Abstract:This paper mainly focuses on the application and feasibility of the Active Front Steering （AFS） control system. Considering that tire behaves nonlinearly during emergency steering and that the vehicle states involved in the AFS control system and the road adhesion coefficient having important effect on the stability are difficult to measure, a controller using nonlinear sliding mode algorithm was designed to synthetically take into account the influence of load transfer, tire nonlinearity, and road on the vehicle stability. Meanwhile, with the use of signals measured from the existing Inertial Measurement Unit (IMU) sensor of ESP system and the application of Unscented Kalman Filter （UKF） algorithm, a state estimator was established to dynamically estimate the vehicle state information and road adhesion coefficient for sliding mode controller. On the above basis, the desired superposition angle is precisely reversely calculated by the nonlinear tire model after the desired tire force is obtained, which verifies the effectiveness of the control system over the "tire-road" attachment capability range. Finally, simulations of fishhook test with high road adhesion and step input with low road adhesion indicate that the state estimation combined with IMU and UKF ensures the feasibility of AFS control system and effectively improves the vehicle stability.

Abstract:The pressure excitation on the front window surface of the vehicle is an important indicator of unsteady flow and aerodynamic noise in the front window area. The complex unsteady flow in this area produces a larger scale vortex structure, which leads to complex unsteady pressure excitation on the front side window surface. In this paper, the turbulent pressure and acoustic pressure excitations on the front window surface of the vehicle are obtained by the hybrid Computational Aeroacoustics（CAA） method based on Acoustic Perturbation Equations（APE）. The Dynamical Mode Decomposition（DMD） is introduced to analyze the pressure excitation on the front side window surface, which demonstrates that the turbulent pressure excitation has the frequency-based regional distribution characteristics and the acoustic pressure excitation acts as radiation sound field characteristics. The turbulent pressure excitation, acoustic pressure excitation and the relative contributions of different excitation sources to the car interior noise are discussed. The main turbulent pressure excitation on the front side window surface identified by the DMD is generated by the wake vortex shedding of the rear view mirror trail. Its characteristic frequency is 59 Hz, which is consistent with the experimental measurement results and it also verifies the validation of the turbulent pressure excitation calculation results. Comparing the DMD modes of the turbulent pressure and acoustic pressure with the same frequency in the space section of the front side window region, the main sound source position in the front side window region is identified. One is located at the pedestal of the rearview mirror, which is generated by the convection of the pedestal vortex in this region. The other is located at the lower edge of the mirror body, resulting from the separation vortex in this area. The latter is identified by the microphone array in the wind tunnel test，which validates of the estimation of the acoustic field.

Abstract:For an Automated Manual Transmission (AMT) with clutch inversed for the pure electric vehicle， this paper presented a program that the synchronizer of the AMT was located on the second axis. The seamless gear shifting can be obtained by switching the clutch to the synchronizer. In order to make the output torque of gearbox change smoothly, addressing the vehicle jerk and friction work as the shift performance indexes, different coordinated control strategies were adopted for different stages during shifting: the motor torque remained unchanged in the torque phase, and the clutch torque was controlled coordinately; the speed difference of the clutch was controlled following the target trajectory by motor speed controlled by PID and adaptive fuzzy PID method in the inertial phase. The dynamic model of MATLAB / Simulink was established and simulated, which showed the feasibility and effectiveness of the coordinated control strategy for gear shifting. Compared with the traditional PID controller applied in the inertial phase, the adaptive fuzzy PID controller can effectively improve the shift quality, the max jerk during gear shift and the friction work decreased, the output torque of the transmission changed seamlessly without power interruption, and the comfort of the vehicle was significantly improved.

Abstract:As the uncertainty of the nonlinearity and load disturbance of the electric brake excitation control of Hybrid Motor Dump Truck (HMDT) impacts on vehicle stability, especially on battery life, a nonlinear variable structure excitation control strategy with feedback precision linearization control and sliding mode control was proposed. This paper established the Simple Input Simple Output（SISO） second-order nonlinear model of electric brake excitation control, and the nonlinear problem was transformed into a linear problem by the precise linearization of the nonlinear model. Then, the speed closed-loop excitation controller was designed with sliding mode variable structure control. At the same time, in order to weaken the system buffeting, ensure system stability under the bad working condition of the mine, and ensure stable speed and braking current, the Luenberger load disturbance state observer was designed. MATLAB simulation results show that, compared with PID controller, NVSC controller has the advantages of dynamic performance and quick response. Under load disturbance fluctuation, the motor speed and braking feedback current are stable and the system is robust, which guarantees the battery life and HMDT stability.

