Table of Content

25 April 2023, Volume 45 Issue 4 Previous Issue    Next Issue

Select

Review of Autonomous Driving Decision-Making Research Based on Reinforcement Learning Lisheng Jin,Guangde Han,Xianyi Xie,Baicang Guo,Guofeng Liu,Wentao Zhu 2023, 45 (4):  527-540.  doi: 10.19562/j.chinasae.qcgc.2023.04.001 Abstract ( 535 )   HTML ( 54 )   PDF (1155KB) ( 648 )   Save Decision-making technology of autonomous vehicle is promoted by the development of reinforcement learning， and intelligent decision-making technology has become a key issue of high concern in the field of autonomous driving. Taking the development of reinforcement learning algorithm as the main line in this paper， the in-depth application of this algorithm in the field of single-car autonomous driving decision-making is summarized. Traditional reinforcement learning algorithms， classic algorithms and frontier algorithms are summarized and compared from the aspect of basic principles and theoretical modeling methods. According to the classification of autonomous driving decision-making methods in different scenarios， the impact of environmental state observability on modeling is analyzed， and the application technology routes of typical reinforcement learning algorithms at different levels are emphasized. The research prospects for the autonomous driving decision-making method are proposed in order to provide a useful reference for the research of autonomous driving decision-making. Figures and Tables | References | Related Articles | Metrics

Select

Efficient Automatic Driving Instance Segmentation Method Based on Detection Yanyan Chen,Hai Wang,Yingfeng Cai,Long Chen,Yicheng Li 2023, 45 (4):  541-550.  doi: 10.19562/j.chinasae.qcgc.2023.04.002 Abstract ( 198 )   HTML ( 12 )   PDF (4730KB) ( 210 )   Save The instance segmentation algorithm based on deep learning has achieved excellent performance in large-scale general scenarios. However， the segmentation of multi-objective instances for complex traffic scenes is still challenging， especially in the balance between high accuracy and fast inference speed， which is crucial to driving safety of intelligent vehicles. In view of this， based on the real-time algorithm Orienmask， a multi-head segmentation framework is proposed based on the one-stage detection method. Specifically， the proposed framework comprises of a backbone， a feature fusion module and a multi-head mask construction module. Firstly， complete high-dimensional feature maps are obtained by adding residual structures to the backbone.Secondly， in order to generate discriminative feature representations， the feature pyramid module is reconstructed by introducing in self-calibrate convolutions and the information propagation path is improved by global attention mechanism， so as to further optimize the feature fusion module of the proposed framework. Finally， a multi-head mask construction mechanism is proposed to significantly improve the segmentation performance of different targets by refining the size distribution of instances in the traffic scenes. The proposed algorithm has been tested and validated on the open-source dataset BDD100k， and has achieved an average intersection ratio of 23.3% and 19.4% （mAP@0.5：0.95） on bounding boxes and segmentation masks， respectively. Compared with the baseline， the average index are increased by 5.2 % and 2.2 %. At the same time， the road experiment on the self-built real-vehicle platform also proves that the proposed algorithm can adapt to actual driving environments and meet the demands of real-time segmentations. Figures and Tables | References | Related Articles | Metrics

Select

Study on Eco-driving of PHEVS Based on Hierarchical Control Strategy Yapeng Li,Xiaolin Tang,Xiaosong Hu 2023, 45 (4):  551-560.  doi: 10.19562/j.chinasae.qcgc.2023.04.003 Abstract ( 194 )   HTML ( 17 )   PDF (5400KB) ( 254 )   Save The development of intelligent transportation system technology provides a great opportunity to further improve the driving performance of automotive vehicles. The eco-driving of plug-in hybrid electric vehicles （PHEV） involves three issues， namely， how to use dynamic traffic information for longitudinal driving speed planning， optimal rapid planning of global battery state of charge （SOC）， and the real-time energy management of the power system. This paper devises a hierarchical control strategy that combines the accuracy model with both calculation efficiency and solution accuracy to solve these problems. In the upper control layer， dynamic traffic light signal information is incorporated into the velocity optimization process to improve driving comfort. In the middle control layer， the SOC fast global optimal planning is realized based on the convex optimization by fitting the powertrain model. Finally， in order to eliminate the error caused by the fitting model， based on the original nonlinear model， an adaptive equivalent consumption minimization strategy （A-ECMS） is established in the lower control layer through feedback control. The results show that the driving comfort is improved by 75.92% compared with the strategy without optimization in velocity， and the fuel economy is improved by 7.39% and 10.91% respectively compared with that of two often used linear programming-based energy management strategies （EMSs）. Figures and Tables | References | Related Articles | Metrics

