It is difficult to directly obtain the longitudinal vehicle speed of commercial vehicle in the anti-lock braking system (ABS). In order to estimate the longitudinal speed in the braking process, a strong tracking-cubature Kalman filter algorithm was proposed. Then, according to the ABS control requirements, a commercial vehicle ABS neural network sliding control algorithm was established. It uses the sliding mode algorithm to control the slip rate of ABS, and uses the neural network to adjust the parameters of the sliding control algorithm. Through Matlab/Simulink and TruckSim co-simulation, simulations tests were conducted on high, medium, low, and off-road surfaces, respectively. The simulation results show that the strong tracking cubature Kalman filter algorithm is accurate in estimating the longitudinal speed, and the neural network sliding control algorithm of ABS has good effect.

This paper proposes a policy learning approach for autonomous driving based on the DRL and multi?type sensor data. Different Convolutional Neural networks (CNNs) are employed to deal with the data from different sources (i.e., high?dimensional data from camera and low?dimensional data from lidar, GPS, etc.). Then the extracted features from CNNs are combined for training the autonomous driving policy. Finally, the TORCS, which is an open?source simulation platform, is chosen to validate the proposed method. The results demonstrate that the multi?type sensor based DRL model can get good performance on velocity and lateral error control.

A new solver is implemented based on immersed boundary method (IBM) and pressure implicit with splitting of operators (PISO) algorithm. By this IBM-PISO solver, the external flow of Ahmed body is simulated using the delayed detached-eddy (DDES) turbulence model. First, the time-averaged features and shedding frequency of 2D cylinder cases are used for the verification of the IBM module and IBM-PISO solver. Then the instantaneous flow field, time-averaged features, and vortical flow structures of Ahmed body are investigated. The effect of slant angle and the transport of vortical structures are compared. The results show that the IBM-PISO solver produces accurate results at separation zone and main vortical flow structures including recirculation vortices, horse-shoe vortices at the upper surface and at the foot of the front surface, C-pillar vortices, and lateral vortices. The transport in the wake enhances the turbulence of C-pillar vortices and renders their domination over the far-wake flow field. The effect of slant angle is based on controlling the flow separation at the slant and confined to the near-wake region.

The prediction capability of steady numerical simulation on aerodynamic forces and surface pressure of generic vehicles was clarified. First, a detailed study on mesh scheme for MIRA reference car model was carried out with baseline data derived from full-scale wind tunnel test. Results show that the grid-independent strategy requires 10 mm for car surface mesh, Y ⁺=30 for the first layer grid and 5 layers around the car. Based on this mesh scheme, aerodynamic force calculation of MIRA model could yield satisfactory predictions, that at 0° yaw, the error of drag coefficient C _D from test data was 0.34%, error of lift coefficient C _L 1.06%, and at 3°~20° yaw conditions, error of C _D and side force coefficient C _S were respectively no more than 5% and 9%, which indicate the reliability of aerodynamic forces computation for simplified cars like MIRA model. For surface pressure, of all 254 pressure taps all over the car, error of C _P over 50% are less than 20%, while surface pressure was poorly calculated mostly where flow is so complicated that separation or reattachment tends to happen, such as the underbody and the leeward side..

The pressure pulse caused by high?speed trains passing each other produces aerodynamic impact on the windbreak, that may result in structural instability of the windbreak and bring security risks to the train. Taking CRH3 as the research object, the pressure pulse on windbreak impacting by different wind speed and different train speed were studied by simulation. In the simulation, the overset mesh method, which can simulate the real train movement, and the chamber energy dissipation windbreak, which retains the hole characteristics, are used. The results show that the pressure pulse around high?speed train presents “positive?negative?negative?positive” trend. The pressure pulses of two trains couple with each other and act on the windbreak on the side of the line. With the train speed increasing, the pressure extremum increases, the commutation time reduces and the rate of change increases. With the wind speed increasing, the cross?wind coupling with the train wind amplifies the negative pressure inside the windbreak. The positive pressure extremum is outside the length of the train and the negative pressure extremum is inside the length of the train. The position of the positive pressure extremum is nearest to the nose cone of the head train. The pressure extremum will move backward under cross?wind.

