To solve the problems of high false detection rate of interferential point clouds and high missing detection rate of distant sparse point clouds in automatic driving scene, the obstacle point cloud detection algorithm based on improved PointPillars was proposed. The point cloud in pillars was encoded by the aggregation module and shared multilayer perceptron（MLP）. The salient and detailed features were mapped into pseudo-image features by stacking the maxpooling and average-pooling. To solve the problem of pseudoimage feature with insufficient attention and utilization, the deep and shallow feature maps were fused by attention and residual second block（ARSB） module to optimize the gradient and enhance the coordinate attention(CA) to effective targets. The results show that the improved algorithm has high detection accuracy for global point clouds. The detection precision of the improved algorithm is better than those of the classical 3D detection algorithms of PointPillars and STD methods, especially for the detection of car category. The detection speed is fast, which meets the requirements of realtime. 

To enhance the vertical ride comfort of straddle monorail vehicle under speedincrease conditions, the parameter matching optimization was conducted on the running wheel system and suspension system. The orthogonal experimental design and the variance analysis were employed to clarify the sensitivity of various parameters to the vertical ride comfort of monorail vehicle, and the main influencing factors were screened out. Taking the rootmeansquare value of vertical acceleration as optimization objective, the collaborative optimization of the running wheel system and suspension system parameters was achieved based on the multidisciplinary platform of modeFRONTIER and Adams by the secondgeneration improved genetic algorithm. The results show that the vertical ride comfort of the vehicle is significantly improved after parameter optimization. According to the "Specification for Dynamic Performance Assessment and Test Verification of Railway Vehicles" (GB/T 5599—2019), the ride indices at measuring points of A and B are increased by 14.5% and 9.7%, respectively. The vertical ride index values at all measured points are below 2.5 after optimization, reaching the "Excellent" grade specified in the standard. The dynamic performance evaluation confirms that the optimized vehicle fully meets the requirements under speedincrease conditions.

To investigate the formation mechanism of central sagging in glass during the preheating stage of the glass forming process and the effects of two geometric parameters of glass thickness and aspect ratio, the numerical model of the fluid-thermal-structural coupled multiphysics was established based on the process flow of industrial preheating furnace. The high-temperature infrared thermography was employed to obtain the surface temperature distribution of the glass at the end of the preheating stage for verifying the fluid-thermal coupled numerical model. The developed fluid-thermal-structural coupled multiphysics model was used to simulate the formation process of central sagging during preheating for elucidating the underlying deformation mechanism. The effects of the two geometric parameters on glass sagging deformation were systematically investigated. The results indicate that geometric parameters play a critical role in glass deformation. The thinner glass exhibits more severe central sagging due to the reduced bending stiffness. The aspect ratio significantly affects the deformation behavior. When the aspect ratio is decreased from 6.32 to 1.58, the sagging depth is increased from 4.65 mm to 9.34 mm, which is attributed to the expansion of thermal strain region in the transverse direction.

To solve the problems of traditional method with poor practicability and low recognition rate for the front vehicle behavior, the new recognition method based on LSTM network model was proposed. In the process of vehicle detection, due to the difficulty in extracting features of vehicle targets and occluded targets in remote scenes, the phenomenon of wrong detection and missing detection occurred. The vehicle detection algorithm YOLOv5-BA improved by YOLOv5 in complex background was adopted. The feature fusion idea of weighted bidirectional feature pyramid network (BiFPN) was introduced, and the adaptive feature fusion module ASFF was introduced in the detection part to improve the detection performance. The results show that the high detection accuracy of the improved algorithm reaches 97.3% on KITTI data set. On this basis, combined with DeepSort algorithm, the vehicle detection and tracking are realized. By the forward vehicle behavior recognition model based on LSTM network, the average accuracy of the model on the data set is 92.3%, which can be used in the practical application of vehicle behavior recognition.

