Table of Content

25 November 2020, Volume 42 Issue 11 Previous Issue    Next Issue

Select

A Model-based Description Method for　Key Performance Indicators　of DSRC System in　Vehicle Platoons Shi Jia, Luo Yugong, Qi Yunlong, Li Keqiang 2020, 42 (11):  1449-1457.  doi: 10.19562/j.chinasae.qcgc.2020.11.001 Abstract ( 475 )   PDF (1282KB) ( 352 )   Save In view of the lack of model-based description method of communication performance indicators in the current research of vehicle platoon communication system, the end-to-end delay is described based on three-parameter Burr distribution in this paper, according to the features of wireless network communication delay. According to the probability distribution law of wireless signal power intensity, with consideration of the effects of obstacle obstruction in the line-of-sight propagation, the description method of packet error rate (PER) is established based on Nakagami-m distribution. Then a communication protocol is formulated for vehicle platoon application, and the communication system platform is developed based on dedicated short range communication. Finally, multi-scenario vehicle tests are conducted to verify the correctness of the description method. The results show that the method proposed can characterize the performance features of vehicle platoon communication system under actual conditions, and reflect the effects of environmental factors on communication performance, suitable for the control of vehicle platoons References | Related Articles | Metrics

Select

Clustering Evaluation Method of Autonomous Driving Safety for Multi-dimensional Logical Scenario Zhu Bing, Zhang Peixing, Zhao Jian 2020, 42 (11):  1458-1463.  doi: 10.19562/j.chinasae.qcgc.2020.11.002 Abstract ( 421 )   PDF (801KB) ( 321 )   Save Accurate and reliable safety evaluation is the basis for the promotion and application of autonomous vehicles. However, the driving environment is complex and changeable, and traditional methods can’t meet the needs of its safety evaluation. A multi-dimensional logical scenario-oriented clustering evaluation method is established in this paper for autonomous driving safety evaluation. Firstly, the results of ergodic test hazard parameters in a multi-dimensional logic scenario are clustered based on Gaussian model; then, two basic evaluation indicators of the danger field dispersion and danger field range are proposed by using the clustering result. An autonomous driving safety clustering evaluation parameter, namely scenario hazard rate is built by coupling the two indicators. Finally, a black box autopilot algorithm is evaluated using the method proposed in the paper. The results show that the method proposed in this paper can comprehensively consider the statistical law in a multi-dimensional logical scenario, obtain a quantitative index, and achieve a comprehensive scientific evaluation References | Related Articles | Metrics

Select

Research on Behavior Recognition Algorithm of Surrounding Vehicles for Driverless Car Cai Yingfeng, Tai Kangsheng, Wang Hai, Li Yicheng, Chen Long 2020, 42 (11):  1464-1472.  doi: 10.19562/j.chinasae.qcgc.2020.11.003 Abstract ( 523 )   PDF (1173KB) ( 562 )   Save The behavior recognition of surrounding vehicles is very important to improve the decision-making rationality and the control safety for driverless cars. Traditional methods of surrounding vehicle’s behavior recognition suffer from low accuracy and poor robustness, without considering the interaction between adjacent traffic objects. In order to solve these problems, SLSTMAT (Social-LSTM-Attention) algorithm is proposed to achieve high accuracy for surrounding vehicle’s behavior recognition. The social characteristics of target vehicle are innovatively introduced and extracted by convolutional neural network. Based on deep learning,the recognition model for vehicle behavior is established. Attention mechanism is applied to capture multiple time-step information in the behavior time window. The algorithm is verified by HighD trajectory data set and real vehicle data. The results show that the accuracy rate of SLSTMAT for surrounding vehicle’s behavior recognition reaches 94.01%, and the precision of behavior recognition reaches 90% at 1s before the target vehicle driving to the lane change point, which has high engineering application value References | Related Articles | Metrics

Select

Gearshifting Strategy for Robot-driven Vehicles Based on Deep Reinforcement Learning Zhou Nan, Chen Gang 2020, 42 (11):  1473-1481.  doi: 10.19562/j.chinasae.qcgc.2020.11.004 Abstract ( 245 )   PDF (1035KB) ( 345 )   Save In order to improve the power performance of the robot-driven vehicle, a gearshifting strategy for robot-driven vehicles in favour of power performanceis proposed based on deep neural network reinforcement learning. Firstly the vehicle longitudinal dynamics model, the kinematics model of the robot driver and the reinforcement learning model of the gearshifting strategy for the robot-driven vehicles described with Markov process are constructed, and the state space, the action space and the reward and punishment mechanism are designed with the vehicle speed, acceleration and throttle opening as the state variables and the gear position as the action variable. Then the gearshifting strategy for the robot-driven vehicles in favour of power performance is solved by using deep neural network reinforcement learning algorithm. The comparison between the simulation and test results with the strategy proposed in this paper and those with other strategyie verifies the effectiveness of the proposed strategy References | Related Articles | Metrics