Abstract:This study starts from setting up the finite element simulation model of B-pillar assembly. Then, the size, free size and laminate orientation of carbon fiber composite B-pillar reinforced plate were optimized based on the principles of constant stiffness. Maximum strength and displacement of the assembly were obtained through Finite Element (FE) simulation of quasi-static three-point bending test on B-pillar assembly. B-pillar sample was fabricated by a Vacuum Infusion Process (VIP) and conducted on a three-point bending test to check the strength index of the B-pillar assembly. Finally, according to 2018 C-NCAP standards, side impact performance of vehicle was elaborated. The comparison results among stiffness, flexural property, side impact intrusion, intrusion velocity and acceleration indicate that the optimization design of CFRP can replace the original B-pillar steel reinforced plate under the premise of ensuring rigidity, strength and side impact performance. CFRP B-pillar reinforcement plate is 1.376kg lighter than that of the conventional one, and the weight loss ratio is up to 76.4%.

Abstract:With the large increase of the capacity of large nuclear / thermal power stations, high power hydraulic coupling is the core component of the main boiler feed water pump, and the strength of working impeller has become an important factor affecting the safety and stability of the power station system. In this paper, the impeller assembly body of a certain hydraulic coupler was taken as the research object. A fluid solid coupling analysis model of the full flow passage was established by the one way fluid solid coupling calculation method. The contact algorithm was used in the end face location of the pump wheel and the turbine sleeve, and the beam element was used to simulate the screw connection effect, calculation and analysis of strength of impeller assembly structure under typical working conditions. The results show that the overall deformation and stress of the assembly increase with the increase of the rotational speed ratio, and the deformation size of the impeller is basically proportional to the length of the rotation radius in the corresponding region. It shows that the centrifugal load is the main reason that affects the strength of the assembly body of the impeller, and the local stress concentration in the connection area of the impeller appears due to the screw pretightening force effect. The inner edge of the turbine sleeve is the weak area of the impeller assembly structure strength, and the analysis results are consistent with the existing results. This research work provides an effective theoretical guidance for the structural design and optimization of the impeller of the autonomous high-power hydraulic coupling.

Abstract:Applying the PowerFLOW software using the Lattice Boltzmann Method (LBM), and combining with the very Large Eddy Simulation Method (VLES), the unsteady flow field of the MIRA notchback model was solved to study the structure and flow characteristics of the external flow field. By analyzing the process of the flow from the A-pillar along the roof to the C-pillar and the rear of the flow field, the structure and flow mechanism of the C-pillar vortex, D-pillar vortex, and partial separation vortex were explored. The transient flow field was analyzed to explore more precise random flow characteristics. In the time domain flow field analysis part, the flow field structure of the wheel, rear wind window and rear part is complicated. The analysis results of the frequency domain further showed the vibration frequency and pulsation characteristics of the vortex. The vibration frequency of the rear part is 12Hz, and the vibration frequency of the side window, top of the engine compartment, roof, and side of the body is 23Hz. In addition, the formation mechanism of the vibration frequency was explored. By analyzing the pressure pulsation, it is found that large vibration energy occurs around the chassis, rear part of the vehicle body, and rear wheel area. It is concluded that the flow field structure in the above area is complex and it contributes a lot to the drag. When comparing the simulation results with the experimental results, the flow field structure is similar, and the number and position of the vortex cores are well consistent, which verifies the reliability of the simulation.

Abstract:Aiming at the problem of large torque ripple in low speed operation of switched reluctance motor, this paper presented a current predictive control strategy of fixed switching frequency to improve the current dynamic tracking ability and to realize the constant speed control of the motor. According to the electromagnetic relationship, the current slope of positive current, negative voltage and zero voltage in the current control cycle is estimated, and the duty cycle of the current cycle is predicted, so that the winding current can be precisely controlled. The simulation and experimental results verify the feasibility of the current predictive control strategy. It shows that the proposed method has a fast dynamic response and can track the reference current very well at low speed, which can restrain the torque ripple and electromagnetic noise.