To solve the insufficient mixing problem of high-viscosity fluids in stirrer, the continuous microfluidic stirrer based on high shear and efficient mass transfer principle was proposed. The orthogonal analysis method was used to investigate the structural parameters of the single-stage stirrer and the enhancement effects on medium mixing. The optimal parameters for the single-stage micro-stirrer were determined with impeller diameter of 30 mm, impeller height of 30 mm, 8 blade and blade installation height of 5 mm above the bottom. The internal flow field and the mixing characteristics of high viscosity fluids with different mass fractions were investigated by the numerical simulation methods. The axial velocity distribution, radial velocity distribution, mass fraction distribution, and the relationship between mixing intensity and axial distance were obtained under various stirring speeds. The results indicate that increasing the rotational speed enhances the intensity of the main circulation zone within the tank, and the distribution range of the high-speed flow region in the mixing tank correspondingly expands. Elevating the concentration of high viscosity fluid reduces the main circulation range of the flow field, which leads to the decrease in mixing intensity at the outlet. Therefore, it is advisable to choose micro-stirrer with three or more stages or increase the stirring speed to facilitate the mixing of high viscosity fluids.

To investigate the effect of SO₂ on the activity of Cu-SSZ-13 catalyst, the catalyst was subjected to sulfurization treatment with different SO₂ concentrations. The structure， surface element content，acid content and other physicochemical properties of the catalyst before and after SO₂ sulfurization were analyzed by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), temperature programmed desorption by ammonia (NH₃-TPD) and temperature programmed reduction by hydrogen (H₂-TPR). The results show that the low-temperature activity of Cu-SSZ-13 catalyst is significantly decreased after SO₂ sulfurization, and the activity of the catalyst sample poisoned by 200×10^-6 sulfur dioxide at 250 ℃ is decreased by nearly 70%. After SO₂ sulfurization, the copper ion in Cu-SSZ-13 reacts with sulfur dioxide to form copper sulfate, which decreases the Cu²⁺ active site and Lewis acid site in the catalyst, while the content of copper sulfate in the catalyst sample is increased with the increase of sulfur dioxide concentration.

Based on the gas-liquid two-phase flow and evaporation model, to investigate the process of the falling film evaporation outside horizontal tube, the liquid film flows of liquid film spreading, liquid film thickness and distribution and the heat transfer characteristics of refrigerant R245fa were compared outside smooth tube and grooved tube under the conditions of low, medium and high heat flux with q values of 500, 5 000 and 20 000 W/m² at falling film Reynolds number Re of 1 690. The results show that the liquid film spreads faster outside the grooved tube than that outside the smooth tube. It takes 0.192 s for the liquid film to completely cover the external wall of the grooved tube and 0.222 s for the smooth tube， respectively. Compared with the smooth tube, the thickness of the liquid film outside horizontal grooved tube is thinner, and the thickness decrease is mainly related to the circumferential angle. At the cross section of the two liquid films meeting, the thickness decrease values are about respective 20%, 10% and 13% when the circumferential angles are ［0°，60°），［60°，160°）and ［160°，180°］. The local heat transfer coefficients outside the horizontal grooved tube are related to the magnitude of liquid film thinning and are respective 1.5-3.0, 1.1-1.2 and 1.1-1.5 times those outside the smooth tube when the circumferential angles are respective ［0°，60°），［60°，160°）and ［160°，180°］. The average heat transfer coefficient of horizontal grooved tube is higher than that of smooth tube. Under the conditions of medium and high heat flux, the sudden local deformations of the liquid film outside the smooth tube can be found. Especially at the cross section of the liquid spraying from the nozzle at the circumferential angle of 110°-180°，the bubble in the liquid film outside the smooth tube is large enough to cause the sharp decrease of liquid film thickness due to the necking effect, which leads to the local dryout outside the smooth tube. However， the local dryout can be effectively avoided for the horizontal grooved tube.

