检索结果-内蒙古大学图书馆

Wgs-Yolo: A Real-Time Object Detector Based on Yolo Framework for Autonomous Driving

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Yue, Shiqin Zhang, Ziyi Shi, Ying Cai, Yonghua Hubei Key Laboratory of Advanced Technology for Automotive Components Wuhan University of Technology Wuhan430070 China Hubei Collaborative Innovation Center for Automotive Components Technology Wuhan University of Technology Wuhan430070 China Hubei Technology Research Center of New Energy and Intelligent Connected Vehicle Engineering Wuhan University of Technology Wuhan430070 China School of Automotive Engineering Wuhan University of Technology Wuhan430070 China School of Automation Wuhan University of Technology Wuhan430070 China

Ensuring the safety and reliability of autonomous driving is critically dependent on the precision and efficiency of object detection systems. This study introduces a WGS-YOLO model for autonomous driving object detection based on YOLO framework. This model is enhanced with a weighted efficient layer aggregation network (W-ELAN) and a grouped spatial pyramid pooling module (GSPP). W-ELAN represents a new attempt at efficient layer aggregation network based on channel Shuffle. Furthermore, the space-to-depth convolution (SPD-Conv) module is integrated to replace strided convolutions and mitigation oversampling issues. Results demonstrate the exceptional performance of the WGS-YOLO model on the BDD100k datasets, significantly enhancing the mean average precision from 54.43% to 66.43% compared to the baseline Yolov7, while achieving a detection speed of 69.93 frames per second. © 2024, The Authors. All rights reserved.

关键词： Object detection

Customer Perceived TCO and BEV Platform Development Approach

学校读者我要写书评

暂无评论

Customer Perceived TCO and BEV Platform Development Approach

作者： Tie-bing ZHANG Miao-hua HUANG School of Automotive Engineering Wuhan University of Technology Hubei Key Laboratory of Advanced Technology for Automotive Component Wuhan University of Technology Hubei Collaborative Innovation Center for Automotive Components Technology Wuhan University of Technology Hubei Research Center for New Energy and Intelligent Connected Vehicle Wuhan University of Technology

The attention to customer perception factors is helpful for automobile companies to better implement "customer-oriented" product development *** this paper, customer perception factors are introduced into TCO framework, and BEV TCO model based on customer perception is established. Compared with TCO model without comprehensive introduction of customer perception factors, comparison calculation shows that introduction of customer perception factors can help identify real customer needs more effectively and improve the pertinence of product *** on MATLAB GUI and '0-1' integer programming method, this paper builds a development tool for BEV platform based on customer perceived TCO, it can realize TCO target setting, analyse TCO optimization alternative schemes and propose decisionmaking suggestion for achieving TCO target. Using the development tool set up in this paper, TCO optimization is carried out on a vehicle model of BEV platform which is under development and achieved optimization goal, indicating that the development tool is beneficial to guide the development of BEV platform.

关键词：

Mechanical Properties of Horsetail Bio-Inspired Honeycombs Under Quasi-Static Axial Load

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Niu, Xiaoqiang Xu, Fengxiang Zou, Zhen Zhu, Yifan Duan, Libin Du, Zhanpeng Ma, Hongfeng Hubei Key Laboratory of Advanced Technology of Automotive Components Wuhan University of Technology Hubei Wuhan430070 China Hubei Collaborative Innovation Center for Automotive Components Technology Wuhan University of Technology Hubei Wuhan430070 China Institute of Lightweight and Safety of New Energy Vehicle School of Automotive and Traffic Engineering Jiangsu University Zhenjiang212013 China Jiangsu XCMG State Key Laboratory Technology Co. Ltd. Xuzhou221000 China