Select

Pedestrian Detection Method Based on Neural Network and Data Fusion Zhu Bo, Huang Maofei, Tan Dongkui, Hu Xudong, Gu Jiaxin 2020, 42 (11):  1482-1489.  doi: 10.19562/j.chinasae.qcgc.2020.11.005 Abstract ( 290 )   PDF (1934KB) ( 324 )   Save In view of the poor results of pedestrian recognition with millimeter wave radar and the inaccuracy of camera in distance detection, a novel pedestrian detection method based on neural network and data fusionis proposed in this paper. The neural network algorithm is used to screen in the suspected pedestrian targets from the targets detected by millimeter wave radar and match it with the pedestrian targets detected by Mobileye. Then the data fusion of matched pedestrian targets between millimeter wave radar and Mobileye is carried out by using Kalman filter algorithm. The results of real vehicle test show that the pedestrian detection method based on neural network and data fusion can accurately match the pedestrian targets detected by the two sensors and enhance the accuracy of pedestrian location detection References | Related Articles | Metrics

Select

Review of Research on HEV Energy Management Based on Adaptive Dynamic Programming Jin Hui, Zhang Zihao 2020, 42 (11):  1490-1496.  doi: 10.19562/j.chinasae.qcgc.2020.11.006 Abstract ( 318 )   PDF (758KB) ( 559 )   Save Hybrid electric vehicles integrate the both advantages of fuel vehicles and battery electric vehicles, which can not only effectively reduce fuel consumption while meeting power demand, but also achieve long-distance driving. The energy management strategy of hybrid electric vehicles can achieve better power distribution, so that the engine and electric motor can work in high efficiency regions. Current energy management strategies can be divided into three categories: rule-based, optimization-based and artificial intelligence-based control strategies. In this paper, the adaptive dynamic programming (ADP) algorithm in artificial intelligence-based category is summarized, the structural development, and the algorithm implementation and the application to energy management strategy research of ADP algorithm are analyzed, with its present inadequacies and future development trends pointed out. References | Related Articles | Metrics

Select

Judging Method for Lateral Stability of Intelligent Vehicle Based on Combined Clustering Gu Xianguang, Meng Kewei, Yao Xinxin, Wang Hongbo 2020, 42 (11):  1497-1505.  doi: 10.19562/j.chinasae.qcgc.2020.11.007 Abstract ( 208 )   PDF (1094KB) ( 350 )   Save On the basis of the research on the traditional judging methods of vehicle stability, a new method of judging vehicle lateral stability is proposed based on the theories of neural network and cluster analysis. A combined clustering method combining SOFM neural network and K-means clustering is adopted to conduct off-line cluster analysis on vehicle driving parameters collected and to obtain all clustering centers and their stability level. The mean value method is applied to online updating clustering centers, the distance between real-time data and clustering center is calculated, and the vehicle stability is judged real-time based on minimum distance criterion. The performance of that stability judging method is analyzed with the tire force method as reference. Finally, the judging results are taken as the basis for the involvement of the stability control strategy, and the effectiveness and accuracy of the method is verified through CarSim/Simulink joint simulation and hardware-in-the-loop test References | Related Articles | Metrics

Select

Research on Parallel Braking Control of Distributed Four-wheel-drive Electric Vehicle Zhu Shaopeng, Jiang Xudong, Wang Yanran, Ye Xingyu, Yu Kai, Xu Linfeng 2020, 42 (11):  1506-1512.  doi: 10.19562/j.chinasae.qcgc.2020.11.008 Abstract ( 214 )   PDF (965KB) ( 307 )   Save Considering the engineering application requirements of regenerative braking system, a distributed parallel braking control strategy is proposed for distributed four-wheel-drive electric vehicle, which mainly focuses on braking safety and stability, while taking into account braking economy. The strategy adopts a hierarchical control structure, including the desired yaw moment formulation layer, the four-wheel electric braking force distribution layer and braking force constraint layer. Based on the direct yaw moment control, the designed parallel braking control strategy optimizes the braking stability of the vehicle, and comprehensively considers the constraints such as braking strength, battery SOC value, and vehicle speed to improve the braking economy. Finally, a real vehicle test is carried out to verify the effectiveness of the designed parallel braking control strategy under the conditions of right angle turn braking and single lane change braking References | Related Articles | Metrics