To solve the issue of point cloud data loss in reflective workpieces, the three-dimensional measurement method based on adaptive multiple exposures was investigated. The binocular structured light measurement system was designed, and the workpiece point cloud was generated by phase retrieval, phase-based stereo matching and three-dimensional point calculation. The initial sequence of exposure times was determined, and the corresponding initial images were acquired to conduct the statistical analysis of overexposure ratio in the initial images. The relationship between initial exposure time sequence and overexposure ratio was constructed for enabling the adaptive calculation of  final exposure time sequence for point cloud generation by the predetermined overexposure ratio. By fusing the multiple exposure images, the point cloud of the reflective workpiece was computed. The system accuracy was evaluated using stepped calibration block, and the experiments on aluminum guide plate and aluminum side plate were conducted. The results show that the average planar distance error of the binocular structured light measurement system is 0.126 9 mm. Compared with the single-exposure imaging method, the proposed approach using fixed overexposure ratio can generate more complete point clouds for different aluminum alloy workpieces with the number of valid points increasing by 55% and 42% for aluminum guide plate and aluminum side plate, respectively. The proposed method effectively enhances the completeness of point clouds for reflective workpieces.

To improve the accuracy and reduce the complexity of the just noticeable difference (JND) model for machine vision, the framework design of existing JND models and the design of task accuracy loss, magnitude loss and spatial distribution loss in the model training constraints were investigated, and the improved machine vision JND model of Smooth-DMV-JND was proposed. The major components of the model were composed of three convolutional neural networks parts with saliency analysis module, edge extraction module and fusion analysis module, and the proposed model was trained with modified objective function by the designed two-stage training pipeline. The performance and complexity of the model were analyzed through experiments, and the application of Smooth-DMV-JND in image compression was presented. The results show that the proposed Smooth-DMV-JND model can provide more accurate estimation of JND for machine vision with shorter analysis time. Smoothed by the Smooth-DMV-JND model with maintaining about 88% classification accuracy, the processed images achieve bit-rate saving values of 17.68% for JPEG compression and 10.69% for BPG compression than that compressing the originals directly. The JND for machine vision can be modeled more effectively by Smooth-DMV-JND, which can guide the removal of redundancy in images and is beneficial to machine vision-oriented image compression.

To solve the problems of high common mode voltage (CMV) and severely conducted electromagnetic interference in five-phase permanent magnet synchronous motor (PMSM) with near-four-vectors-based space vector pulse width modulation (SVPWM), the novel CMV-reduction-based spread-spectrum SVPWM strategy was proposed. To reduce CMV, according to the CMV generation mechanism of the five-phase PMSM driven by voltage source inverter, the vectors in opposite direction were employed to replace the zero vector in the near-four-vectors-based SVPWM strategy. To restrain current ripple and conducted electromagnetic interference, the frequency of SVPWM was adjusted on-line based on the mechanism of current ripple and high-frequency harmonic distribution. The five-phase PMSM experimental platform was constructed to compare the proposed RCMV-3SVPWM strategy with the conventional NFV-SVPWM strategy, and the dynamic tracking performance tests were carried out on the five-phase PMSM. The results show that by the proposed strategy, the CMV can be reduced by 40% with decreased current ripple of 0.30 A, and the conducted electromagnetic interference is effectively suppressed with maintaining good steady-state and dynamic performance.

The high torque density U-shaped permanent magnet in-wheel motor was proposed for the application of distributed drive electric vehicle hub direct drive. The impact of temperature rise on the torque performance of motor was theoretically analyzed, and the motor loss and current angle were identified as the optimization design objectives. The sensitivity analysis method was introduced, and the sensitivity indices of design parameters to the optimization objectives were calculated to determine the highly sensitive design parameters for optimization. Using the response surface method and multi-objective genetic algorithm to implement multi-objective optimization of the motor, the optimal design scheme of the motor was finally determined. The loss characteristics, temperature distribution characteristics and torque performance of the motor before and after optimization were compared, and the operating point characteristics of the permanent magnet were analyzed. The experimental prototype of the wheel hub motor was machined. The results show that the effectiveness of the motor and the optimal design is verified by the theoretical analysis and experimental results, and the output torque of the motor is increased by 25.8% after optimization.