To reveal the degradation rules of adhesively bonded Carbon Fiber Reinforced Plastic (CFRP)/aluminum alloy subjected to high temperature environment, adhesive joints under shear, normal and combined shear and normal stress states were manufactured, and degraded at 80 ℃ (high temperature) for 10, 20, 30, 40, 50 days. The failure strength and failure mode were analyzed, and a response surface of failure criterion was built for the failure prediction. Results show that with increasing proportion of normal stress, the failure strength decreases more obviously and the failure mode changes from cohesive failure to mixed failure of fiber tear, interface and cohesive, which is mainly caused by the degradation of CFRP. The average relative errors of the response surface of failure criterion is about 3.0%, which can be used to predict the failure of adhesive joints under different stress states. Therefore, it is necessary to reduce the proportion of normal stress in adhesively bonded CFRP/aluminum alloy joints to improve the load capacity. Besides, the effect of fiber tear has to be taken into consideration in the failure prediction of CFRP/aluminum alloy joints.

In order to improve the overcharge safety of power battery as well as clarify the occurrence mechanism, the behavior characteristics and the propagation law of overcharge thermal runaway, the experiment on overcharge thermal runaway of NCM battery and battery pack are carried out. The results show that there is an obvious evolution process of overcharge thermal runaway. There is a time interval between the sharp shutdown of battery voltage and the exponential rise of battery temperature. The maximum temperature that each battery reached during thermal runaway process is elevated increasingly along with the thermal runaway propagation. Furthermore, the module structure to some extent can restrain the occurrence of overcharge thermal runaway. Finally, the suggestions for improving the overcharge safety of NCM battery and battery pack are proposed based on the above results.

Aiming at the problems that the pressure of wheels apply on Anti-force Roller Brake Tester is over large or over small, or even completely overhead for the test of multi-axle tractor braking performance, a detecting system for automatic adjustment of axle load was designed. This system was based on the Solidworks software designed for the axle load self-adjusting lifting device. Taking the four axles truck for example, a vehicle-tester multibody dynamic simulation model was developed in Adams virtual prototype technology. Through the simulation, the axis load-height changing curve was obtained in turn. Combined with the established model of axle load distribution for multi-axle vehicles, the height of the non-detection axis was automatically adjusted to realize the purpose of changing the axle load of the testing axle. To verify the accuracy of the simulation model, a real truck comparison test was carried out. At the same time, in order to verify the feasibility of the detection system, tests of five types of trucks with different curb weight was carried out. The results indicate that the maximum error between the established model and vehicle test is 4.47%, and the relative errors are less than 5% by five types of trucks, meeting the practical test requirement.

In order to overcome the defect of fixed safety distance in Optimal Velocity (OV) model and its related derivative generalized force (GF) model and full velocity difference (FVD) model, a car-following model with improving safety distance based on the OV model is proposed in this paper. The improvements of the proposed model are analyzed theoretically. Then, according to the proposed model, the simulation frameworks are developed to analyze the impact of three model parameters on traffic flow stability. These parameters are the density of traffic flow on the road, the maximum velocity for driving vehicle and the time of driver's reaction delay. The simulation can truly reproduce the propagation state of traffic flow. Finally, using the data from U.S. NGSIM Database as samples, the accuracy and effectiveness of the model presented in this paper are verified after testing with cluster analysis by comparing with OV model, GF model and FVD model.

Under the premise of vehicle driving safety, an optimal method of collaborative control for vehicles in Cooperative Vehicle Infrastructure System (CVIS) environment is studied with the maximum speed as the target. The control strategies are simulation experiments under two scenarios, consecutive turning or platoon turning. The results show that the proposed method can realize the coordinated control of the vehicles in the U?turn area. Finally, the control efficiency is evaluated by the total delay time and driving comfort indexes, and the thresholds relative to the two control strategies are obtained.

In order to explore the causes and influencing factors of risky driving behavior of electric two-wheeled vehicles, a questionnaire survey was conducted among 183 electric two-wheeled cyclists. After exploratory factor analysis (EFA), a 6-factor structural model was established. These factors include risk driving behavior, emotion State, altruism, risk perception, driving self-confidence and risk awareness. The relationship between statistical information, the 6 factors and traffic accidents were studied to establish a traffic accident prediction model based on Logistic regression. This model was validated by the Factor Analysis (CFA). Research shows that daily mileage, risk driving behavior, working hours, risk awareness are important parameters affecting traffic accidents.

