Wanrun Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Wanrun Li has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
Yongfeng Du
Lanzhou Univ. of Technology
15 shared publicationsWenhai Zhao
Lanzhou Univ. of Technology
7 shared publicationsYining Wang
Institute of Earthquake Protection and Disaster Mitigation Lanzhou University of Technology Lanzhou China
5 shared publications- Aiqun Li
School of Civil Engineering, Southeast University, Nanjing 210096, China
3 shared publications - Wenhai Zhao
Institution of Earthquake Protection and Disaster Mitigation, Lanzhou University of Technology, Lanzhou 733050, China
3 shared publications - Boyuan Fan
Institution of Earthquake Protection and Disaster Mitigation, Lanzhou University of Technology, Lanzhou 733050, China
2 shared publications
Latest Publications
- 1 October 2026
- Published by American Society of Civil Engineers (ASCE) in Journal of Performance of Constructed Facilities
- Vol. 40 (5)
Adaptive Universal Pavement Distress Detection Based on Transfer Learning and Deep Learning
Various pavements serving in different environments are vulnerable to numerous types of distress. It is crucial to promptly identify and repair fatal pavement distresses. Addressing the issues of time-consuming and labor-intensive traditional pavement distress detection, as well as its lack of precision, an adaptive and universal pavement distress detection method based on deep learning and transfer learning is proposed. First, by summarizing the primary distresses of a large number of different pavements, a classification method based on an improved vision transformer (ViT) deep learning model combined with transfer learning is proposed to categorize these distresses, achieving a Top-1 classification accuracy of more than 97% on four major distress types. Second, more than 3,000 images from multiple public datasets (CrackForest, RDD, CFD, etc.) are integrated using the improved ViT model to construct a universal pavement distress dataset covering local roads, highways, and rural roads. Finally, an improved You Only Look Once v8 (YOLOv8) deep learning model is proposed for multi-information pavement distresses detection. Experimental results demonstrate that while performance on single-pavement datasets is limited by data imbalance, training on the proposed universal pavement distress dataset significantly improves generalization, achieving 82%–89% mean average precision across multiple distress types and enabling robust detection under diverse real-world conditions. Accurate classification and detection of universal pavements have been achieved through the improved ViT and YOLOv8 deep learning models, providing a reference for pavement distress assessment and maintenance.
- 1 October 2026
- Published by American Society of Civil Engineers (ASCE) in Journal of Performance of Constructed Facilities
- Vol. 40 (5)
Adaptive Universal Pavement Distress Detection Based on Transfer Learning and Deep Learning
Various pavements serving in different environments are vulnerable to numerous types of distress. It is crucial to promptly identify and repair fatal pavement distresses. Addressing the issues of time-consuming and labor-intensive traditional pavement distress detection, as well as its lack of precision, an adaptive and universal pavement distress detection method based on deep learning and transfer learning is proposed. First, by summarizing the primary distresses of a large number of different pavements, a classification method based on an improved vision transformer (ViT) deep learning model combined with transfer learning is proposed to categorize these distresses, achieving a Top-1 classification accuracy of more than 97% on four major distress types. Second, more than 3,000 images from multiple public datasets (CrackForest, RDD, CFD, etc.) are integrated using the improved ViT model to construct a universal pavement distress dataset covering local roads, highways, and rural roads. Finally, an improved You Only Look Once v8 (YOLOv8) deep learning model is proposed for multi-information pavement distresses detection. Experimental results demonstrate that while performance on single-pavement datasets is limited by data imbalance, training on the proposed universal pavement distress dataset significantly improves generalization, achieving 82%–89% mean average precision across multiple distress types and enabling robust detection under diverse real-world conditions. Accurate classification and detection of universal pavements have been achieved through the improved ViT and YOLOv8 deep learning models, providing a reference for pavement distress assessment and maintenance.
