Zhenzhong Hu

    He received both his BE and PhD degree in the Department of Civil Engineering at Tsinghua University, China. He was a visiting researcher in Carnegie Mellon University.
    He is now the associate professor in Shenzhen International Graduate School, Tsinghua University, and also the secretary general of the BIM Specialty Committee of the China Graphics Society.
    His research interests include information technologies in civil and marine engineering, building information modeling (BIM) and digital disaster prevention and mitigation.
  • 2026-04-16

    2026-04-16

    The paper "ST-SRNet: A deep learning framework for seismic response prediction of subsea tunnels" has been published in Tunnelling and Underground Space Technology.

    As key infrastructure in nearshore and offshore areas, subsea tunnels face extremely high risks from earthquake disasters. Conventional numerical methods are computationally intensive, making them inadequate for real-time prediction demands under complex marine environments. To address this, this study proposes a rapid prediction method integrating high-fidelity finite element (FE) simulation and deep learning technology. A two-dimensional FE model incorporating bidirectional (horizontal and vertical) ground motions and hydrodynamic pressure effects is established to construct a comprehensive seismic response database. On this basis, the Subsea Tunnel Seismic Response Network (ST-SRNet) is developed, combining one-dimensional convolution, attention mechanisms, long short-term memory (LSTM), and a Feature-wise Linear Modulation (FiLM) module to achieve efficient and accurate prediction of seismic responses at multiple monitoring points. Results show that the model achieves high prediction accuracy, with the coefficient of determination values exceeding 0.95 and peak relative errors controlled below 10%, while also exhibiting strong generalization. An engineering case study further validates the robustness and applicability of the proposed method, providing effective technical support for disaster assessment, structural monitoring, and real-time early warning of subsea tunnels.

    Note: Tunnelling and Underground Space Technology is a Q1 top journal in the field of engineering and technology, with an impact factor of 7.4 in 2025. The first author of the paper is Master’s student Yimu Chen, and Prof. Yutao Guo is the corresponding author. This research was supported by the Shenzhen Science and Technology Program.

  • 2026-03-26

    2026-03-26

    The paper "Typhoon-Induced Risk Evolution in Wind Farms: From Disaster-Inducing Factors Identification to Domino Effect Assessment" has been published in the journal Reliability Engineering & System Safety. 

    In response to the technical accidents caused by typhoons in the complex waters of the South China Sea, this study proposes a Disaster-Inducing and Risk Evaluation (DIRE) framework. This framework integrates two modules: the Disaster Inducing Factor Extraction Model (DIFEM), which identifies the key environmental driving factors that cause damage to wind turbines through a hybrid physical-data fusion analysis; and the Hierarchical Analytical Domino Evaluation System (HADES), which realizes systematic hazard classification, hierarchical structuring, probability assessment, and consequence assessment. Through the analysis of five typhoon-induced disaster events in the South China Sea, this study not only identified the common and specific disaster-causing factors that cause damage to wind turbines, but also quantified the cascading disaster risks. The findings of this study provide reliable data and decision-making support for disaster prevention and mitigation on wind farms.

    Note: "Reliability Engineering & System Safety" is a top journal in the field of engineering technology in the Q1 zone, with an impact factor of 11.0 in 2025. The first author of the paper is doctoral student Li Yilin and postdoctoral researcher Li Yihong, and I am the corresponding author. The research results were supported by the National Key R&D Program of China, the Guangdong Basic and Applied Basic Research Foundation, the Shenzhen Science and Technology Program, and the Shuimu Tsinghua Scholar Program.

  • 2026-03-25

    2026-03-25

    The paper "Digital disaster prevention for ocean engineering: Current progress and future directions" has been published in the journal Ocean Engineering. 