To investigate the flexural behavior of continuous steel-concrete composite beams strengthened with prestressed carbon fiber-reinforced polymer (CFRP) plates in negative moment region, three 6.2 m two-span continuous beams were tested to failure. The specimens included two strengthened beams and one unstrengthened control beam, which had steel box-concrete composite section. The CFRP plates were prestressed to 20% and 35% of the tensile strength respectively by an innovative fabricated anchorage system and installed over the top surface of the concrete slabs. Experimental results show that the ultimate flexural capacity of the central support sections increased 19.4% and 28.5% respectively by the use of prestressed CFRP plates. The crack resistance in the negative moment region was enhanced significantly. The strengthening contributes to the decrease of the mid-span deflection and crack width under service condition, while improving the ultimate deformation at failure stage. The strain distribution along the cross-section conforms to plane-section assumption. The calculation method of the ultimate flexural capacity for central support section is proposed. Good agreement between experimental and theoretical results is achieved.

A damage model for accurate evaluation of the cross section damage of reinforced concrete pier under dynamic load is established. First, the damage criterion of plastic strain energy density is deduced. Then, the performance level of reinforced concrete pier is established based on the condition of crack propagation. The correspondence relationship between damage index and performance level is determined through analyzing the dynamic damage of 33 quasi-static test model of reinforced concrete pier by the software of ABAQUS, and the evaluation model of the cross section damage of reinforced concrete pier under dynamic load is quantified. Finally, the Support Vector Machine algorithm is used to predict the normalized parameters of the plastic strain energy density damage criterion. The accuracy and applicability of the plastic strain energy density damage criterion and the Support Vector Machine algorithm which were used to analyze the cross section damage of the reinforced concrete pier are verified by numerical examples.

Twenty?four reinforced concrete beams were tested to shear failure. The mechanism of shear strengthening was investigated. The influence of the beam design parameters, Loading level of original beam, damage degree of original beam and arrangement of wire rope，on shear strength was investigated. The results show that ropes can effectively delay shear crack development, limit diagonal crack growth, and increase the ultimate shear capacity. The action mechanism of steel wire rope to increase shear capacity is similar to that of web reinforcement in reinforced concrete beams. The design parameters of the original beam have obvious influence on the effect of shear reinforcement; with the increase of loading level, the degree of improvement of shear ultimate bearing capacity gradually decreases; the damage degree of the original beam has some influence on the effect of shear reinforcement. When the crack of the original beam is less than 0.2 mm, the effect of glue injection is not ideal and the increase is small. When the width is more than 0.2 mm, the effect of glue injection is improved and the bearing capacity is improved. However, when the crack width is greater than 0.5 mm, the reinforcement effect is in a stable stage; the U?shaped beam is better than the oblique beam; and the U?shaped beam is better than the U?shaped closed beam.

The three?point bending test was carried out on concrete notched beams with different amount of coarse and fine polypropylene(PP) fibers. Based on the double?K fracture theory, the influence of mixing method of PP fibers with different sizes on the fracture toughness and failure mechanism were discussed.The real cohesive toughness, the increment of initiation toughness and the bridge connection toughness were gained by the quantitative relationship between the initiation toughness, the unstable toughness and the cohesive toughness. In addition, The theoretical cohesive toughness was calculated from three kinds of ordinary concrete bilinear softening constitutive curves, which was compared with the real cohesive toughness to determine the bilinear softening traction?separation law of PFRC, which was suitable for different PP fiber incorporation conditions. The results show that PFRC has higher initiation toughness, unstable toughness and fracture energy compared with plain concrete. The enhancement effect of bridge connection toughness from bridging stress is very significant when 2 or 3 sizes of PP fibers are mixed. The 3 sizes of PP fibers play the role of bridging at different stages of crack propagation, and it serves a well function of toughening and strengthening. The bilinear softening traction?separation law proposed by Xu and Reinhardt is well applicable for PFRC when the correction coefficient is 6.

A new type of reinforced Engineered Cementitious Composites (ECC) beam was proposed with ultra high strength bars, named ultra high strength rebar reinforced ECC beam (UHSRRE). The tensile bearing capacity of ultra high strength reinforcement was expected to be efficiently utilized until the requirements of serviceability limit state were not met in UHSRRE. Theoretical analysis on the flexural behavior of UHSRRE was conducted. Three kinds of moments at different phases, including cracking moment, yielding moment and ultimate moment, were estimated with three fundamental hypotheses to calculate the flexural capacity. These hypotheses were the constitutive relationships of ECC and ultra high strength rebar, plane section assumption of mean strain, and the tensile stress of ECC in the tensile area of beam. In addition, the requirements of boundary and minimum reinforcement ratio were defined. These theoretical models were verified by the experimental results .