Select

Control Strategy of Assistant Driving for Post-impact Based on Optimal Method Bing Zhou,Jiabao Wei,Tian Chai,Xiaojian Wu,He Wang 2023, 45 (4):  561-571.  doi: 10.19562/j.chinasae.qcgc.2023.04.004 Abstract ( 117 )   HTML ( 9 )   PDF (6485KB) ( 154 )   Save In a collision accident， violent yaw and lateral motion often occur at the same time after the side impact of a vehicle. It is difficult for ordinary drivers to correctly deal with such emergency situation， obviously the vehicle may lose stability， or even cause more serious accidents. In order to reduce the secondary collision caused by vehicle instability and the misoperation of the driver， this paper proposes a two-stage auxiliary driving control strategy that takes over the vehicle driving authority after the collision. In the first stage， the cost function is designed by integrating the two indicators of vehicle stability and lateral displacement. Through hierarchical control， the vehicle can quickly return to stability after collision and reduce the lateral displacement to reduce the risk of secondary collision. After that， according to the stability region divided by the phase plane method， a switching criterion for the control system is formulated. After judging that the vehicle state is stable， the control system switches to the path tracking control in the second stage. The model predictive path tracking controller will drive the vehicle back to the original lane to reduce the impact on the adjacent lane. Finally， the effectiveness of the control strategy proposed in this paper is verified through simulation experiments in different strength side impact. Figures and Tables | References | Related Articles | Metrics

Select

Study on Improved Point Cloud Registration Algorithm Enhanced by Double Down-sampling Zhongsheng Chen,Chaolin Li,Wang Zuo,Xinglin Hou 2023, 45 (4):  572-578.  doi: 10.19562/j.chinasae.qcgc.2023.04.005 Abstract ( 80 )   HTML ( 3 )   PDF (976KB) ( 94 )   Save In the field of autonomous driving， the real-time requirement of point cloud registration is high. Existing ICP algorithm and its variants have such problems as high requirements for initial pose and slow registration speed. In order to deal with the above-mentioned problems， an improved fast point cloud registration algorithm is proposed in this paper. Firstly， a double down-sampling method is used to preprocess point cloud data. By this way， the amount of point cloud data can be greatly reduced rapidly， while retaining the original features. Then the Intrinsic Shape Signature （ISS） is introduced to optimize the Super 4-Points Congruent Sets （Super4PCS） algorithm to reduce its time complexity. Finally， the linear least squares optimization-based iterative closest point （ICP） algorithm is used for fast and precise registration. The effectiveness of the algorithm is tested and compared by using the Stanford and autonomous driving Kitti point cloud data. The results show that the proposed algorithm has good robustness， and the registration accuracy and speed are significantly increased compared with the existing algorithms. Figures and Tables | References | Related Articles | Metrics

Select

Research on Trajectory Tracking Control of Hybrid Tracked Unmanned Platform Bin Zhang,Yuan Zou,Xudong Zhang,Fengchun Sun,Zhe Wu,Yihao Meng 2023, 45 (4):  579-587.  doi: 10.19562/j.chinasae.qcgc.2023.04.006 Abstract ( 142 )   HTML ( 7 )   PDF (5229KB) ( 151 )   Save To improve the trajectory tracking performance of tracked unmanned platform， a hierarchical trajectory tracking algorithm considering longitudinal speed planning is proposed and verified by co-simulation experiment and real vehicle experiment. Based on the establishment of the vehicle differential equation including the slip ratio of the track and the sideslip angle of the center of mass， the hierarchical trajectory tracking algorithm framework is constructed. Speed planning algorithm of the upper level based on the pseudo spectrum method plans the longitudinal speed according to the road information， and sends the planned speed as the target speed to the lower layer trajectory tracking algorithm based on linear time-varying model predictive control （LTV-MPC）. By establishment of the prediction model and constraints， the algorithm based on LTV-MPC solves the target speed of motors on both sides through quadratic programming. Through the real vehicle experiment and the co-simulation of MATLAB/Simulink with RecurDyn， it is verified that the proposed algorithm has good trajectory tracking effect under different ground conditions. Figures and Tables | References | Related Articles | Metrics