Abstract:A High Step-up DC/DC (HS) converter was proposed. It steps up the voltage by using a special structure, and the boost structure is composed of two inductors, a capacitor and two diodes. Combing the special boost structure with the double-switch boost DC converter, HS converter is obtained, which depends mainly on charging and discharging of the inductors and capacitors to realize the step-up function. In addition, Repeated HS (Re-HS) and Generalized HS (G-HS) converters were proposed. The voltage gain of the Continuous Conduction Mode (CCM) for HS converter was analyzed. The voltage gain, voltage stress and ripple current of HS converter was deeply studied and compared with other converters. The comparative analysis shows that the HS converter has the characteristics of high voltage gain, low voltage stress, and low current ripple. By building the actual circuit of the HS converter, the correctness of the proposed topology is verified.

Abstract:The quasi-Z source inverter (qZSI) has a unique impedance network that couples the power supply to the conversion circuit and has a special buck-boost function. Inductors are the core components of the impedance network, which occupy a large proportion of qZSI, and thus the size and weight of the two inductors can be significantly reduced by a coupled inductor design. In this paper, a novel design of coupled inductor is proposed to reduce the input current ripple of qZSI. Then, the loss of the coupled inductor and the temperature rise are analyzed. Moreover, the analytical expressions of the loss and temperature of the magnetic core and the winding are derived. A multiphysics model using the finite element software Ansys and a 6 kW qZSI prototype were built for simulation and verification.

Abstract:The transient output prediction of solar power source is of great significance for power grid stability analysis, power quality control and fault diagnosis. To this goal, the discrete model and linear prediction model of ideal solar power source were established. Then, a regularized least square prediction scheme was proposed to estimate the unchanged model parameters. When the power source model parameters vary, the prediction model parameters are continuously updated in real-time by the Sliding Rectangle Window (SRW) Recursive Least Square (RLS) method. Unlike the standard RLS, SRW-RLS adopts a data update strategy based on sliding rectangle window, which improves the tracking performance and prediction accuracy. The experimental results show that the proposed prediction schemes achieve good prediction accuracy and SRW-RLS is able to adapt well to the changes in the parameters of power source model.

Abstract:In order to realize the quantitative analysis and prediction on the operation state of the transformer, the interchange complex matter element was built between dissolved gases in transformer oil and typical faults. Analytic Hierarchy Process (AHP) and maximum information entropy were used to determine the subjective and objective weights influencing the transformer health level, respectively. The quantitative analysis of the transformer health level was proposed based on matter element maximum information entropy. The Support Vector Machines (SVM) algorithm was adopted to predict the operation condition of transformers, the parameters （c and g） were optimized by grid-search, Genetic Algorithm (GA) and Particle Swarm Optimization（PSO），and the optimal prediction model was established. This method provides a good guiding value for the elimination of transformer faults, overhaul decisions and online predictions.

Abstract:An adaptive lateral stability control system based on Fuzzy Neural Network（FNN） is proposed to solve the uncertain stability problems of Distributed Drive Electric Vehicles（DDEV） such as insufficient steering and lateral instability. The system consists of an upper direct yaw moment controller and a lower torque distribution controller. The upper direct yaw moment controller can get the desired direct yaw torque according to the error of side slip angle caused by uncertain factors. The lower torque allocation controller distributes the output of the upper controller to each hub motor according to the tire's load, which can adjust the vehicle attitude efficiently and improve the steering ability and lateral stability of the vehicle. Simulation results show that the proposed control system can significantly improve the lateral stability of DDEV，and show the better control effect than the traditional fuzzy control.

Abstract:In order to solve the problems such as large trigger voltage, large voltage snapback margin and slow turn-on speed of Electrostatic Discharge（ESD） protection devices based on the Silicon Controlled Rectifier（SCR） structure, a dual-directional SCR（DDSCR） device embedded with PMOS and triggered by the RC circuit（DUT3） was designed. Three types of devices including the conventional DDSCR（DUT1），DDSCR embedded with PMOS （DUT2） and DUT3 were fabricated in a 0.35 μm Bipolar-CMOS-DMOS process. Their ESD characteristics were measured by the transmission line pulse system. The test results show that, compared with DUT1, the DUT2 trigger voltage decreases from 31.3 V to 5.46 V, the holding voltage increases from 3.59 V to 4.65 V，and the voltage snapback margin of DUT3 is very small, but the high leakage current up to 10-2 A makes it unsuitable for ESD protection. By introducing an RC circuit to provide a fixed gate voltage for the embedded PMOS in the DUT2, the modified DUT3 shows not only a further reduced voltage snapback margin but also a shorter response time of only 12.6 ns. Compared with DUT1, the turn-on speed of DUT3 increases by about 71.5%, and the leakage current can be stabilized at the order of 10-10 A. This optimized DUT3 is suitable for ESD protection in the low-voltage integrated circuits with requirements of high-speed, small snapback margin and narrow ESD design windows.