Inspired by the microstructures of horsetails, a series of novel bio-honeycombs are proposed by introducing the microstructure of horsetails into regular hexagonal honeycomb (RH) cells. The bio-honeycombs can be divided into two categories: one is that joint ribs of bio-honeycombs pass through the midpoint of the RH cell walls and are perpendicular to them (MP type). Another one is that joint ribs coincide with center lines of feature angles of the RH cell (CC type). To investigate out-of-plane crashworthiness of bio-honeycombs, finite element models of them are established. Preliminary findings suggest that introduced bionic units and the outer RH frame will produce an obvious interaction effect including severe plastic strain formed on intersections of them and more folding lobes formed on cell walls, which make a huge contribution to the improvement of their crashworthiness. Subsequently, the effect of the d-to-l ratio (d/l;ζ) and the t-to-l ratio (t/l) of MP_6 (MP type with six joint ribs) bio-honeycombs on their mechanical response are explored. Results demonstrate that smaller cell size of bio-honeycomb is conducive to improving the utilization rate of materials. Furthermore, t/l determines the deformation modes of MP_6 bio-honeycombs, and the strain hardening effect of materials will appear when ζ being in 0.030-0.056. Finally, theoretical expression is derived to estimate the mean crush force (MCF) of MP_6 bio-honeycombs, which conform that the theoretical solution is in good agreement with the numerical simulation. © 2023, The Authors. All rights reserved.

关键词： Finite element method

Crystal Structure, Cation Distribution and Electrical Properties in NixAlyMn3-x-yO4 (0 < X < 0.6;0 < Y < 0.3) Spinel-Based Ntc Chip Thermistors

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Shao, Gang-Qin Cen, Quan-Jin Chen, Zhi-Hua Zhang, Ji-Wei Xie, Meng-Xiang Li, Qiang Cao, Xiu-Hua Wang, Wen-Hao Wang, Yan-Pei Wu, Hou-Ran Hu, Chun-Yuan State Key Laboratory of Advanced Technology for Materials Synthesis and Processing Wuhan University of Technology 122 Luoshi Road Hubei Wuhan430070 China State Key Laboratory of Advanced Materials and Electronic Components Guangdong Fenghua Advanced Technology Holding Co. Ltd. 18 Fenghua Road Guangdong Zhaoqing526060 China

关键词： Manganites

Aerodynamic Performance of a Sedan Under Wind-Bridge-Tunnel Road Condition 5th

学校读者我要写书评

暂无评论

Aerodynamic Performance of a Sedan Under Wind-Bridge-Tunnel ...

5th Symposium on Fluid-Structure-Sound Interactions and Control, FSSIC 2019

作者： Zhang, Qianwen Su, Chuqi Wang, Yiping Hubei Key Laboratory of Advanced Technology for Automotive Components Wuhan University of Technology Wuhan430070 China Hubei Collaborative Innovation Center for Automotive Components Technology Wuhan University of Technology Wuhan430070 China Hubei Research Center for New Energy & Intelligent Connected Vehicle Wuhan University of Technology Wuhan430070 China

ISBN: (纸本)9789813349599

Due to the complexity of the wind environment around bridge-tunnel section in canyon, the flow field around vehicle is complex and changeable. In this paper, the processes of a sedan pulling out of a two-way four-lane tunnel and passing through the pylon area under canyon wind condition were simulated numerically using the dynamic mesh technique. The results showed that canyon wind and the terrain topography have a significant effect on the aerodynamic performance of the vehicle. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Bridge-tunnel section Canyon wind Unsteady aerodynamics

A Novel Gear Defect Detection Method Through Improved Yolov5 Network Using Attention Mechanism and Feature Fusion

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Zhang, Shuwen He, Minqi Zhong, Zhenyu Zhu, Dahu Hubei Key Laboratory of Advanced Technology for Automotive Components Wuhan University of Technology Wuhan430070 China Hubei Collaborative Innovation Center for Automotive Components Technology Wuhan University of Technology Wuhan430070 China

This work is motivated by the misdetected and undetected errors in metallic gear defect detection when the defect features are very close to the interferences of oil stains and image shadow. In this paper we present a novel gear surface defect detection method based on the improved YOLOv5 network model to solve this problem. The presented method combines the approach using the attention module CBAMC3 constituted by CBAM and C3 modules together with the analysis of the module BiFPN_concat, which makes the network capable of extracting and fusing defect features, respectively. The cosine annealing function is then employed to improve the learning ability of the network by modifying the learning rate. The experimental results indicate that the improved YOLOv5 network enhances both the recall rate by 13.1% and the mAP@0.5 by 12% based on the constructed gear defect dataset compared with the original YOLOv5 network. © 2022, The Authors. All rights reserved.

关键词： Feature extraction

Online estimation of state of health for lithium-ion batteries in electric vehicles based on voltage integral and temperature 7

学校读者我要写书评

暂无评论

Online estimation of state of health for lithium-ion batteri...