Select

Intelligent Vehicle Driving Risk Assessment Method Based on Trajectory Prediction Xiang Gao,Long Chen,Xinye Wang,Xiaoxia Xiong,Yicheng Li,Yuexia Chen 2023, 45 (4):  588-597.  doi: 10.19562/j.chinasae.qcgc.2023.04.007 Abstract ( 215 )   HTML ( 7 )   PDF (3026KB) ( 232 )   Save This paper proposes a driving risk assessment method based on trajectory prediction. Firstly， a driver’s risk field （DRF） with Gaussian cross-section characteristics along both sides of the prediction trajectory is established to characterize the uncertainty of the driver’s behavior. Then， taking the risk consequences of the vehicle and the surrounding static and dynamic obstacles in specific states into consideration， the environmental event cost is established， and the quantitative perception risk that adapts to the uncertainty of complex driving scenarios is obtained. Finally， the quantitative perception risk time series in the prediction interval is then fused based on Bayesian theory to realize prediction of potential collision risks in future driving. The real vehicle trajectory and simulation results show that compared with the classic TTC index method， the risk assessment method of DRF based on the integration of interaction information between self-vehicle and surrounding environment in the future can identify the driving risk changes of complex traffic scenarios faster and more accurately， which provides a reference for the study of vehicle collision risk problems in complex scenarios with multiple surrounding vehicles. Figures and Tables | References | Related Articles | Metrics

Select

Distributed Charging Control of Electric Vehicles Considering Distribution Grid Load Zhongqiang Wu,Changxing Zhang 2023, 45 (4):  598-608.  doi: 10.19562/j.chinasae.qcgc.2023.04.008 Abstract ( 140 )   HTML ( 5 )   PDF (2364KB) ( 117 )   Save To address the problem that large-scale plug-in electric vehicle access poses a threat to the safe operation of the distribution grid， a distributed consistent optimal charging management scheme for electric vehicles considering the load of the distribution grid is proposed. The scheme adopts a two-tier optimization structure. The upper layer establishes a mathematical model with the objectives of reducing the distribution grid load fluctuation and satisfying the power demand of charging stations， and uses a particle swarm optimization algorithm to solve the power allocated by the grid to charging stations in each time period. The lower layer establishes a mathematical model with the objectives of maximizing the overall user benefit and reducing the charging current fluctuation in the case of short supply of power of charging stations. A distributed charging protocol based on multi-agent is proposed to solve for the optimal charging power of electric vehicles. The protocol is distributed， plug-and-play for electric vehicles， and has good robustness and scalability. The scheme ensures the safe operation of the distribution network and improves the overall customer satisfaction. The feasibility of the scheme is verified by simulation. Figures and Tables | References | Related Articles | Metrics

Select

Optimization of Temperature Model in Axial Flux Motor Based on Genetic Algorithm for EVs Zhaozong Li,Shuo Zhang,Chengning Zhang 2023, 45 (4):  609-618.  doi: 10.19562/j.chinasae.qcgc.2023.04.009 Abstract ( 90 )   HTML ( 4 )   PDF (3802KB) ( 113 )   Save In recent years， segmented armature axial flux motors have been widely used in the field of electric vehicles with the high torque density and compact axial size. However， due to the complex material composition of the contact area between segmented armatures and cooling fins， and the difficulties in determining the pressure at each position， the thermal conductivity of this region is always the difficulty of temperature prediction for such motors. For the heat transfer behavior of non-ideal contact surface， a research method of building a weighted model based on the three-dimensional thermal resistance grid model is proposed in this paper to fine-tune the unknown thermal conductivity. Firstly， the topology of the prototype is introduced， and the thermal resistance grid model and the weighted model framework of the segmented armature single sector are established. The unknown thermal conductivity in the weighted model is trained by genetic algorithm， and the model is used to replace the traditional single-sector thermal resistance grid model of the motor. Finally， the method is verified by the experimental bench of the prototype motor. Figures and Tables | References | Related Articles | Metrics

