing; and further analyzes and summarizes the common path planning algorithms, such as A* search algorithm, genetic algorithm, ant colony algorithm, particle swarm optimization algorithm, and the common path smoothing algorithm B-spline curve method; and summarizes the problems of the current UAV path planning model construction and path planning search algorithm. Finally, some potential future development trends of UAV path planning are proposed, including the construction of reasonable path planning system, the study of advanced online path planning algorithm, and the cooperative path planning of multi-UAV.

Abstract: In traditional image-based waterline identification method, a fixed camera is used to capture the waterline area, which requires complex equipments and has difficulty to measure the waterline of the outer chord of ships. This will decrease the identification efficiency. Based on this, a dynamic identification method of waterline area based on aerial image captured by UAV is proposed. Though UAV is more convenient than the fixed cameras, the waterline area in the images may be unsettled due to the unstable flight track. In addition, water trace, ripples in the surface of water and different exposure environment all have the adverse impact on the recognition of waterline. Thus, a statistic-based selection for contour combined with the prior knowledge of waterline characteristics is adopted to locate the interesting horizontal areas containing the waterline. After comprehensively utilizing the color and spatial information of pixels, the waterline is accurately segmented by knowledge-guided K-means++ and watershed algorithm under L×a×b color space. The experimental results show that the proposed method can effectively process the influence from light conditions, water traces and so on, as well as rapidly identify the waterline area from a dynamic video. Its strong robustness for the complex environment in ports provides the foundation for obtaining the accurate weigh of cargoes.

Abstract: Moth-flame optimization algorithm (MFO) has some drawbacks in solving optimization problems, such as low precision and high possibility of being trapped in local optimum. A modified MFO algorithm based on chaotic initialization and Gaussian mutation is proposed. Firstly, the cube chaotic map is used to initialize the moth population, which makes the moth more evenly distributed in the search space. Then, Gaussian mutation is adopted to disturb a few poor individuals to enhance the ability of escaping the local optimum. Finally, Archimedes curve is introduced to expand the search scope and strength the exploration ability in the unknown field. A series of experiments are carried out on CEC14 test function set and 21 extensible Benchmark functions. Compared with standard moth-flame optimization algorithm, genetic algorithm, artificial bee colony algorithm, particle swarm algorithm, differential evolution algorithm, flower pollination algorithm, and butterfly optimization algorithm, the results demonstrate that the proposed algorithm is strengthened in obtaining solutions with better quality and convergence.

Abstract: In order to improve the super-resolution reconstruction quality of single image, an improved learning based super-resolution approach was proposed in this paper. To tackle the problem of low details of semi-coupled dictionary learning super-resolution algorithm, the paper presented learning strategy where detail constraint factor and semi-coupled dictionary learning were performed in turn. In reconstruction stage, detail constraint factor was designed by the gradient in both horizontal and vertical direction. Combined with semi-coupled dictionary learning, detail constraint factor was used to further improve the super-resolution reconstruction quality. In order to improve the contribution of detail constraint factor on preserving boundary information, the adaptive regular parameter was explored via the approximate Laplacian distribution of edge difference. Compared with the semi coupled dictionary learning super-resolution algorithm, the peak signal-to-noise ratio of this method was increased by 1.5% on average. Experiments demonstrated that the proposed method could achieve better reconstruction effect in both subjective and objective evaluation and improve the quality of super-resolution.

Abstract: In order to improve the performance of alkyd resin in water resistance, drying and hardness, acrylic prepolymer was used to modify the alkyd resin, and the emulsion was prepared by using emulsion phase inversion method. The synthesis process was investigated. The results showed that the proper resin synthesis condition was that the amount of initiator was 2%; the amount of chain terminator was 1.5%; mixed monomer dropping time was 4 h; prepolymer holding time was 3 h; esterification reaction temperature was 210 ℃; and esterification time was 150 min. The properties of propanol alcohol emulsion prepared by suitable emulsification process conditions were improved, such as shortened drying time, enhanced hardness and reduced particle size.

Abstract: High carbon high chromium steel had high hardness, but there were network eutectic carbides in the steel, which made the steel easy to break when used. In order to improve the structure and mechanical properties of GX160CrMoV12 steel, OM, SEM observation, hardness test, impact toughness test and XRD test were used to study the effect of quenching and tempering temperature on the structure and mechanical properties of GX160CrMoV12 steel. The results showed that the structure of GX160CrMoV12 steel after quenching was martensite + residual austenite + carbide. As the quenching temperature increased, the eutectic carbide gradually dissolved, the size of martensite became larger, the content of residual austenite increased, and the hardness increased first and then decreased. The optimal quenching temperature was 1 040 ℃; the matrix of GX160CrMoV12-Ti-Ni-Y steel was tempered martensite after tempering. As the tempering temperature increased, the hardness decreased and the impact toughness increased. The optimal tempering temperature was 200 ℃.