Select

Study on Coordinated Control of Trajectory Tracking and Yaw Stability for Autonomous Four-wheel-independent-driving Electric Vehicles Zhang Lei, Zhao Xianhua, Wang Zhenpo 2020, 42 (11):  1513-1521.  doi: 10.19562/j.chinasae.qcgc.2020.11.009 Abstract ( 321 )   PDF (984KB) ( 493 )   Save In order to improve trajectory tracking precision and dynamics stability of intelligent vehicles under extreme conditions such as high speed and low adhesion conditions, this paper proposes a coordinated control strategy of trajectory tracking and direct yaw-moment based on the layered control structure for autonomous four-wheel-independent-driving electric vehicles, which consists of an upper and a lower controller. Firstly, in the upper controller, a model predictive controller is developed based on the three-degree-of-freedom (3-DOF) vehicle dynamics model considering multiple nonlinear constraints while optimizing the prediction time domain and control time domain through theoretical analysis. The steering angle of the front wheels and the additional yaw-moment are derived via optimization derivation. Then, in the lower controller, the optimal distribution of wheel longitudinal force is realized with the objective of minimizing the load rate of tires. Considering the real-time requirement of the control strategy, the effective set algorithm is adopted for the optimal torque distribution. Finally, the effectiveness of the proposed control scheme of coordinated controller of trajectory tracking and yaw stability is verified under different vehicle speeds and road adhesion coefficients in the CarSim-Simulink joint simulation. References | Related Articles | Metrics

Select

State of Charge Estimation for Traction Battery Based on EKF-SVM Algorithm Liu Xingtao, Li Kun, Wu Ji, He Yao, Liu Xintian 2020, 42 (11):  1522-1528.  doi: 10.19562/j.chinasae.qcgc.2020.11.010 Abstract ( 296 )   PDF (925KB) ( 426 )   Save In view of that　the single SOC estimation algorithm cannot concurrently meet the requirements of multi-indicators, an algorithm combining the extended Kalman filtering (EKF) and support vector machine (SVM) is proposed. By dynamically tracking the model parameters and estimating the open-circuit voltage in real-time, the preliminary SOC estimation is obtained by using EKF algorithm. Furthermore, by training the DST condition data output from EKF algorithm, SVM model is obtained and its regression prediction ability is utilized to perform error compensation on preliminary estimation, further reducing the error of SOC estimation. The results of simulation show that compared with EKF and EKF-BP algorithms,the proposed EKF-SVM algorithm has better robustness and adaptability and can achieve accurate estimation of battery SOC,with the maximum absolute error of about 1% References | Related Articles | Metrics

Select

Fault Diagnosis of Traction Battery Pack Based on Improved Convolution Neural Network and Information Fusion Peng Yunsai, Xia Fei, Yuan Bo, Wang Zhicheng, Luo Zhijiang 2020, 42 (11):  1529-1536.  doi: 10.19562/j.chinasae.qcgc.2020.11.011 Abstract ( 211 )   PDF (830KB) ( 337 )   Save A fault diagnosis method for lithium-ion battery based on improved CNN and information fusion is proposed in this paper. Firstly, the Concat layer is added before the fully connected layer to fuse the features of different levels, so an improved CNN model is established. Then, the MCE criterion is used to optimize the cross entropy loss function, solving the problem that the non-label dimension gradient is not processed. In view of the insufficient practicality of the convolution kernel structure manually determined, the BIC criterion is used to determine the optimal structure of convolution kernel of the model. After diagnosis by using the improved CNN mentioned above, the conditions for confirmed diagnosis are introduced for judgment. For the diagnosis results not meeting the conditions, the general CNN network is further used for auxiliary diagnosis. By fusing preliminary diagnosis result and auxiliary diagnosis result using D-S evidence theory, the results of final diagnosis are obtained. The test results show that the method proposed can effectively enhance the accuracy of fault diagnosis on traction battery packs. References | Related Articles | Metrics