Select

Switching Functional Hybrid Control Strategy for Permanent Magnet Synchronous Motor of Electric Vehicle Zhiheng Wu,Aimin Liu 2023, 45 (4):  619-627.  doi: 10.19562/j.chinasae.qcgc.2023.04.010 Abstract ( 97 )   HTML ( 7 )   PDF (2456KB) ( 139 )   Save In order to improve the working efficiency of the driving system of electric vehicles equipped with permanent magnet synchronous motor （PMSM）， enhance the stationarity and response speed of the motion process， so as to improve the overall dynamic control performance of the driving system of electric vehicles， in this paper， switching functional hybrid control technology is proposed and designed based on the analysis of electric motor operating characteristics and feedback control principle derivation. The control technology can effectively improve the dynamic and static performance and robustness of the vehicle motor control system. In order to verify the effectiveness of the proposed control technology， a simulation model is established for simulation and analysis. The experimental platform is built for experimental verification. The simulation and experimental results show that the proposed control technology has the advantages of fast output response， no overshooting or oscillation， which can improve the efficiency of the motor， optimize the output characteristics of its drive system， and enhance the control performance of the electric vehicle drive system. Figures and Tables | References | Related Articles | Metrics

Select

A Supercapacitor SOC Estimation Method Based on Weighted Fusion Considering Ambient Temperature Variation Chun Wang,Tao Tang,Yongzhi Zhang 2023, 45 (4):  627-636.  doi: 10.19562/j.chinasae.qcgc.2023.04.011 Abstract ( 91 )   HTML ( 6 )   PDF (4009KB) ( 131 )   Save Accurate estimation of the state of charge （SOC） of supercapacitors plays an important role in electric vehicle hybrid energy storage system， which directly determines the starting， climbing and accelerating performance of electric vehicles. Therefore， this paper proposes a supercapacitor SOC estimation method based on fuzzy entropy weighted fusion. Firstly， the Thevenin model parameters are identified by using the particle swarm algorithm under -10， 10， 25 and 40 ℃， and the nearest neighbor method is adopted to establish the mapping relation between the parameters and temperatures. Then， the fuzzy entropy is utilized to design a SOC weighted fusion estimation method based on three typical Kalman filters. Finally， the SOC estimation method of adaptive weighted averaging and residual normalized weighted fusion is selected to further evaluate the performance of the proposed method in this paper. The results show that supercapacitor SOC estimation method based on fuzzy entropy weighted fusion can improve the supercapacitor SOC estimation accuracy， especially in high ambient temperature environment （40 ℃） . Figures and Tables | References | Related Articles | Metrics

Select

Study on Stability Control Strategy of Heavy Vehicles Based on Phase Space Three-Dimensional Dynamic Stability Domain Yuhang Kang,Shaohua Li,Zekun Yang 2023, 45 (4):  637-646.  doi: 10.19562/j.chinasae.qcgc.2023.04.012 Abstract ( 143 )   HTML ( 11 )   PDF (4191KB) ( 195 )   Save Heavy-duty vehicles have the characteristics of large moment of inertia and slow control response. For heavy-duty commercial vehicles， a three-dimensional phase space analysis method is designed based on the sideslip angle， yaw rate and vertical load transfer coefficient to judge the real-time stability state of vehicles. For different vehicle driving conditions， AFS control and AFS /DYC hierarchical control are adopted， and the extension control method is designed based on adhesion coefficient to compensate the control output of front wheel angle and yaw moment， so as to ensure the robustness of the controller under different working conditions. The validity of the proposed method is verified by TruckSim /Simulink co-simulation and hardware in the loop experiment， with the superiority of the dynamic stability domain in phase space demonstrated in the simulation by the comparison of \beta -\dot{\psi }-LTR judgment and \beta -\dot{\psi } judgment. The simulation and experimental results show that compared with the way tooperate the vehicle only according to the driver’s intention， the designed AFS/DYC control strategy based on Extension H_{\mathrm{\infty }} method can ensure better stability of the vehicle under different working conditions， especially on the low adhesion road， and thus effectively reduce the probability of traffic accidents under extreme working conditions. Figures and Tables | References | Related Articles | Metrics

Select

Research on Integrated Design of Battery Pack and Car Body Based on Torsional Stiffness Yubo Lian,Bengang Yi,Yingying Cui,Hongsheng Tian,Junfei Yan,Chen Cheng 2023, 45 (4):  647-653.  doi: 10.19562/j.chinasae.qcgc.2023.ep.003 Abstract ( 189 )   HTML ( 17 )   PDF (3062KB) ( 377 )   Save The torsional stiffness of the body-in-white is an important mechanical performance index of the load-bearing body， which has a direct impact on the handling stability of the vehicle， and is also an important index to evaluate the lightweight level of the body. Since the new generation of pure electric body has a large structural frame difference from the traditional body， in the early stage of body design， the force transmission path is redefined with the goal of improving torsional stiffness. In this paper， based on theoretical analysis and topology optimization， the optimal force transmission path of the torsional stiffness of the body is found. Through the integrated design of the battery pack and the body， the body forms a plurality of closed structures that are close to a circular ring. On the premise of no extra weight adding to the car body t， the torsional stiffness of the body-in-white can reach 40，000 N·m/（°）， and the lightweight coefficient of the body can reach 1.75， achieving the leading level of pure electric vehicles， and at the same time greatly improving the handling stability of the vehicle. Figures and Tables | References | Related Articles | Metrics