    The increasing frequency of typhoons and other extreme climate events has intensified the risks of natural hazard-triggered technological accidents (Natech). Traditional disaster-prevention approaches, largely dependent on static analyses, are no longer sufficient for real-time risk identification or proactive mitigation in ocean engineering. This review synthesizes recent progress in digital disaster prevention, focusing on three core areas: disaster-inducing factor identification, disaster mechanism modeling, and structural safety assessment. The paper integrates physics-based numerical modeling, data-driven simulation, and system dynamics to analyze hazard triggers and cascading failures. It highlights applications of digital twins and deep learning in scenario-based risk analysis and early-warning systems, particularly for offshore wind farms. Furthermore, it proposes a forward-looking digital technology system that integrates environmental sensing, interpretable modelling, and resilience-oriented decision support. Overall, digital disaster prevention provides a significant pathway toward more adaptive, predictive, and resilient safety management in ocean engineering.

    Note: Ocean Engineering is a top journal in the field of engineering technology, classified in Q1, with an expected impact factor of 5.5 in 2025. I am the first author and corresponding author of the paper. The research was funded by the Shenzhen Science and Technology Program and the Guangdong Basic and Applied Basic Research Foundation.

  • 2026-03-05

    2026-03-05

    The paper titled “Dynamics-Based Adaptive Scheduling for Large-Scale Edge Computing Networks in Smart Cities” has been published in  IEEE Internet of Things Journal.

    Addressing the critical challenge of real-time and efficient task scheduling in large-scale distributed edge computing networks caused by the exponential growth of task requests from the rapid proliferation of IoT devices in smart cities, this paper proposes a novel Dynamics-based Adaptive Scheduling (DAS) method. By innovatively transforming the complex task scheduling problem into a two-dimensional geometric model and leveraging the principles of forward dynamics simulation, DAS achieves efficient and non-iterative task assignment, successfully reducing computational complexity to O(n log n). Simulation results demonstrate that the DAS method significantly outperforms existing benchmark strategies: it reduces service response time by 9% to 15%, improves system load balance by 93% to 98%, and decreases overall energy consumption by 2% to 12%. This solution effectively resolves scheduling efficiency and reliability issues in large-scale edge computing environments.

    Note: The IEEE Internet of Things Journal is a top journal in the field of engineering and technology, ranked in the Q1 zone, with an impact factor of 8.9 in 2025. The first author of the paper is Master's student Feng Yixiao, and Professor Ren Zhengru from Shanghai Jiao Tong University serves as the corresponding author. The research was supported by the National Key R&D Program of China.

  • 2026-02-20

    2026-02-20

    The paper "Multi-task deep learning-based failure diagnosis for tubular joints: Automated identification and description generation" was published in the journal Advanced Engineering Informatics.

    This study addresses the critical yet vulnerable tubular joints in offshore platforms and bridges, proposing a multi-task deep learning model for simultaneous failure identification and description generation. Traditional failure inspection methods are time-consuming and labor-intensive, while existing intelligent methods often focus on single-task or single-modal designs, which fail to fully leverage multimodal data and struggle to capture coexisting failure modes, limiting comprehensive diagnostic assessments. To overcome these challenges, a multimodal database is constructed from 141 experimental programs, including 409 failed joints with images, textual descriptions, and component-level labels, further augmented to 1,227 images and 3,681 sentences. The proposed model employs a pretrained encoder for image feature extraction, a multi-label classification decoder for failure identification, and an attention-based image captioning decoder for failure description generation. Experimental results demonstrate that the model achieves a BLEU-4 score of 76.70 and an mAP of 0.9467, outperforming single-task baselines. Validation on real engineering images further shows its preliminary transferability, highlighting the potential of the multi-task learning framework for automated failure diagnosis.

    Note: Advanced Engineering Informatics is a top-tier Q1 journal in the field of engineering technology, with an impact factor of 9.9 in 2025. The first author of the paper is Zhang Wenhao, a PhD student at Southern University of Science and Technology, with Professor Hou Chao as the corresponding author. The research was funded by the National Natural Science Foundation of China (NSFC) , the National Key Research and Development Program of China, and Shenzhen Science and Technology Program.