The influence of carbonization on the distribution of nitrite ions in hardened cement paste is verified by electron probe microanalysis (EPMA) and X-ray diffraction (XRD), and the migration and adsorption of free nitrite ions in cement paste is clarified, which is important to accurately evaluates the corrosion-resistant effect of nitrite ion on reinforced concrete. The results reveal that a new hydrated crystallization phase NO₂?AFm is produced after the hydration of nitrite-containing cement paste and is evenly distributed. During the carbonation process, NO₂?AFm re-decomposes and generates nitrite ions, diffusing to non-carbonized area. This results in a decrease of NO₂?content in carbonized area and an increase in non-carbonized area. Nevertheless, due to the decomposition of C?S?H gel during the carbonization process, the free nitrite ions adsorbed on its surface are re?transformed into free nitrite ion, increasing the content of free nitrite in the carbonization area. Meanwhile, the migration of silicon from carbonized area into non?carbonized area, contributes to a higher amorphous content of C?S?H gel in non-carbonized area, which increases the solidified effect of nitrite ion, and eventually leads to a higher nitrite ion content in carbonized area than that in non-carbonized area.

In order to solve the problems of low user satisfaction, poor rationality and livability, a dynamic design method of building livability based on multi source spatiotemporal data is proposed to improve the existing design method of livability design. The key points of the architectural road traffic landscape design are divided into the construction of road visual modeling, the construction of the roadbed landscape and the construction of the separation zone. Based on the results of the road traffic and landscape analysis, the related multi-source spatiotemporal data are obtained, and the relationship between the length and the thickness of the building floor is fitted through the power relation, and the floor board is obtained. The relationship between the thickness, the length of the short side of the floor and the ratio between the length and width of the floor, so as to get the design result of the minimum plate thickness at the building's livability, and complete the dynamic design of the vibration comfort of the building. According to the reliability theory, the functional function of building structure livability is defined, and the result of the function calculation is combined with the correlation criterion of the acceleration limit value. In order to guide the livability design of the building, the dynamic construction of the livability of the building is guided by the different fitness of the acceleration corresponding to the acceleration of the human body. The experimental results show that the average user satisfaction is about 95.8%, and the livability coefficient of the building is 0.9.

To achieve a low?cost, real?time, non?invasive, high?accuracy soft cylinder pressure sensor, we propose a novel pressure identification method involving frequency?amplitude modulated Fourier series and an extended?Kalman?filter?optimized engine model, which is based on theories regarding explosion and burning together with Newton′s laws. The extended?Kalman?filter is used to estimate the optimal output of this iteration model. The optimal output of the speed iteration model is utilized to separately compute the frequency and amplitude of the cylinder pressure cycle?by?cycle. A standard engine working cycle, identified by the 24th order Fourier series, is determined. Using the frequency and the amplitude to modulate the Fourier series yields a complete pressure model. A commercial engine (EA211) provided by the China FAW Group corporate R&D center is used to verify the method. Test results show that this novel method possesses high accuracy and real?time capability, with an error percentage for speed below 10% and the percentage corresponding to power of pressure below 2% when Air?Fuel Ratio coefficient is setup at 0.85. Error percentage for speed below 1.5% and the percentage corresponding to power of pressure below 1.5% when Air?Fuel Ratio coefficient is setup at 0.95.Thus, the accuracy and feasibility of the proposed method are verified.

Based on the special structure of internal/external meshing gear motor, the main ways of leakage were analyzed. Through the calculation formula of leakage, four different connection modes were analyzed. Meanwhile, the test bed was built to measure the volume efficiency of the motor. The results show that the leakage of the end gap of the gear motor is independent of the motor speed, and it is only related to the pressure difference of the motor inlet and outlet. With gear motor connected in different ways, the radial clearance leakage has a tiny change. As the motor inlet and outlet pressure difference becomes bigger, the internal leakage of the motor gradually increases and the motor volume efficiency decreases. The mechanical efficiency increases with the increase of the pressure of the motor inlet and outlet. The conclusion is significant for the design, manufacture and use of the internal / external meshing gear motor.

