Haibin Zhang

Dr. Haibin Zhang

School of Civil Engineering and Artchitecture, Hainan University, Haikou, Hainan, China
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
21 October 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Xinzhe YuanXinzhe Yuan
Xinzhe Yuan
DTDustin Tanksley
Dustin Tanksley
LLLiujun Li
Liujun Li
Haibin ZhangHaibin Zhang
Haibin Zhang
Genda ChenGenda Chen
Genda Chen
DWDonald Wunsch
Donald Wunsch
Contemporary deep learning approaches for post-earthquake damage assessments based on 2D convolutional neural networks (CNNs) require encoding of ground motion records to transform their inherent 1D time series to 2D images, thus requiring high computing time and resources. This study develops a 1D CNN model to avoid the costly 2D image encoding. The 1D CNN model is compared with a 2D CNN model with wavelet transform encoding and a feedforward neural network (FNN) model to evaluate prediction performance and computational efficiency. A case study of a benchmark reinforced concrete (r/c) building indicated that the 1D CNN model achieved a prediction accuracy of 81.0%, which was very close to the 81.6% prediction accuracy of the 2D CNN model and much higher than the 70.8% prediction accuracy of the FNN model. At the same time, the 1D CNN model reduced computing time by more than 90% and reduced resources used by more than 69%, as compared to the 2D CNN model. Therefore, the developed 1D CNN model is recommended for rapid and accurate resultant damage assessment after earthquakes.

3 May 2025 in Journal of Civil Structural Health Monitoring
Verified
Journal Article
Open access
PDF file
Tarutal Ghosh MondalTarutal Ghosh Mondal
Tarutal Ghosh Mondal
ZSZhenhua Shi
Zhenhua Shi
Haibin ZhangHaibin Zhang
Haibin Zhang
Genda ChenGenda Chen
Genda Chen
This study focused on the problem of domain shift in deep learning-based bridge element segmentation. The impracticability of accounting for all possible variabilities vis-à-vis structural shape, size, color, texture, illumination, and other operational conditions in the training process leads to the deterioration in the model performance when applied to test data from novel unseen domains. In such situations, rebuilding the model with labeled training data from the target domain becomes prohibitively expensive and time-consuming in many practical cases. Recent advancements in unsupervised domain adaptation techniques are known to provide viable solutions to this problem. However, it was observed in this study that the performance gain afforded by the domain adaptation techniques is not significant enough to adequately close the domain gaps commonly encountered in vision-based robotic bridge inspections. This study, therefore, proposed a class-wise histogram matching-based data augmentation technique that seeks to complement the domain adaptation strategy, leading to a significantly improved adaptation in situations where no labeled data are available from the target domain. The proposed framework is validated with two case studies concerning deep learning-based bridge element segmentation in inspection images collected by unmanned aerial vehicles (UAVs). It produced a mean intersection-over-union which is 21.2% and 21.3% higher than a benchmark domain adaptation method. In the future, this study can be extended to other relevant application areas, including but not limited to autonomous vision-based bridge defect detection and post-disaster structural reconnaissance.

16 August 2023 in Sustainability
Verified
Journal Article
Open access
PDF file
Haibin ZhangHaibin Zhang
Haibin Zhang
Wenyu LiaoWenyu Liao
Wenyu Liao
Genda ChenGenda Chen
Genda Chen
Hongyan MaHongyan Ma
Hongyan Ma
The exploitation of coal and the disposal of waste plastic present significant environmental and economic challenges that require sustainable and profitable solutions. In response, we propose a renewable construction composite material of coal-based thermoplastic composite (CTC) that can be made from low-grade coal and plastic waste. We developed and tested the hot-press fabrication method for this CTC, using coal with a maximum particle size of 4.75 mm and recycled high-density polyethylene (HDPE). The effects of the coal fraction (50–80 wt%) on compressive properties, thermal properties, microstructure, and ecological and economic efficiencies of the CTC were investigated. Test results revealed that the compressive strength and modulus decrease as the coal fraction increases. However, the thermal properties, including thermal conductivity and specific heat, increase with higher coal contents. Compared to concrete, the CTC has about half the thermal conductivity and twice the specific heat, making it a more energy-efficient construction material. Microstructure testing helped to reveal the mechanisms behind the above behaviors of CTC from the observation of binder volume, bonding quality between coal and HDPE, and porosity variation. The life cycle analysis indicated that the CTC production reduced embodied energy, carbon footprint, and cost by up to 84%, 73%, and 14%, respectively. Therefore, we recommend the CTC with 50–70% coal fraction as an innovative construction material with satisfied mechanical and thermal properties, better cost efficiency, and a reduced ecological impact.