Select

Influence of Valve Overlap Angle on GDI Engine Performance and Particulate Emission Xiaona Li,Fangxi Xie,Huili Dou,Jiangwei Liu,Yu Liu 2023, 45 (4):  654-662.  doi: 10.19562/j.chinasae.qcgc.2023.04.014 Abstract ( 85 )   HTML ( 6 )   PDF (5095KB) ( 97 )   Save In this paper， the influence of valve overlap angle formed by variable intake timing （IVT）， variable exhaust timing （EVT） and variable intake and exhaust timing strategy （IEVT） on combustion and particulate emission of the direct injection gasoline engine is studied. It is found that the increase of positive valve overlap angle under the three timing strategies under low load will lead to the increase of residual exhaust gas in the cylinder，prolonged ignition delay and increased combustion duration， and reduce fuel consumption and HC emissions at first and then increase， and reduce NO x emission. At the same valve overlap angle， EVT has the largest amount of residual exhaust gas， while IVT can improve the pumping loss by 15.6%. Compared with IVT and EVT， IEVT still burns stably at the overlapping angle of 60 ° CA and reduces heat transfer loss and exhaust loss， with reduction offuel consumption by 8.67%， reduction of NO x by 96.57%， and reduction of the total number of particles by 89.43%. Figures and Tables | References | Related Articles | Metrics

Select

Research on NO x Emission Control Method of Real Driving Based on Durable Aging Vehicle Shenglong Xu,Jun Song,Wei Yuan 2023, 45 (4):  663-671.  doi: 10.19562/j.chinasae.qcgc.2023.04.015 Abstract ( 95 )   HTML ( 6 )   PDF (5619KB) ( 136 )   Save In order to meet the RDE （real driving emission） sampling requirements of 160，000 km in-use vehicles stipulated in China 6 emission regulations， the aggressive RDE test cycle on test bench is carried out on two durable aging vehicles under the critical environment of 1 ℃. By adjusting the engine control strategies， the operation parameters of two extreme working conditions with high NO x emission are optimized： rapid acceleration after cold start and rapid acceleration to super high vehicle speed after hot start， then the comparative verification tests between the new and old strategies in variable emission combinations include roller cycles and RDE are carried out. The results show that the VVT， excess scavenging coefficient， lambda target value and aging catalyst window control have great impact on NO x emission of durable aging vehicle under cold and hot engine’s high load conditions. Appropriate control strategies can reduce the total NO x emission by more than 40%. It is an effective RDE development method to firstly carry out the emission development based on durable aging vehicles under aggressive RDE cycle， and then validate it on the real road. Figures and Tables | References | Related Articles | Metrics

Select

Influence of Entrainment Effect of Two-Phase Flow on Combustion Mode in Diesel Engine Miao Yang,Xuedong Lin,Degang Li,Yingshu Liu 2023, 45 (4):  672-680.  doi: 10.19562/j.chinasae.qcgc.2023.04.016 Abstract ( 90 )   HTML ( 4 )   PDF (4726KB) ( 77 )   Save Premixed combustion and diffusion combustion are the basic combustion modes of diesel engines. In this paper， the two typical combustion chambers of the reentrant combustion chamber and flared combustion chamber of high-pressure common-rail direct injection diesel engine are selected to study the mixture formation mechanism based on the entrainment effect under different background airflow conditions by simulation calculation. The results show that the entrainment effect caused by the small-scale vortex on the surface of the oil spray is the main reason for the formation of the pre-mixture when the oil spray penetrates the background air flow at the initial stage of injection. The entrainment effect is directly affected by different background airflow under certain injection condition. For the reentrant combustion chamber， at the initial stage of injection， there is strong entrainment effect and a fast premixing rate. Most of the fuel is rapidly diffused and combusted under the action of strong tumble inside the combustion chamber. Therefore， it belongs to the premixed diffusion combustion mode （PDC）. Although the proportion of premixed combustion is small， the high-temperature zone is wide. For the flared combustion chamber， although there is relatively weak entrainment effect and a low premixed combustion rate at the initial stage of injection， a secondary premixing process is formed after the oil spray impinges on the wall convex platform， which extends the formation duration of the pre-mixture and increases the proportion of premixed combustion to form a double-premixed diffusion combustion mode （DPDC）. Through the DPDC combustion mode， the high-temperature area is reduced， thus effectively inhibiting the generation of NO x . Figures and Tables | References | Related Articles | Metrics