The continuous meshing zone and "double conjugate" phenomenon of the circular spline and flexspline under no?load conditions can effectively improve the meshing performance of the double?arc harmonic gear device. For this reason, a multi?objective optimization design method of the tooth profile parameters of the double?circle flexspline is proposed. For the non?public tangential double?arc tooth profile, the conjugate tooth profile was designed by using the relationship between the improved kinematics method and the precise rotation angle. To reduce the difference between the conjugate blank area and the conjugate tooth profile as the optimization goal, a harmonic drive multi?objective optimization model was established. Considering the existence conditions of conjugate tooth profiles, single target analysis and multi?objective optimization are performed. The results show that the two objective functions increase with the increase of the radial deformation coefficient and the double circular arc inclination, and decrease with the increase of the tooth thickness ratio; Radial deformation coefficient and double circular arc inclination have the greatest influence on the two objective functions, and tooth thickness ratio has less influence on the difference of minimum conjugate tooth profile. Convex arc radius coefficient and concave arc radius coefficient have almost no effect on the minimum blank conjugate area; this optimization method can achieve a reasonable selection of parameters of the soft tooth profile, and make the freewheeling state between the circular spline and the flexspline is less than 1.2°, and the difference between the conjugate profiles is about 0.83 μm, which can meet the requirements of engineering applications. Guaranteed no?load continuous meshing interval and "double conjugate" transmission accuracy.

For clarifying the heat generation mechanism of hydrostatic bearing oil film accurately, a calculation method of the oil film temperature is proposed, which considered the influence of hot-oil-carrying(HOC) . The factor of HOC was defined and deduced, and the occurring conditions was determined theoretically when the phenomenon of HOC appears. The temperature rise equation of lube feeding under the influence was established. The finite volume method was used to simulate the temperature field of the film. The temperature rise curves of oil film under different working conditions was obtained, which were in good agreement with the experimental results. The results show that, for this structure bearings, the phenomenon doesn't appear under any load condition when the rotating speed is less than 10 r/min. When the rotating speed is greater than 60r / min, the phenomenon occurs in the entire load range, from 0 t to 32 t. The temperature rise of the film is aggravated, while the phonmenon of HOC occurs, and the rotation speed affects the temperature rise more than the load does.

Accurate modeling of the contact stiffness of bolted joints is therefore crucial to predict the dynamic performance of CNC machine tools. This paper presents a contact stiffness model of a bolted joint based on multi-scale theory. The model uses a series of stacked three-dimensional sine waves to describe the multiple scales of roughness of the contact surface, and each frequency level is considered a layer of asperities, stacked iteratively on top of each other. The relationship between the contact area ratio and frequency level can be deduced. Moreover, the contact stiffness at each frequency level can be regarded as a spring in series in the model, therefore, the total stiffness can be obtained by summing the contact stiffness at each frequency level. The influences of contact load, material characteristic parameters and multi-scale parameters on the contact stiffness model are analyzed by mathematical simulation. An experimental setup consisting of the section beam specimen was used to validate the numerical model of the bolted joint for the case of equal pre-tightening forces. The relative error between the multi-scale natural frequencies and experimental frequencies was less than 9.94%, suggesting the multi-scale model can be effectively used in predicting the dynamic characteristics of CNC machine tools.

A tool path generation method for active control machining errors is proposed taking into consideration that, in optical freeform surface slow tool servo, the location and quantity of cutting location points selection are lack of basis. The cutting contact points and cutting location points are analyzed using this method. The experiment of slow tool servo for sine?wave surface was carried out. The results show that the proposed tool path generation method has the characteristics of predicting machining surface accuracy and significantly reducing the number of cutting location points.

During cleaning process of rice combine, the rice threshed mixtures could adhere to or block the sieve surface. In order to study the interface adhesion law of rice threshed mixtures under different vibration modes, three modes of vibration test bench were designed, including linear vibration, circular arc vibration and jitter. The adhesive mass on working face under three vibration modes kept stable when the vibration time exceeded 15 s. When the vibration amplitude was increased from 10 mm to 30 mm, the adhesive mass on the working face under three vibration modes was effectively reduced. As the vibration frequency was increased from 2 Hz to 3.5 Hz, the adhesive mass on the working face under three vibration modes was quickly reduced. The acceleration peak value of linear vibration working face was obviously higher than that of circular arc vibration and jitter. The linear vibration working face has the best anti-adhesion effect. The influences of test factors on the anti-adhesion capabilities of linear vibration working face in descending order are amplitude, frequency and vibration time. The influence of test factors on the anti-adhesion capabilities of circular arc vibration and jitter working face in descending order are frequency, amplitude and vibration time.