21 November 2018 in Sensors
Verified
Journal Article
Open access
PDF file
Haibin ZhangHaibin Zhang
Haibin Zhang
SHShuang Hou
Shuang Hou
JOJinping Ou
Jinping Ou
Concrete compressive strength is an important parameter of material properties for assessing seismic performance of reinforced concrete (RC) structures, which has a certain level of uncertainty due to its inherent variability. In this paper, the method of concrete strength validation of finite element model using smart aggregate (SA)-based stress monitoring is proposed. The FE model was established using Open System for Earthquake Engineering Simulation (OpenSEES) platform. The concrete strengths obtained from the material test, peak stress of SA, and estimated concrete strength based on SA stress were employed in FE models. The lateral displacement monitored by Liner variable differential transformer and vertical axial load monitored by load cell in the experiment are applied in the model. By comparing the global response (i.e., lateral reaction force and hysteretic loop), local response (i.e., concrete stress, rebar strain, and cross-section moment) and corresponding root-mean-square error obtained from experiment and numerical analysis, the capabilities of validation of FE model using SA-based stress monitoring method were demonstrated.

1 January 2026 in Natural Science of Hainan University
Verified
Journal Article
Open access
PDF file
WLWanqiu Liu
Wanqiu Liu
QWQi Wang
Qi Wang
XHXujie Huo
Xujie Huo
YJYilin Jiang
Yilin Jiang
TSTianshu Sun
Tianshu Sun
JOJinping Ou
Jinping Ou

热带气旋( “飓风”或“台风”)是全球最具破坏性的自然灾害之一,每年导致数百人死亡,并造成数百亿元人民币的经济损失。现有数据表明,全球气候变化使得热带气旋的移速减缓,但由热带气旋引发的降雨量与移速呈现反比关系;此外,虽然全球热带气旋总数呈下降趋势,但4~5级飓风(即超强台风)活动频率整体上升。2017年大西洋飓风季造成美国1 250亿美元的经济损失;2025年美国德克萨斯州飓风引发强降雨,导致逾130人丧生。海南地处台风生成及移动的核心路径,是中国受台风影响最频繁的地区之一。2014年超强台风“威马逊”登陆文昌市,成为有气象记录以来登陆中国的最强台风(登陆强度70 m·s−1),海南岛因灾死亡或失踪24人,直接经济损失达110亿元人民币。2024年,超强台风“摩羯”以仅次于超强台风“威马逊”的强度(登陆强度62 m·s−1),同样登陆文昌市,其破坏力却远超超强台风“威马逊”,造成海南岛死亡4人,直接经济损失超600亿元人民币。 超强台风“摩羯”(国际编号:2411)于2024年9月1日在菲律宾以东洋面生成,6日在文昌市翁田镇沿海登陆(中心最低气压915 hPa),后穿过海口市和澄迈县等地移入琼州海峡,随后移入北部湾海面。海南岛12级及以上风圈覆盖区域受影响超9 h,远超2014年超强台风“威马逊”的3 h,文昌测得的最低气压和最强风速均突破中国所有台风过程陆地实测值。4~8日,海南岛普降暴雨到大暴雨,最大降水量为乐东县尖峰镇691.2 mm,文昌市降水量突破当地历史9月的日最大降雨量极值。海口市转移群众超过10万人,倒塌房屋约400间,房屋损坏超过3万间,农作物受灾面积达570 km2,直接经济损失超过26亿元人民币;文昌市房屋受损达2.5万余间,农作物受灾面积约180 km2,绝收面积高达110 km2,经济损失约达327亿元。近年来海南省通过修订工程建设与设施设备使用规范,并完善灾前应急预案,显著提升了台风防御能力,使得人员伤亡数量明显下降。但随着沿海地区人口与经济持续增长,超强台风所带来的风险不断攀升,造成的经济损失也愈发难以承受。 在世界范围内,从气象角度的热带气旋生成发展基础理论及预报研究已成体系,相关研究成果在降低热带气旋灾害上发挥了重要作用,但超强台风登陆的偶发性和区域差异显著等特征,使得其强大动力对地表影响研究相对匮乏,尚缺乏充分的数据和经验来系统揭示其致灾规律,对于其适应性及防灾减灾策略还未形成成熟的解决方案。作为重大灾害评估及科学研究的基本手段,系统的现场调研至关重要。超强台风“摩羯”过境海南岛后,海南大学土木建筑工程学院组织近百名师生,成立多学科交叉的台风灾害调研团队,赴受灾区域开展实地调研工作,主体调研历时十余日。调研范围覆盖超强台风“摩羯”10~17级风圈区域,以受灾明显的9类建构筑物(建筑围护结构、高层建筑、高耸结构/钢结构、临时建筑、农业温室建筑、道路基础设施、光伏结构、电力通信设施和风力发电系统)以及海岸侵蚀和树木倒伏状况为调研对象,涉及调研单体(场地)超1 000个(处)。调研采用“空−天−地”多源数据采集方法,通过实地走访了解建构筑物背景信息,记录结构损伤形态与程度,结合手持激光测距仪和加速度传感器等设备量化破坏参数;利用卫星遥感与无人机航拍数据分析海岸侵蚀及树木倒伏情况;采用问卷调查与主观感知量化高层建筑风致振动舒适度。数据处理上,依据功能和区位分布等特征对调研对象进行分类,并对灾害损失情况进行等级划分,从而总结灾害损伤规律,同时选取典型单体对象,深入分析灾损成因。调研工作得到了海南省气象局、海南省住房和城乡建设厅、海南省交通运输厅、海南省应急管理厅、海南省自然资源和规划厅等单位的支持和指导。临高县东英镇人民政府、定安县水冲坡村、耕读山房和美鳌村村委会等单位在实地考察和信息收集方面提供了大量帮助。海南信义矿业有限公司、海南浙联钢构集团有限公司、海南海控中能建工程有限公司、文昌鲁能希尔顿酒店、海南智普农业科技发展有限公司和海南雅邦集成房屋有限公司等企业也给予了积极配合与支持。 调研结果显示,海口市和文昌市多处气象站录得的风压超建筑设计荷载标准,部分地区因地形效应出现风速增强现象,如海口湾附近高层建筑群因狭管效应导致局地风速增幅达20%~30%,迎风面建筑围护结构承受的动态风压峰值超过5 kPa(超出常规建筑设计荷载标准近50%);现有建筑围护结构、临时建筑、电力设施和农业温室建筑抗超强台风能力不足,针对台风灾害易损区域,亟需修订相关建设规范;文昌市砂质海岸线整体呈现米级后退,需高度重视并制定长期保护方案;强降雨和树木倒伏等次生灾害严重影响公路和城市道路使用寿命,需结合排水系统和行道树设计等提高韧性;树木倒伏状况与其种类、所处地理位置及周边建筑环境都有直接关系,合理的空间布局能降低灾害损失。超强台风“摩羯”的防灾救灾工作再一次证明,通讯、交通和水电的保障是应急处置的核心,对于降低灾害损失、减少人员伤亡具有关键作用,要加强相关标准建设以及新防灾减灾技术的应用。超强台风的作用区域及破坏程度与其所处的地理位置、地形和土地利用等因素直接关联,具有显著地域性。鉴于有记录以来登陆中国的2次最强台风均在文昌市登陆,并给海南岛造成了巨大的经济损失,建议在海南岛成立台风防灾减灾研究中心,系统研究海南岛台风灾害发生规律、防灾避灾减灾与灾后恢复重建等方面的创新技术、法律法规及政策措施,开展国际交流与合作,服务政府科学决策。