- 13 August 2026
- Published by Wiley in The Structural Design of Tall and Special Buildings
- Vol. 35 (8) , e70170
Dynamic Buckling Analysis of Wind Turbine Tower Under Near‐Fault and Far‐Fault Ground Motions
The wind turbine tower, a typical thin‐walled structure, is susceptible to buckling under dynamic loads, particularly during seismic events. The dynamic buckling behavior of wind turbine towers under near‐fault and far‐fault ground motions is investigated herein. A 2‐MW wind turbine located in northwestern China was selected as the prototype, with the blades and nacelle modeled as concentrated masses. For incremental dynamic analysis (IDA), 42 natural ground motion records were employed as input excitations, classified into three categories based on epicentral distance and pulse characteristics: 14 near‐fault pulse‐like ground motions, 14 near‐fault non‐pulse‐like ground motions, and 14 far‐fault ground motions. The dynamic buckling behaviors were analyzed by using phase plane trajectories and the Budiansky–Roth (B–R) criterion. A ductility factor accounting for dynamic buckling effects was developed. Local plastic dynamic buckling at the lower section of the turbine doorway was identified as the primary failure mode. Under near‐fault ground motions, dynamic buckling is primarily governed by geometric nonlinearity, whereas buckling induced by far‐fault ground motions predominantly influenced by both plastic deformation and geometric nonlinearity. Pulse‐like ground motions demonstrate a higher propensity to induce dynamic buckling compared to non‐pulse‐like ground motions. At a 90% reliability level for elastoplastic deformation capacity, the ductility factors are 7.41, 9.48, and 3.80 for near‐fault pulse‐like, near‐fault non‐pulse‐like, and far‐fault ground motions, respectively.
- 13 August 2026
- Published by Wiley in The Structural Design of Tall and Special Buildings
- Vol. 35 (8) , e70170
Dynamic Buckling Analysis of Wind Turbine Tower Under Near‐Fault and Far‐Fault Ground Motions
The wind turbine tower, a typical thin‐walled structure, is susceptible to buckling under dynamic loads, particularly during seismic events. The dynamic buckling behavior of wind turbine towers under near‐fault and far‐fault ground motions is investigated herein. A 2‐MW wind turbine located in northwestern China was selected as the prototype, with the blades and nacelle modeled as concentrated masses. For incremental dynamic analysis (IDA), 42 natural ground motion records were employed as input excitations, classified into three categories based on epicentral distance and pulse characteristics: 14 near‐fault pulse‐like ground motions, 14 near‐fault non‐pulse‐like ground motions, and 14 far‐fault ground motions. The dynamic buckling behaviors were analyzed by using phase plane trajectories and the Budiansky–Roth (B–R) criterion. A ductility factor accounting for dynamic buckling effects was developed. Local plastic dynamic buckling at the lower section of the turbine doorway was identified as the primary failure mode. Under near‐fault ground motions, dynamic buckling is primarily governed by geometric nonlinearity, whereas buckling induced by far‐fault ground motions predominantly influenced by both plastic deformation and geometric nonlinearity. Pulse‐like ground motions demonstrate a higher propensity to induce dynamic buckling compared to non‐pulse‐like ground motions. At a 90% reliability level for elastoplastic deformation capacity, the ductility factors are 7.41, 9.48, and 3.80 for near‐fault pulse‐like, near‐fault non‐pulse‐like, and far‐fault ground motions, respectively.
- 12 June 2026
- Published by Elsevier in Structures
- Vol. 90
Experimental investigation of the seismic performance of innovative self-centering prefabricated beam–column joint
- 12 June 2026
- Published by Elsevier in Structures
- Vol. 90
Experimental investigation of the seismic performance of innovative self-centering prefabricated beam–column joint
- 16 April 2026
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 252
Research on centrifugal stiffening effect of wind turbine rotating blades based on visual monitoring data
- 16 April 2026
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 252
Research on centrifugal stiffening effect of wind turbine rotating blades based on visual monitoring data
- 2 April 2026
- Published by Elsevier in Engineering Failure Analysis
- Vol. 192
Dynamic buckling of wind turbine towers subjected to seismic loads under operating conditions
- 2 April 2026
- Published by Elsevier in Engineering Failure Analysis
- Vol. 192
Dynamic buckling of wind turbine towers subjected to seismic loads under operating conditions
- 10 August 2025
- Published by SAGE Publications in Structural Health Monitoring
Adaptive wind turbine blade surface defect detection based on deep learning and transfer learning
Addressing the issues of low efficiency and insufficient accuracy in traditional methods, a novel adaptive automatic detection framework for wind turbine blade surface defects is proposed, which combines a deep learning model with a transfer learning strategy. First, the super-resolution generative adversarial network deep learning model, which has been pre-trained using transfer learning, is employed to perform super-resolution reconstruction on the acquired images of wind turbine blade surface defects. Second, the pretrained weights from transfer learning and frozen training are utilized to optimize the training speed and detection accuracy of the You Only Look At Coefficients deep learning model. This optimized model is then employed to detect multiple types of defects on the blade, resulting in instance segmentation outcomes for the surface defects on the blade. Finally, defect detection was conducted on four types of blade surface defects at a wind farm in western China and across multiple datasets. The results demonstrated that the integration of deep learning with transfer learning achieved high-precision instance segmentation detection for various multi-scale defect types. Furthermore, this approach enabled the direct output of masks, locations, and size information for blade surface defects, providing a foundation for the assessment of blade surface defects. This study offers a novel solution for the detection of surface defects on wind turbine blades and provides valuable references for the application of deep learning in related fields.