Wheeled farm tractors are widely used in farming and accidents involving tractor rollover remain an ongoing problem. Investigating the significance variation of factorial effects on tractor stability on various terrains provides a theoretical reference for designing tractors with higher stability and provides operators criteria for avoiding rollover. In this paper, adopting the scale?model?based experimental approach, we simulated the progress that the tractor running along the side slope in random road surfaces. In order to interpret the rollover trend, we developed an evaluation index, called the load loss rate (LLR), which characterizes the strength of contact between the ground and tractor uphill wheels. The results showed that the reasonable increase in the front weight has a significant effect on improving the tractor stability in all kinds of road conditions. However, for medium uneven road surfaces such as farmland and rotary farm, the large diameter and low?pressure tire and the wide rear wheelbase will be more effective. Meanwhile, the optimization of the tractor configuration has been significantly impaired in higher uneven terrain such as plowed fields. In addition, the interaction of the configurational factors has no significant effect on the overall machine stability under all road conditions. The results of this paper may provide a theoretical reference for the targeted optimization of tractor configuration, which can help to improve the stability of tractors in complex road conditions.

Based on the theory of crescent?tail propulsion, a new type of bionic multi?link fishbone that conforms to the body and caudal fin (BCF) propulsion model is designed. First, the main design parameters, which influence the fish's propulsion are determined, including the swing angle of the body, the swing angle of caudal vertebra and the swing angle of tail. The skipjack, which has smaller body size among Tuna bream, is taken as the prototype of multi?link fishbone. Second, the fitting function of multi-link fishbone is obtained through reverse engineering technology. Third, the fitting function is used as the boundary condition of the design of fishbone structure. Finally, combined with the design parameters, the six?part connecting rod system, which includes the caudal vertebra, the caudal fin, the torso drive and the caudal vertebra, is designed by the inverse kinematics principle of the body. The feasibility of the mechanism is simulated by ADAMS software. The prototype experiment is carried out and the data obtained are analyzed. Experimental results show that the structure of multi?link fishbone can carry out the fish propulsion.

The stamping process of handbrake fixed plate on the automobile was simulated using the finite element simulation software AutoForm, and the influence mechanism of process parameters on the drawing forming was investigated. Orthogonal experiments were used to optimize parameters and the whole process was simulated. The simulation results indicate that the quality of forming parts is improved by optimizing the process parameters on the basis of the rational design of the drawbead. The optimum condition of process parameters are blankholder force 600 kN, die radius 6 mm, die gap 1.50 mm, friction coefficient 0.15, drawbead coefficient 0.5. The correctness of the simulation results is proved by practical experiments, which means that simulation software Autoform can be applied to guide the actual production.

In order to understand the effect of inner damage evolution on the formability of sheet metal, a prediction method of forming limit coupled with GTN damage model was developed based on the M-K groove instability theory. The damage-related parameters of high strength steel 22MnB5 were calibrated accordingly. The critical void volume fraction was chosen as the instability criterion, which improves the physical significance of the prediction method. And the effects of loading path on the instability of sheet and damage evolution were also studied based on the theoretical calculation results. Meanwhile, the NAKAZIMA test was conducted, and the forming limit data under different loading path was obtained. The deformation characteristics of 22MnB5 sheet were also analyzed. By comparing with the experimental results, the validity of the proposed method was verified.

According to the structure of the stretch bending equipment and the loading condition, the loading capacity of the jaw is equated to the tensile strength of the drawing cylinder and rotation angle of the turning arm, and a trajectory design method for profile stretch bending based on incremental control is proposed. By dividing the target shape into several forming sections and calculating the tensile strength increment and rotation angle increment, the independent distribution and control of the loading capacity are realized. The Z-shaped stainless steel curved roof beam is selected as the research object. The profile stretch bending FE simulation model composed of connector units is established. The calculated loading increment is applied in predicting the forming effects of the workpiece, which realizes the parameterized regulation of the loading trajectory. The forming defect analysis and pertinent modification are conducted according to the shape error and the surface warpage in the simulation results, which improve the forming effects. The modified trajectory is adopted in the forming test, and the simulation results meet well with the experimental results.