24 September 2026 in Infrastructures
Verified
Journal Article
Open access
PDF file
RLRui Li
Rui Li
QBQichang Ban
Qichang Ban
ZWZilong Wang
Zilong Wang
Zhexiong ShangZhexiong Shang
Zhexiong Shang
ZWZhenzhen Wang
Zhenzhen Wang
Ri NaRi Na
Ri Na
Digital image correlation (DIC) is a crucial technique for capturing and monitoring cracks in concrete during laboratory structural tests. However, existing deep learning-based crack detection methods predominantly rely on supervised learning, which is costly due to the need to collect DIC datasets through destructive structural tests. Furthermore, conventional 2D-based DIC image analysis is inadequate for dynamic visualization and analysis of crack developments in 3D structures. To address these challenges, this study introduces an integrated framework for automated segmentation and management of DIC-based cracks using synthetic data, unsupervised learning, and building information modeling (BIM). First, a deep learning-based synthetic image generation method is presented to simulate realistic DIC cracks without structural tests. Second, an unsupervised learning network, enhanced by a customized multi-stage data augmentation strategy, is proposed for label-free DIC crack segmentation. Third, a DIC-enriched BIM management system is developed to automatically integrate multimodal DIC data with 3D physical models of structures while enabling interactive data management. The performance of the proposed framework was assessed through both synthetic datasets and real-world three-point bending tests on three concrete beams. The results demonstrate that the proposed segmentation method outperforms existing SOTAs, and the developed system enables multimodal DIC data visualization for subsequent SHM applications.