Select

Steady Flow Test and Combustion Analysis of Different Spiral/ Tangential Air-Passage Valves Zhang Wei, Wang Yingxiao, Peng Yiyuan, Chen Zhaohui, Jiang Qianyu 2020, 42 (11):  1537-1544.  doi: 10.19562/j.chinasae.qcgc.2020.11.012 Abstract ( 178 )   PDF (1358KB) ( 332 )   Save Steady-flow test and combustion analysis of the D19 four-valve and dual-intake-port engines are carried out using the steady-flow test bench and CFD simulation. The results show that with the same valve under the maximum opening there is severe interference of airflow between the two air passages, which limits the increase of the vortex ratio and the flow coefficient; while the different valves designed for the maximum opening of the valve can make full use of the space in the cylinder and make the airflow in the cylinder form mutual staggered flow to avoid serious interference of airflow between the airways so as to increase the flow coefficient and vortex ratio. The increase of the air intake can make most of the fuel and air more fully mixed.The increase of the vortex ratio makes the fuel and gas mixture more uniform and the combustion is more sufficient. As a result, the cylinder heat release is increased, with Soot and CO greatly reduced and NOx slightly increased References | Related Articles | Metrics

Select

Study on the Effect of Lubricating Oil Additives on the Microphysical Properties of Diesel Particulate Matter Xu Guangju, Gao Jian, Li Mingdi, Zhao Yang, Lin Ling 2020, 42 (11):  1545-1551.  doi: 10.19562/j.chinasae.qcgc.2020.11.013 Abstract ( 130 )   PDF (1132KB) ( 257 )   Save In this paper, three different lubricating oil additives are used to configure the test fuel. The diesel engine bench tests are conducted and particulate samples are collected. Then the effect of lubricant additives on particle morphology, structural characteristics and graphitization degree are studied by means of SEM/TEM/Raman spectra. The results show that the particle diameter depends on the PAHs content of the fuel. The microcrystalline size of particles decreases with the increase of fuel cetane value. The larger the microcrystalline size, the higher the degree of ordering of carbon layer structure.The larger the curvature of the basic carbon particle, the worse the electron resonance stability of the carbon layer References | Related Articles | Metrics

Select

Design and Test of Active Roll Vehicle Liu Pingyi, Ke Chengpeng, Gao Ruolin, Li Haitao, Wei Wenjun, Wang Ya 2020, 42 (11):  1552-1557.  doi: 10.19562/j.chinasae.qcgc.2020.11.014 Abstract ( 209 )   PDF (1571KB) ( 298 )   Save For the current problems of vehicle traffic jam , difficulty in parking, and heavy pollution emissions in cities, a narrow vehicle roll method suitable for urban traffic is proposed. The steering wheel angle information and vehicle speed information are obtained in real time while the vehicle is driving, and the required roll angle when the vehicle is stable is calculated and the roll is carried out while steering so as to prevent the vehicle from rollover in the process of steering. In this paper, an active roll vehicle using with a new roll mechanism is designed and the vehicle roll angle parameter and roll execution strategy are analyzed. The active roll can be realized through the roll mechanism. The prototype S-track and O-track steering tests are carried out on outdoor roads for verification. The test results show that the lateral acceleration of the active roll prototype can be reduced by 95% during the steering process, effectively preventing the rollover danger caused by the centrifugal force when the vehicle is turning, verifying the feasibility of the active roll method and active roll mechanism scheme References | Related Articles | Metrics

Select

Study on the Enhancement of the Crash Performance of aPLI Legform in a Vehicle Model Hou Song, Luo Kun, Xie Bin, Wu Changpeng 2020, 42 (11):  1558-1565.  doi: 10.19562/j.chinasae.qcgc.2020.11.015 Abstract ( 412 )   PDF (1773KB) ( 297 )   Save In view of the application difficulties of the brand-new aPLI legform, a relevant application research is carried out based on the pedestrian protection development of a vehicle model. Firstly, the parameters of Flex-PLI and aPLI are compared to analyze the characteristics and injury mechanism of aPLI legform. Then, through benchmarking the vehicle models with the same platform, the deviation between simulation and test results of injury is contained within 10%, laying a foundation for the modification and improvement of the vehicle model. Finally, aiming at the problem of the low scoring rate of aPLI legform, three modification schemes are proposed, including adding energy absorbing boxes on auxiliary bumper, adjusting the stiffnesses of the main and auxiliary energy absorbing boxes, and adding the front cross beam on bonnet, to achieve the matching of the stiffnesses of vehicle front-end, resulting in a significant improvement of pedestrian protection performance of the vehicle model. References | Related Articles | Metrics