The purpose of this study is to analyze the effect of shot peening on the surface finish of the workpiece, in which the AA6061 workpiece is taken as example. Firstly, a FEM model of shot peening is established, based on which the surface morphology of single shot peening and multiple shot peening are simulated. Corresponding experiments of single shot peening and multiple shot peening are carried out and the practical surface morphology of the workpiece after the shot peening is observed by the Laser Scanning Confocal Microscope, by which the validity of the established FEM model of shot peening is verified. Then, 9 groups of orthogonal simulations are executed to study the influence of the initial roughness of target material, shot size, shot speed and shot peening coverage on the final surface roughness of the workpiece described by the 3D roughness parameter S _q. The prediction model of the surface roughness of the workpiece after the shot peening is established by means of the BP neural network model. Based on this prediction model the predicted surface roughness and the simulated surface roughness are compared and the mean error is 3.46%. This prediction accuracy can meet the requirement of the shot peening production. Furthermore, the controlling map of AA6061 workpiece after shot peening is constructed, which is important to control the final surface finish in the practical shot peening processing. The methodology presented in this paper can be extended to the application of other metallic material besides aluminum alloy, which would be helpful for the simulation, prediction and controlling the surface roughness of the workpiece after the shot peening processing.

The yaw velocity and the side angle of the center of mass play important roles as the main parameters in the stability of the vehicle. In this paper, a layered structure is proposed to control the stability performance of the vehicle by using the sliding mode variable structure strategy for the upper system. A two degree of freedom vehicle model is used to calculate the ideal value of the control variable. The vehicle stability control system is designed according to the yaw velocity and the deviation and deviation derivative of the sideslip angle. The optimal additional yaw moment is calculated by the control parameters. The stability of the closed loop system is judged by Lyapunov decision method. The experimental results show that the sliding mode variable structure controller designed in this paper can improve the stability performance of the vehicle obviously and improve the safety of the vehicle in the emergency.

Information theory is widely used in feature selection methods. Traditional feature selection methods employ cumulative summation to select features. Different from previous feature selection methods, this paper proposes a nonlinear feature selection method that also considers dynamic change of selected features. The experimental results demonstrate that the proposed method achieves the best classification performance in terms of average classification accuracy and highest classification accuracy. To verify the effectiveness of the proposed method, the proposed method is compared with seven very competitive feature selection methods on three different classifiers on eight real-world data sets.The experimental results shows that this algorithm has strong classification superiorities.

In order to solve the problem of low efficiency and low accuracy of cryptographic software security verification, a formal cryptographic software security verification method based on Hoare logic is proposed. The main properties of the cryptographic software are described. Through the analysis of the random vector of the cryptographic software runtime, the matrix expression of the cryptographic software is obtained, which is linearly transformed, and the variance and covariance matrix of the cryptographic software runtime data are calculated to obtain the cryptographic software. The principal component space of the runtime data matrix, and analyzes the principal components of the cryptographic software operation; analyzes the cryptographic software security formal verification process by using the logic model of the cryptographic software and the state set of the software operation, and establishes the running model of the cryptographic software. And set the security attributes, mark and process the faults that occur when the cryptographic software is running, obtain the security results of the cryptographic software running, and realize the formal verification of the cryptographic software security. The experimental results show that the method has high efficiency when verifying the security of cryptographic software, and can accurately verify the security of cryptographic software.

A remote sensing image fusion method based on cooperative empirical wavelet transform to fuse multi-source remote sensing image is proposed. Firstly, the algorithm performed principal component analysis on multi-source images to obtain a common image. Secondly, empirical wavelet transform was performed on intensity components of the common image to obtain filter banks. Then these filters were used to represent multi-spectral image intensity components and panchromatic images in multiscale. Finally, fusion image was obtained by inverse transformation. The algorithm adopts the cooperative adaptive decomposition method, which is beneficial to separate high frequency and low frequency information of the source image, and effectively improve the clarity of the remote sensing fusion image. QuickBird satellite data verifies the effectiveness of the algorithm. Visual perception and objective evaluation criteria indicate that has better advantages than other similar algorithms.

In order to solve the problem that the existing recaptured image algorithms have weak theoretical basis and low forensics rate, a new recaptured image identifying algorithm is put forward based on local plane linear point. Firstly, the proposed algorithm establishes the mathematical model in the imaging process, and provides concepts and properties of the local plane linear point from the model. Then the local plane linear point was extracted from image as the characteristic value. Finally the support vector machine is applied to classify the recaptured image with the characteristic value. The results show the proposed method can not only identify the recaptured image but also have better identification rate, and the dimension of the characteristic vector is also lower than those obtained by other algorithms.

The collection statuses of the iris image are different at different times, so the accuracy of single recognition algorithm in the multi-category iris recognition may be poor.This paper proposes an iris secondary recognition algorithm based on decision particle swarm optimization and stable texture. Use six image processing algorithms to extract stable texture features.The Gabor filtering and Hamming distance constitute the first recognition, and the Haar wavelet and BP neural network constitute the second recognition, complete secondary recognition of multi-category irises by sequence structure.Gabor filtering and neural network are adaptively optimized according to the Markov decision process and different iris libraries.The results show that the proposed algorithm can effectively improve accuracy of iris recognition.