Abstract: In order to explore the influence of shot peening angle and shot peening pressure on the surface integrity of DC53 steel, the effects of different shot peening angles and shot pressures on the residual stress distribution and hardness distribution on the surface of DC53 steel are studied by experiment, and the morphology after shot peening and the effect of shot peening on roughness are analyzed. The results show that after shot peening, DC53 steel introduces a residual compressive stress layer with a certain depth. The stress shows a “U”-shaped distribution with depth increasing. When the shot peening pressure is 0.5 MPa, and the shot peening angle is 90°, the maximum stress tends to stabilize about 1 075 MPa, and the stress layer depth is about 410 μm. The surface hardness value and hardened layer depth of DC53 steel after shot peening are obviously increased. When the shot peening angle is 90° and the shot peening pressure is 0.55 MPa, the surface hardness increases from 58.3 HRC to 62.4 HRC,and a hardened layer of about 330 μm is formed. The surface morphology changes significantly after shot peening, compared with the original specimens, the surface roughness value decreases first and then increases with the increasing of the shot peening angle and shot peening pressure. The research results can provide a reference for the reasonable selection of shot peening parameters of DC53 steel.

Abstract: In order to improve the structural stability of the chip without affecting the function of the chip, the surface of the chip material COC (cyclic olefin copolymer) was studied. The highly hydrophilic surface of COC material after air plasma treatment has obvious timeliness. By adjusting parameters such as plasma cleaning power, time and placement temperature, the timeliness after hydrophilic treatment could be amplified and the recovery of COC hydrophobicity could be realized. The experimental results showed that: using this method, the contact angle of COC surface changed from 90° to 25°±2.1° and then to 83°±1.5°. The bonding strength increased from instability to 0.051 MPa. X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR) showed that the treatment process only modified the surface and did not affect the deep structure of the material itself. The main reason for the improvement of bonding strength was the introduction of oxygen-containing functional groups such as hydroxyl, carboxyl and carbonyl. Finally, the feasibility of this method in microfluidic devices was verified by droplet generation experiments.

Abstract: To study the effect of high temperature on rock, siliceous sandstone was selected to study the physical and mechanical properties after high temperature treatment, and the high temperatures ranged from 200 ℃ to 800 ℃. In addition, samples at room temperature (25 ℃) were set as the control group. The effects of high temperature on rock were evaluated by mineral composition analysis, longitudinal wave velocity, uniaxial compressive strength change and failure morphology. It was found that high temperature had a significant effect on the rock composition, which led to changes in rock wave velocity, strength and failure mode. The strength decreased with the temperature increased, and a turning point occured at 400 ℃. The longitudinal wave velocity and peak strain at the rock failure also changed significantly at 400 ℃. To further study the influence of high temperature on the rock, the thermal damage factor DT was introduced to establish the relationship of thermal damage and the rock strength. There is a three order polynomial function relationship between DT and high temperature.

Abstract: Wind load of the long span roof was numerically investigated by adopting CFD simulation method. The present numerical method was firstly verified by comparing the simulated base force of the section roof model with the force balance wind tunnel experimental data. Effects of the PV panel density and the supporting buildings on the wind load distribution on the roof were then investigated. The wind load distribution mechanism was analyzed based on the simulated flow field. Finally , the base force and the wind pressure distribution on the whole roof were presented and compared with those of the old scheme roof. The CFD simulation method was verified to be capable in predicting the mean wind load on the long span hollow grid roof structure. Wind effect on the roof was dominated by the horizontal wind load. The vertical wind load was mainly wind suction which was favorable for the structure. Inhomogeneous distribution of local wind suction was observed on the roof ，especially at the bottom surface of the roof. Large wind suction was easy to cause local wind-induced damage in those zones. In comparison with the PV panel density , the supporting buildings could have more remarkable influence on both the base wind forces and local wind pressure distribution on the roof structure .The existence of the supporting buildings could not only increase the vertical wind suction on the whole roof body, but also could enlarge the local wind suction on some part of the bottom of the roof.

Abstract: In order to realize the resource utilization of sewage sludge ash, the effect of desulphurization gypsum on the strength and hydration products of the sewage sludge ash cementitious system at different stages was studied by means of paste strength measurement, X-ray diffraction, infrared spectroscopy, thermogravimetric analysis and scanning electron microscopy. The result showed that the hydration products of the sewage sludge ash cementitious system were mainly ettringite, portlandite and hydrated silicon aluminum (iron) calcium gel. Desulphurization gypsum promoted the early hydration of calcium silicate in the cementitious system, generated more hydration products, and improved the degree of polymerization of the hydrated gel and strength of the sample; at the later stage of hydration, desulphurization gypsum accelerated the alkali-induced reaction of cementitious system, promoting the formation of more gel phase. The microstructure of the cementing system was mainly composed of granular products with low development degree, and there were many holes in the structure. After adding desulfurization gypsum, a large number of plate-like calcium hydroxide and short rod ettringite crystals were produced, interlacing and filling in the interlayer space of the gel, which improved the compactness of microstructures and macroscopic strength.

Abstract: Studies were initiated for the shear lag effect of the wide single box bridge with multi-cell during cantilever construction and its evolution in the construction. A single-box with five cells four-span continuous PC bridge was taken as an example, and the whole construction process was simulated numerically. Analyses were focused on the normal stresses in the roof and bottom plate of several critical sections, caused by concrete pouring, prestressing, closure and system transformation singly. The results showed that the transverse distributions of normal stresses could be regarded as the superposition of “global” shear lag effect of the whole section and “local” shear lag effect of each web, and the former origins from the shear stress difference in webs. Moreover, the proportions of the two effects were related to the cell number, cantilever length and load type. The global effect increased with the cell number. The shear lag effect caused by the concrete pouring and prestressing in the root section was remarkable in the roof but not the bottom plate, and the global effect increased with the cantilever length. Additionally, the global effect caused by the concrete pouring was more evident than that caused by prestressing.