Select

Noise Source Identification of Vehicle Synchronous Belt Based on Sound Array Technology Shi Yaochen, Chen Guoping, Li Zhanguo, Tang Wusheng, Zhao Xilu 2020, 42 (11):  1566-1570.  doi: 10.19562/j.chinasae.qcgc.2020.11.016 Abstract ( 151 )   PDF (1308KB) ( 291 )   Save Through the analysis on the noise generation mechanism of vehicle synchronous belt transmission,it is pointed out that the main noise sources of vehicle synchronous belt transmission are meshing impact noise and transverse vibration noise. A transmission noise test device for ZA type synchronous belt is designed based on the principle of sound array measurement.Noise tests are conducted at different speeds and the sound pressure nephogram and the frequency characteristic curve of noise are obtained,with the noise generation mechanism and sound source location of synchronous belt verified.The test results show that the noise source identification of vehicle synchronous belt transmission by using sound array technology is accurate and reliable.This study provides a theoretical basis for noise control of vehicle synchronous belt transmission. References | Related Articles | Metrics

Select

Design and Analysis of a Novel Shifting Synchronization Mechanism Based on Torsional Vibration Attenuation Luo Yutao, Cheng Zhihao, Xu Xiaotong 2020, 42 (11):  1571-1576.  doi: 10.19562/j.chinasae.qcgc.2020.11.017 Abstract ( 147 )   PDF (1092KB) ( 308 )   Save In view of the defects of large impact and prone-to-wear of engaging sleeve and gear ring during the shifting of automated manual transmission in an electric vehicle, a kind of torsional vibration coupler with torsional vibration attenuation function is proposed, dedicated to shifting synchronization. With vehicle jerk, shifting engagement time and sliding-friction work as indicators, a comparative simulation is conducted on torsional vibration coupler, inertial synchronizerand thesynchronization scheme of motor-speed-regulation withoutsynchronizerby using software Adams. The results show that torsional vibration coupler is optimum in all indicators of synchronizing engagement except that it is inferior to inertial synchronizer in shifting engagement time References | Related Articles | Metrics

Select

Optimization of Vehicle Aerodynamic Drag Based on Discrete Adjoint Method and Surrogate Model He Yansong, Cao Lipeng, Zhang Zhifei, Li Yun, Chen Zhao 2020, 42 (11):  1577-1584.  doi: 10.19562/j.chinasae.qcgc.2020.11.018 Abstract ( 182 )   PDF (1574KB) ( 229 )   Save The discrete adjoint method and the surrogate model are introduced to propose a fast and effective optimization strategy for reducing vehicle aerodynamic drag in this paper. Firstly, through the aerodynamic modelling and simulation analysis on vehicle, the changing tendency of drag coefficient and measured point pressure are obtained, which are then compared with the results of wind tunnel test. They agree well, showing that the accuracy of model meets requirements. Then with drag coefficient as objective, the discrete adjoint method is adopted to conduct sensitivity analysis on vehicle surface, and the components with higher sensitivity such as front bumper, rear-view mirror, tail wing and rear bumper are determined to be the objects of optimization. Next, the sample space is constructed by Hammersley design of experiment and the sample point model are parameterized by using mesh free form deformation technology with the corresponding drag coefficients calculated. Finally, the surrogate model is constructed by Kriging interpolation method, and the multi-island genetic algorithm is selected to perform a global optimization on surrogate model. The results of optimization indicate that the drag coefficient of vehicle reduces by 3.29%. References | Related Articles | Metrics

Select

Fault Diagnosis of Position Sensor for Synchronizer Actuator Motor in Dry Dual Clutch Transmission Zhao Zhiguo, Fu Shihao, Jiang Siwen, Chen Jiayi 2020, 42 (11):  1585-1594.  doi: 10.19562/j.chinasae.qcgc.2020.11.019 Abstract ( 223 )   PDF (1498KB) ( 294 )   Save In view of that the Hall position sensor of the synchronizer actuator motor in a dry dual clutch transmission (DCT)is susceptible to environment and is inconvenient in maintenance, the fault diagnosis and fault-tolerant control of the Hall position sensorare performed in the gear-selecting process of dry DCT in this paper for ensuring the smooth shifting and safe driving of DCT vehicle. Firstly, the simulation model for the actuator motor of dry DCT synchronizer is built and verified,fuzzy sliding mode observer is designed to estimate its line back electromotive force (EMF) real-time, and the relationship between the line back EMF in brushless DC motor and rotational angle signal is utilized to achieve the indirect estimation of the rotor position of synchronizer actuatormotor. Then, based on the difference between the estimated value and the actual sensor output, the Hall position sensor fault is diagnosed, and the active fault-tolerant control of the Hall sensor fault of synchronizer actuatormotor is realized by means of analytical redundancy. Finally, simulation and verification test are conducted on the fault diagnosis method proposed for Hall position sensor, with a result showing that the delay of the reconstructed Hall position signal is less than 5.5%, and the method is applicable to the fault diagnosis and corresponding fault-tolerance control of the Hall sensor in synchronizer actuator motor References | Related Articles | Metrics