Based on the hierarchy mechanism of the conventional chicken swarm optimization (CSO) algorithm, an improved chicken swarm optimization (ICSO) algorithm is proposed to enhance the solution accuracy and the convergence rate of the conventional CSO algorithm. The ICSO algorithm introduces several improved factors that learned from the grey wolf optimizer (GWO) and the particle swarm optimization (PSO), to extend the searching ability of the algorithm. Moreover, a duplicate remove operator is also introduced to improve the diversity of the population. Experimental results show that the accuracy of the solution and the convergence rate of the proposed algorithm are better than other benchmark algorithms.

To overcome the poor classification ability of the fast affine invariant feature extracting algorithm, an extended contour-based fast affine invariant feature extracting algorithm is developed. Firstly, the flaw in the proving and calculating procedures of the fast affine invariant feature extracting algorithm is proposed. Secondly, the definition of extended contour is proposed which is used to amend the fast affine invariant feature extraction algorithm. Thirdly, through simulation experiments, the parameter setting intervals in the algorithm are fixed and the numbers and powers of the component functions are determined. The simulation experiments are carried out to compared the proposed algorithm with the original fast affine invariant feature extracting algorithm and the multi-scale auto-convolution algorithm. The experimental results show that the proposed algorithm has better classification and recognition ability than the original fast affine invariant feature extracting algorithm and the multi-scale auto-convolution algorithm.

To solve the problems in the existing methods on estimating the frequency of MSK signals, especially under the condition of low signal-to-noise ratio (SNR), an algorithm of frequency estimation based on dual chaotic oscillator is proposed. By considering both the dynamic frequency of the chaotic oscillator system and the frequency of the signal to be measured, the chaotic oscillator’s characteristics of signal sensitivity and noise immunity are fully utilized to realize the signal frequency effective estimation under the low SNR conditions. Theoretical analysis and experiments show the effectiveness of the algorithm. Compared with the classical phase locked loop (PLL) and 2P-POWER methods, the new algorithm can accurately estimate the signal frequency under the condition of low SNR (eg. E_b/N₀= 0 dB), and the relative error is under 10^-2 orders of magnitude.

Considering the requirements of the interference power threshold for the primary users and the signal to interference plus noise ratio (SINR) for the secondary users, this paper proposes a particle swarm optimization based on genetic thought, namely genetic particle swarm optimization (GPSO). This scheme studies the issue of minimizing secondary users' transmit power in the CRN. The GPSO algorithm introduces selection, crossover, and mutation operations in the fitness value calculation, speed update, and location update phases. Compared with the Lagrange multiplier method and PSO, the GPSO algorithm reduces the transmit power and obtains a higher SINR.

Because the most existing ship detection methods do not perform well for various high sea clutter situations, according to the characteristic difference between man?made object and natural background, we propose a novel saliency detection model by computing the eigenvalues of region covariances. To the best of our knowledge, there is no existing reference in the literature about exploring the relationship between the different eigenvalues of region covariances and saliency so far. We further integrate the feature maps linearly by assigning adaptive weight value based on information entropy. The proposed approach has a remarkable ability to pop out targets and suppress distractors against clutter introduced by heavy clouds, islands as well as ship wakes. Furthermore, our model is fast and efficient which has great potential in engineering applications. Extensive experiments have been carried out on optical satellite images and experimental results demonstrate that our approach outperforms 5 existing salient object detection methods in the Area Under the Receiver Operating Characteristics.

In order to improve the quality of restored images, a single image dehazing approach based on sky segmentation and superpixel-level dark channel model is proposed. First, a simple but effective multi-threshold sky segmentation method is presented, which divides the image into sky area and non-sky area. Second, the atmospheric light is estimated in sky area. Third, the transmission maps of the two types of areas are estimated respectively. For sky area, the transmission is directly estimated by the linear relationship between the dark channel values of clear images and the corresponding hazy images. For non-sky area, the transmission is estimated by superpixel-level dark channel values. Finally, the haze-free image can be restored by the atmospheric scattering model. Experimental results show that, compared with the existing methods, the proposed method can segment the sky more accurately with better adaption of the thresholds, and can estimate the atmospheric light and the transmission map with higher accuracy and efficiency. Moreover, the dehazed images obtained by the proposed method have many advantages such as high contrast, natural color and abundant details.