Huanwei Xu

Huanwei Xu


Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2018 in Journal of Mechanical Science and Technology
Verified
Journal Article
Closed access
PDF file
HXHuanwei Xu
Huanwei Xu
WLWei Li
Wei Li
MLMufeng Li
Mufeng Li
CHCong Hu
Cong Hu
SZSuichuan Zhang
Suichuan Zhang
XWXin Wang
Xin Wang

1 June 2024 in Engineering Research Express
Verified
Journal Article
Closed access
HXHuanwei Xu
Huanwei Xu
XXXuyuan Xiao
Xuyuan Xiao
ZZZewei Zhao
Zewei Zhao
Zhonglai WangZhonglai Wang
Zhonglai Wang
Production defects caused by irresistible factors such as process design problems or differences in steel properties in strip production affect the economic benefits of the enterprise and threaten production safety. Traditional defect detection methods are difficult to achieve real-time and high-precision detection, so developing surface defect detection methods based on deep learning is of great significance for strip production. In order to effectively improve the accuracy of the deep learning model in detecting surface defects on hot-rolled strip, in this work we propose a real-time detection model for surface defects on strip steel based on the YOLOv8n model. Firstly, the newly convolutional layer Con5v is designed to replace the original convolutional layer in the neck, and an attention mechanism is added in front of each Con5v to improve the algorithm’s ability to extract small target information. Secondly, an additional set of upsampled feature extraction units is added to the neck in order to enhance the spatial information of the feature map. Subsequently, a set of feature fusion units is incorporated and the convolutional layers in it are improved to provide better feature maps. Thirdly, the number of decoupling detection heads is increased to receive more high-quality features. The final experimental results show that YOLOv8n-GAM (YOLOv8 Nano Model with Global Attention Mechanism) achieves 81.4mAP and 82.0FPS on the NEU-DET dataset and 71.2mAP and 55.0FPS on the GC10-DET dataset, which are 5.7% and 6.9% higher than those of YOLOv8n, respectively. The model proposed in this paper achieves a comprehensive performance improvement in strip steel.

10 February 2021 in Mathematical Problems in Engineering
Verified
Journal Article
Open access
PDF file
HXHuanwei Xu
Huanwei Xu
XZXin Zhang
Xin Zhang
HLHao Li
Hao Li
GXGe Xiang
Ge Xiang
An ensemble of surrogate models with high robustness and accuracy can effectively avoid the difficult choice of surrogate model. However, most of the existing ensembles of surrogate models are constructed with static sampling methods. In this paper, we propose an ensemble of adaptive surrogate models by applying adaptive sampling strategy based on expected local errors. In the proposed method, local error expectations of the surrogate models are calculated. Then according to local error expectations, the new sample points are added within the dominating radius of the samples. Constructed by the RBF and Kriging models, the ensemble of adaptive surrogate models is proposed by combining the adaptive sampling strategy. The benchmark test functions and an application problem that deals with driving arm base of palletizing robot show that the proposed method can effectively improve the global and local prediction accuracy of the surrogate model.

14 September 2020 in Journal of Mechanical Science and Technology
Verified
Journal Article
Closed access
PDF file
HXHuanwei Xu
Huanwei Xu
LLLiangwen Liu
Liangwen Liu
MZMiao Zhang
Miao Zhang

5 October 2023 in Journal of Electrochemical Energy Conversion and Storage
Verified
Journal Article
Closed access
PDF file
HXHuanwei Xu
Huanwei Xu
SXShizhe Xiong
Shizhe Xiong
LWLi Wei
Li Wei
LWLingfeng Wu
Lingfeng Wu
Zhonglai WangZhonglai Wang
Zhonglai Wang
Temperature is a critical factor affecting the performance and safety of battery packs of electric vehicles (EVs). The design of liquid cooling plates based on mini-channels has always been the research hotspots of battery thermal management systems (BTMS). This paper investigates the effect of adding vortex generators (VGs) to the liquid cooling channel on the heat dissipation capacity and temperature uniformity of the battery. The shape of the vortex generators (triangle, trapezoid, and semicircle), placement position (middle, inlet, and outlet of the channel), different flowrates, and different numbers of channels on the heat dissipation of the battery are systematically analyzed. The research results indicate that (1) compared to the triangular and trapezoidal vortex generators, the semicircular vortex generators have a lower cost in terms of pressure drop while maintaining the same heat dissipation efficiency. The pressure drop of the semicircular vortex generators is 15.89% less than that of the trapezoidal vortex generators and 20.49% less than that of the triangular vortex generators. (2) The effect of adding vortex generators is more obvious when the flowrate is small in the cooling channels. When the flow velocity is 0.025 m/s, the heat dissipation performance can be increased by 7.4%. (3) When the cross-sectional area of the inlet is fixed, the heat dissipation effect of more channels is better. The average temperature of three and seven cooling channels decreases from 311.23 K to 310.07 K, with a decrease of 8.87%. (4) The temperature difference can be effectively reduced when the vortex generators are concentrated near the outlet of the flow outlet. Its temperature difference is 1.8 K lower than that when the vortex generators are placed near the inlet, with a decrease of 10.5%.

1 January 2022 in Computer Modeling in Engineering & Sciences
Verified
Journal Article
Open access
PDF file
HXHuanwei Xu
Huanwei Xu
WWWenzhang Wei
Wenzhang Wei
HHHanjin He
Hanjin He
XYXuerui Yang
Xuerui Yang
The surrogate model technology has a good performance in solving black-box optimization problems, which is widely used in multi-domain engineering optimization problems. The adaptive surrogate model is the mainstream research direction of surrogate model technology, which can realize model fitting and global optimization of engineering problems by infilling criteria. Based on the idea of the adaptive surrogate model, this paper proposes an efficient global optimization algorithm based on the local remodeling method (EGO-LR), which aims at improving the accuracy and optimization efficiency of the model. The proposed algorithm firstly constructs the expectation improvement (EI) function in the local area and optimizes it to get the update points. Secondly, the obtained update points are added to the global region until the global accuracy of the model meets the requirements. Then the differential evolution algorithm is used for global optimization. Sixteen benchmark functions are used to compare the EGO-LR algorithm with the existing algorithms. The results show that the EGO-LR algorithm can quickly converge to the accuracy requirements of the model and find the optimal value efficiently when facing complex problems with many local extrema and large variable spaces. The proposed algorithm is applied to the optimization design of the structural parameter of the impeller, and the outflow field analysis of the impeller is realized through finite element analysis. The optimization with the maximum fluid pressure (MP value) of the impeller as the objective function is completed, which effectively reduces the pressure value of the impeller under load.