In order to realize the optimization of the intrinsic characteristics of the powertrain mounting system and its robustness evaluation, this paper proposed a sensitivity analysis method based on variance to find the key parameters that have a significant impact on the system performance. A multi-objective optimization model of the mounting system is constructed, which solves to obtain a reasonable deterministic optimization scheme. Meanwhile, in order to evaluate the robustness of the intrinsic characteristics of the mounting system under the uncertainty variation of the key parameters, the skewness and kurtosis are introduced, and the robustness evaluation is conducted by combining the mean value and standard deviation, and the Monte Carlo simulation method based on Latin Hypercube Sampling is used to analyze the influence of the uncertainty factors of the mounting stiffness on the robustness of the intrinsic characteristics of the mounting system. The results show that the variance-based sensitivity analysis method can not only effectively optimize the intrinsic characteristics of the mounting system and improve the optimization efficiency, but also more accurately evaluate and enhance the robustness of the system by comprehensively considering the sensitivity of the mounting stiffness.

In order to efficiently realize the measurement of vehicle wheel brows, a non-contact vehicle wheel fender clearance measurement system is designed based on image processing technology, and corresponding measurement software is created with the Graphical User Interface (GUI) of MATLAB. The verification results show that for actual clearance of at least 15 mm, the error of the wheel fender clearance calculated by the system under different shooting distance conditions is controlled within 2%, which exhibiting a high degree of measurement accuracy.

In order to improve the vehicle driving range of a pure electric light commercial vehicle, this paper proposes to reduce the electric power consumption per 100 kilometers by reducing power consumption and increasing coasting energy recovery power. Firstly, for ECO energy-saving mode, the paper proposes the energy-saving control technology based on different vehicle usage scenarios and working conditions, which significantly reduces the energy consumption of the vehicle in the whole scenario. Secondly, in the view of the coasting energy recovery deceleration of commercial vehicles with empty and full load varies greatly, the paper proposes the energy recovery control based on the vehicle mass to achieve the same coasting deceleration under different loads to recover more energy. Then, for the problem that coasting energy recovery is usually designed with low adhesion coefficient road surface, the paper proposes adaptive adjustment of coasting energy recovery based on road adhesion coefficient. Finally, the pure electric light truck is tested on real road. The results show that the ECO mode reduces the electricity consumption per 100 kilometers by 2.9%, and by 2.1% for scenarios of turning on the air conditioning and opening the car door. The energy recovery strategy based on the vehicle’s mass shows a 2.7% reduction in electricity consumption per 100 kilometers.

Based on the study of the three parameter control objectives of Pedal Map, a calibration control program for Pedal Map of electric vehicle is designed, and effective control of three parameter objectives are realized through MATLAB traversal interpolation. Simulation calculations and real vehicle testing results verify that the control objective based on the three parameters can effectively achieve the calibration and control of the pedal characteristics of electric vehicle. The experimental and simulation comparison results of the longitudinal acceleration curve show that the control program can accurately achieve the calibration control objective requirements and successfully achieve automatic calibration.

In order to realize automatic test and effectiveness verification of Automatic Mechanical Transmission (AMT), an automatic testing sequence is constructed based on the Hardware-in-the-Loop (HiL) testing system and automated testing software for the low gear shifting stuck fault of AMT front auxiliary gearbox. The fault diagnosis response is tested, and the test results show that the fault diagnosis response meets expectations, verifying the accuracy of the test sequence and fault diagnosis. Based on the test sequence, the normal gear shifting at high speed, gear shifting without fault and the low gear shifting of the front auxiliary gearbox under fault response are established. Test results show that the Transmission Control Unit (TCU) recognizes and responds in time when the front auxiliary gearbox low gear shift stuck fault occurs. The response is in line with expectations, and it also can keep the vehicle running normally under the fault diagnosis response, which verifies the effectiveness of the fault diagnosis response.

The steering system produces stick-slip vibration and noise due to the relative rotational friction between various components, resulting in unsmooth steering. To solve this problem, this paper studies the practical problems of stick-slip caused by worm gear deceleration mechanism of steering gear. The source of the problems are identified through tests, and the relevant influencing factors are determined by establishing a theoretical force model and CAE simulation analysis model. The torque anti-drop function MSP (Mechanical Stuck Prevention) of the steering gear is developed simultaneously. The software functional architecture and module interfaces are designed and the activation parameters are set through the debugging and optimization of the software functions. Finally, the output square wave signal of the motor torque 0.03 N·m is locked, which satisfies the software real-time control requirement, and improves the handing comfort of the vehicle.

In order to improve the accuracy, stability and flexibility of the vehicle chassis geometric parameter detection system, a vehicle chassis geometric parameter measurement system consisting of LED type 3D surface measurement sensors is designed based on stereo photogrammetry, using a structured LED light source to form a “spatial measurement area” based on the entire wheel side surface and wheel cover edge. By using 3D photography technology to obtain point cloud data under dynamic measurement, and combining with x-line software, a 3D wheel hub model is established to analyze and calculate the front toe camber angle of the chassis, the linearity and repeatability of the measurement system. The results show that linearity error of this measurement system is 1.528', repeatability error of front toe angle and camber angle are ±0.03' and ±0.05' respectively, verifying stability and reliability of the measurement system.

In order to analyze the influence of random and impulse road surface on the smoothness of commercial vehicles, a commercial vehicle is taken as the research object to construct the vehicle dynamics model containing front and rear suspension, tire, body, steering system and human body-seat system by using the ADAMS/Car software. The smoothness analysis of impulse and random input is carried out for the vehicle, and simulation results show that the smoothness of the vehicle increases after increasing the vehicle speed under the excitation of pulse road surface; the effect of smoothness is more significant on Class C road compared with Class B road under the random road surface input condition, which is more effective in the evaluation and improvement of driving comfort of commercial vehicles.

To tackle the challenge of exhaust gas leak detection, this paper proposes an exhaust gas leakage detection method based on distributed fiber optic vibration sensing technology. Firstly, sensors are mounted on the vehicle’s exhaust pipe to detect variations in both intensity and phase of the optical fiber light signal, which arise from local pressure disturbances associated with exhaust gas leaks. A distributed vibration sensing system that utilizes a dual Mach-Zehnder interferometer setup is applied with a differential cross-correlation algorithm for demodulating vibrations captured at both ends. Secondly, a cross-correlation delay algorithm is used to precisely pinpoint any leakage locations. In simulation test, leak scenarios are emulated using a piezoelectric ceramic device along with a designated fiber length of 2 m, achieving remarkable positioning accuracy of 0.01 m and validating the feasibility of this proposed method.

The rainwater pollution on vehicle side window affects driver’s safety critically, it is necessary to study this kind of water management situations. By improving the key parameters of windshield water bar, and in combination of vehicle climate wind tunnel test results of different windshield water bars of a SUV, this article quantiataively analyzes water pollution area of the front side window and the exterior rear view mirror area by image processing technology, and proposes an evaluation method considering the lateral and rearview field comprehensively. It is found that increasing the height of the windshield water bar and the water storage area can reduce the water flow to the side window and weaken the water pollution of the side window. Finally, a water bar scheme for the front wind window with better visual field and engineering feasibility is obtained.