Select

Noise Reduction Design and Wind Tunnel Test Verification of an SUV Rearview Mirror Miaoyan Song,Guocheng Zhou,Hongqing Chen,Bao Chen 2023, 45 (4):  681-689.  doi: 10.19562/j.chinasae.qcgc.2023.04.017 Abstract ( 116 )   HTML ( 9 )   PDF (5722KB) ( 118 )   Save To investigate the effect of automobile rearview mirror arm on the aero-acoustics， based on a certain SUV automobile rearview mirror， the detached eddy simulation method is used to analyze the flow field and near field noise characteristics. By changing the shape of the arm， two kinds of noise-reduction models are designed， and the aerodynamic noise wind tunnel tests of three rearview mirror models are carried out in the FL-53 aero-acoustic wind tunnel of the Institute of Aerodynamics of the Aviation Industry. The results show that by changing the curvature around the mirror arm， the size of detached eddy is reduced， the direction of vortex shedding is changed， and the noise near the wake zone is reduced. Besides， the wind velocity can affect the noise reduction in some frequency bands. In terms of far-field directivity， the sound pressure level of the three rear-view mirrors in the wake flow field is larger， and the noise reduction models cannot change the far-field directivity. Figures and Tables | References | Related Articles | Metrics

Select

Gear Rattle Optimization of Balance Shaft of HEV Yuepu Liu,Ying Guan,Yan Zhang,Liang Zhao,Qiang Chen,Jingling Yang,Tuo Shen 2023, 45 (4):  690-698.  doi: 10.19562/j.chinasae.qcgc.2023.04.018 Abstract ( 262 )   HTML ( 9 )   PDF (11413KB) ( 205 )   Save In order to optimize the balance shaft gear rattle noise of the longitudinal hybrid vehicle, by comparing the structural differences with the transverse hybrid vehicle, it is identified that the increase of the inertia of the flywheel, the decease of the crankshaft mode, and the increase of the drive gear angular acceleration of the balance shaft lead to the gear rattle of the balance shaft. However, the conventional single-stage TVD matching presents the law of "one thing decreases and the other grows", which cannot cover the optimization objective of the angular acceleration of the engine at the full speed segment. Therefore, a two-stage TVD is developed to reduce the angular acceleration of the drive gear of the balance shaft, which can solve the problem of gear rattle. The Simpack software is used to build a multi-body dynamic model of gear rattle, which studies the mechanism of gear rattle generation and propagation of the balance shaft. Then a double anti-backlash balance shaft is developed to further optimize the noise of gear rattle by reducing the force impact on both sides of the gear teeth during meshing. The results show that the combined use of two-stage TVD and double anti-backlash balance shaft can solve the gear rattle problem, which has important engineering significance for the design of the balance shaft gear and the optimization of rattle problem. Figures and Tables | References | Related Articles | Metrics

Select

Analysis of Vehicle Diagnostic Trouble Codes Based on Association Rule Mining Jie Hu,Hao Geng,Yuanjie Li,Huangzheng Geng,Minmin Tong 2023, 45 (4):  699-708.  doi: 10.19562/j.chinasae.qcgc.2023.04.019 Abstract ( 119 )   HTML ( 5 )   PDF (2887KB) ( 133 )   Save Based on the principle of on-board diagnosis， this paper combines analysis of diagnostic trouble codes （DTCs） generated and saved during vehicle self-diagnosis with association rule mining， proposes an improved FP-tree algorithm suitable for exploring DTCs data association， and establishes the vehicle DTCs association diagram according to the obtained association rule， which is applied to the historical data analysis process and vehicle maintenance process. Interesting association rules in data is explored to provide association visualization results. The miscellaneous DTCs read in the maintenance process is analyzed to reduce the complexity of the DTCs and the main DTCs are analyzed to shorten the maintenance time based on the DTCs and help the maintenance personnel to locate the faults. Figures and Tables | References | Related Articles | Metrics