Abstract:In order to deal with the problem of harmonic look in the traditional Delay Locked Loop (DLL), a DLL-based 90°phase-shifter with a Schmitt Frequency Selector（SFS） was proposed. The SFS and dual delay lines were employed to achieve wider locking frequency range. In addition, the proposed SFS exhibits high capability of frequency noise suppression, which improves the stability of the proposed phase-shifter. The proposed phase-shifter, fabricated in SMIC 55 nm CMOS technology, occupies an active area of 0.131 mm2 and utilizes a 1.2 V supply voltage. The test results show that the proposed phase-shifter has an operating frequency ranging from 250 to 800 MHz and consumes 5.98 mW at 800 MHz. Furthermore, the measured peak-to-peak and root-mean-square (rms) jitters of 90°phase-shifted clock are 25.9 and 2.8 ps, respectively.

Abstract:As the scale of tasks in the cloud environment continues to expand, the problem of high energy consumption in cloud computing centers has become increasingly prominent. In order to solve the problem of task assignment in a cloud environment and to effectively reduce energy consumption, a Modified Particle Swarm Optimization algorithm (M-PSO) was proposed. First, a cloud computing energy consumption model, which takes into account the processor's execution energy consumption and task transmission energy consumption, was introduced. Based on the model, the task assignment problem was defined and described, and the particle swarm optimization algorithm was used to solve this problem. In addition, a dynamically adjusted inertia weight coefficient function was constructed to overcome the local optimization and slow convergence problem of the standard PSO algorithm, and the strategy can effectively improve the system performance. Finally, the performance of the introduced algorithm model was evaluated by simulation experiments. The results show that the M-PSO algorithm can effectively reduce the total energy consumption of the system compared with other algorithms.

Abstract:The Delay Tolerant Network (DTN), as an intermittently connected network, transfers messages through opportunistic encounters between nodes. It is difficult to intercept messages because of the multi-hop and random delivery method, which shows that DTN is very suitable for covert transmission of information. However, it also limits transmission efficiency. In order to develop transmission efficiency and to guarantee the security of covert transmission, a routing algorithm was proposed. Firstly, an efficient message forwarding strategy was designed to improve the delivery efficiency of messages by utilizing the static social characteristics and real-time encounters of nodes. Secondly, because massive copies of message lead to high probability of delivery but also reduce the security of covert transmission, the number of copies was dynamically set according to the topology of the network. Finally, the proposed algorithm was simulated. Compared with the classical DTN algorithm, the simulating results show that the proposed algorithm can improve the message delivery rate of DTN，reduce network overhead, and increase the security of messages.

Abstract:Human Visual System（HVS） first roughly perceives global areas，then centers on the detailed local areas for the perception of image quality. In this paper，a novel blind Image Quality Assessment（IQA） algorithm was proposed for tone-mapped images by combining local and global features. First，the global features were extracted based on color moments，global entropy and bright/dark pixels' distribution under overexposure/underexposure conditions. Then，local contrast，local entropy and wavelet energy based on blocks were utilized to extract local features. Finally，global features were combined with local features to constitute a final feature vector. And all these feature vectors mentioned above were trained using Support Vector Regression（SVR） to generate a model，which bridges the feature space with quality space. Extensive experiments on a public ESPL-LIVE HDR database have demonstrated that the proposed method has a high consistency with subjective evaluation and outperforms state-of-the-art no-reference IQA metrics.

Abstract:A new method of high resolution radar target recognition based on Convolution Neural Network (CNN) was presented. To solve the problem of slow convergence of loss function values during the training process when small samples are applied to the deep CNN, High Resolution Range Profile (HRRP) features were firstly extracted by using the improved CNN combined with the Batch Normalization (BN) algorithm, and then classified by using a Support Vector Machine (SVM). The experimental results using high-fidelity electromagnetic simulation data of military vehicles validate the effectiveness of the proposed method.