Due to the complex factors influencing the liquefaction resistance of sand and the limitations of laboratory tests in terms of mesoscale observation and sample preparation methods, the effects of sand particle size and distribution on liquefaction resistance under different cyclic stress conditions  were systematically analyzed based on discrete element numerical simulations. The results show that the variations of cyclic stress ratio (CSR) do not significantly alter the fundamental accumulation pattern of excess pore water pressure in saturated sand. Under the low CSR conditions, the development of excess pore water pressure exhibits distinct two-stage characteristic of slow first and then rapid progressing. When CSR is less than or equal to 0.25, the coefficient of uniformity (Cu) plays dominant role in influencing the liquefaction resistance of sand with optimal Cu value for obtaining the maximum resistance. The median particle size (D50) is generally not the sole dominant factor for controlling the liquefaction resistance of sand and the influence becomes significant only when the particles are relatively uniform with small Cu value.

Based on three-dimensional thermoelasticity theory, the thermo-mechanical response of the laminated plates with the temperature-depended material properties was investigated under the combined action of temperature and load. The state-space method was employed to formulate the governing state equation with displacements and stresses as the primary state variables. The state variables were subsequently discretized in the in-plane directions by the differential quadrature method. Based on the interlaminar continuity conditions, the transfer relationship of the state variables between the top and bottom surfaces of the laminated plate was derived. The exact solutions of displacement and stress were obtained by combining the temperature field and the mechanical boundary conditions on the surfaces of laminated plate. The effects of temperature variation, number of layers and material properties on the displacement and stress distribution in the laminated plate were analyzed. The results show that as the external temperature rise is increased, the displacements and stresses within the laminated plate are increased proportionally, and the temperature dependence of material properties significantly influences this behavior. Under the identical thermal and mechanical loading conditions, the laminated plates of identical geometry with different material constituents exhibit pronounced reduction in the x-direction displacement and the xy-plane shear stress with the increasing of layer number. The structural performance of laminated plates in thermally varying environments can be enhanced through strategic optimization of layer numbers and constituent materials.

To solve the problem of poor detection performance for densely arranged rebars in reinforced concrete, the non-destructive examination method based on millimeter wave radar was proposed. By establishing the millimeter wave radar signal model and applying the range migration algorithm, the near-field synthetic aperture detection system was constructed to collect data from reinforced concrete. The range migration algorithm was employed to perform radar imaging and reconstruct rebar layouts at different depths. On this basis, the Kuan filter was utilized to suppress the image clutter. The filtered images were used to quantitatively calculate the rebar diameter and spacing. The experimental results show that the calculations based on coordinates from Kuan filtered images achieve high accuracy with detection errors for both rebar diameter and spacing less than 2 mm, and the error in concrete cover thickness is less than 5 mm. Compared with the effects of mean filtering and median filtering, the rebar images obtained through Kuan filtering are more distinct.

To investigate the evolution of the axial compressive stress-strain relationship of ultra-high performance seawater sea-sand concrete (UHPSSC)  under seawater environment, the axial compression tests were conducted. Based on the experimental data, the prediction model for the stress-strain curve of UHPSSC was developed with considering the seawater immersion duration. The effects of the ceramsite sand replacement ratio and seawater immersion duration on the stress-strain relationship of UHPSSC were analyzed. The results indicate that as the ceramsite sand replacement ratio is increased, the peak stress, peak strain and elastic modulus all exhibit decreasing trend. For the given ceramsite sand replacement ratio, with the increasing of seawater immersion duration, both the peak stress and elastic modulus of UHPSSC initially are increased and subsequently decreased. In the early stage of immersion, the enhancing effect of seawater on the mechanical properties of UHPSSC outweighs the deteriorating effect, while the deteriorating effect becomes dominant in the later stage. The evaluation of the prediction model demonstrates that the curves predicted by the newly proposed model show good agreement with the experimental curves.