2021 7th International Conference on Manufacturing technology and Applied Materials, ICAMMT 2021

作者： Chen, Kailun Li, Qizhong Wang, Yujiu Huang, Miaohua Hubei Key Laboratory of Advanced Technology for Automotive Components Wuhan University of Technology Wuhan430070 China Hubei Collaborative Innovation Center for Automotive Components Technology Wuhan University of Technology Wuhan430070 China Hubei Research Center for New Energy and Intelligent Connected Vehicle Wuhan University of Technology Wuhan430070 China

In order to achieve accurate online State of health estimation of power battery, this paper proposes a method based on voltage integration and temperature. Based on the NASA ternary lithium battery accelerated life test, three groups of battery experimental data are divided according to the difference in temperature and discharge rate. According to the single battery data and battery experimental data required to record in the Chinese national standard documents, the feasibility of voltage integration as a characterization of battery health is verified. Then we discussed the charging and discharging behavior of electric vehicles and combine the Particle Swarm Optimization algorithm to select the voltage integral of the fixed voltage interval in the charging stage as the equivalent health factor of the battery health state, and establish a battery health state estimation model based on the Gaussian process regression with the temperature and voltage integral as input. The proposed method is verified by experimental data. The proposed method is suitable for online estimation of battery health status under different temperatures and different discharge rates. The root mean square error and average absolute error of battery health status estimation are both less than 1.4%. © Published under licence by IOP Publishing Ltd.

关键词： Charging (batteries)

A Novel Gear Defect Detection Method Through Improved Yolov5 Network Using Attention Mechanism and Feature Fusion

学校读者我要写书评

暂无评论

SSRN 2022年

This work is motivated by the misdetected and undetected errors in metallic gear defect detection when the defect features are very close to the interferences of oil stains and image shadow. In this paper we present a novel gear surface defect detection method based on the improved YOLOv5 network model to address this problem. The presented method combines the approach using the attention module CBAMC3 constituted by CBAM and C3 modules together with the analysis of the module BiFPN_concat, which makes the network capable of extracting and fusing defect features, respectively. The cosine annealing function is then employed to improve the learning ability of the network by modifying the learning rate. The experimental results indicate that the improved YOLOv5 network enhances both the recall rate by 13.1% and the mAP@0.5 by 12% based on the constructed gear defect dataset compared with the original YOLOv5 network. © 2022, The Authors. All rights reserved.

关键词： Feature extraction

High Purity Hydrogen Production by Metal Hydride System:A Parametric Study Based on the Lumped Parameter Model

学校读者我要写书评

暂无评论

Journal of Wuhan University of technology(Materials Science) 2021年第1期36卷 127-135页

作者： KOUA Koua Alain Jesus TONG Liang YANG Tianqi XIAO Jinsheng School of Materials Science and Engineering Wuhan University of TechnologyWuhan 430070China Hubei Key Laboratory of Advanced Technology for Automotive Components Hubei Collaborative Innovation Center for Automotive Components Technology and Hubei Research Center for New Energy&Intelligent Connected VehicleWuhan University of TechnologyWuhan 430063China Reliability Engineering Institute School of Energy and Power EngineeringWuhan University of TechnologyWuhan 430070China Hydrogen Research Institute Universitédu QuébecàTrois-RivièresQC G9A 5H7Canada

The simulation of hydrogen purification in a mixture gas of hydrogen/carbon dioxide (H2/CO2) by metal hydride system was *** lumped parameter model was developed and *** validated model was implemented on the software Matlab/Simulink to simulate the present *** simulation results demonstrate that the purification efficiency depends on the external pressure and the venting *** increase in the external pressure and enough venting time makes it possible to effectively remove the impurities from the tank during the venting process and allows to desorb pure *** impurities are partially removed from the tank for low external pressure and venting time during the venting process and the desorbed hydrogen is *** parameters such as the overall heat transfer coefficient,solid material mass,supply pressure,and the ambient temperature influence the purification system in terms of the hydrogen recovery *** increase in the overall heat transfer coefficient,solid material mass,and supply pressure improves the hydrogen recovery rate while a decrease in the ambient temperature enhances the recovery rate.

关键词： hydrogen production hydrogen purification carbon dioxide metal hydride lumped parameter parametric study