- 10 August 2025
- Published by SAGE Publications in Structural Health Monitoring
Adaptive wind turbine blade surface defect detection based on deep learning and transfer learning
Addressing the issues of low efficiency and insufficient accuracy in traditional methods, a novel adaptive automatic detection framework for wind turbine blade surface defects is proposed, which combines a deep learning model with a transfer learning strategy. First, the super-resolution generative adversarial network deep learning model, which has been pre-trained using transfer learning, is employed to perform super-resolution reconstruction on the acquired images of wind turbine blade surface defects. Second, the pretrained weights from transfer learning and frozen training are utilized to optimize the training speed and detection accuracy of the You Only Look At Coefficients deep learning model. This optimized model is then employed to detect multiple types of defects on the blade, resulting in instance segmentation outcomes for the surface defects on the blade. Finally, defect detection was conducted on four types of blade surface defects at a wind farm in western China and across multiple datasets. The results demonstrated that the integration of deep learning with transfer learning achieved high-precision instance segmentation detection for various multi-scale defect types. Furthermore, this approach enabled the direct output of masks, locations, and size information for blade surface defects, providing a foundation for the assessment of blade surface defects. This study offers a novel solution for the detection of surface defects on wind turbine blades and provides valuable references for the application of deep learning in related fields.
- 31 March 2025
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 230
Wind turbine blade rotational condition monitoring based on RBs-YOLO deep learning model
- 31 March 2025
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 230
Wind turbine blade rotational condition monitoring based on RBs-YOLO deep learning model
- 1 March 2025
- Published by AIP Publishing in Journal of Renewable and Sustainable Energy
- Vol. 17 (2)
Investigation on vibration behavior of onshore wind turbine tower based on long-term monitoring data
To study the structural characteristics of onshore wind turbines under a range of environmental and operational conditions, a 2.0 MW onshore wind turbine tower was monitored for a period of 3 years. Based on the long-term field monitoring data, a statistical analysis of the vibration of the wind turbine and environmental factors was conducted. Then, a method based on covariance analysis was applied to determine the modal parameters of the wind turbine tower. In addition, the operating conditions of the wind turbine were divided. The effects of different operating conditions on the first three orders of frequency and damping ratio of the wind turbine tower were thoroughly examined. Finally, Campbell diagrams for the wind turbine in the X and Y directions were generated. The results indicate that the wind turbine blade speed at the third frequency (0.35 Hz) was nearly identical to the tower's fundamental frequency, which could potentially lead to structural resonance. Additionally, the blade speed of 7 rpm in the wind turbine causes the jump phenomenon, which is consistent with the non-linear jump phenomenon observed in the Sommerfeld effect during the transition from pre-resonance to post-resonance. This phenomenon provides an explanation for the peak of the RMS (root mean square) of the tower at wind speeds of 3 and 8 m/s vibration amplitude, thereby confirming the existence of resonance in the wind turbine. Consequently, the long-term monitoring of the wind turbine provides a basis for the design and assessment of the condition of wind turbine towers.
- 1 March 2025
- Published by AIP Publishing in Journal of Renewable and Sustainable Energy
- Vol. 17 (2)
Investigation on vibration behavior of onshore wind turbine tower based on long-term monitoring data
To study the structural characteristics of onshore wind turbines under a range of environmental and operational conditions, a 2.0 MW onshore wind turbine tower was monitored for a period of 3 years. Based on the long-term field monitoring data, a statistical analysis of the vibration of the wind turbine and environmental factors was conducted. Then, a method based on covariance analysis was applied to determine the modal parameters of the wind turbine tower. In addition, the operating conditions of the wind turbine were divided. The effects of different operating conditions on the first three orders of frequency and damping ratio of the wind turbine tower were thoroughly examined. Finally, Campbell diagrams for the wind turbine in the X and Y directions were generated. The results indicate that the wind turbine blade speed at the third frequency (0.35 Hz) was nearly identical to the tower's fundamental frequency, which could potentially lead to structural resonance. Additionally, the blade speed of 7 rpm in the wind turbine causes the jump phenomenon, which is consistent with the non-linear jump phenomenon observed in the Sommerfeld effect during the transition from pre-resonance to post-resonance. This phenomenon provides an explanation for the peak of the RMS (root mean square) of the tower at wind speeds of 3 and 8 m/s vibration amplitude, thereby confirming the existence of resonance in the wind turbine. Consequently, the long-term monitoring of the wind turbine provides a basis for the design and assessment of the condition of wind turbine towers.
- Open Access
- 1 January 2025
- Published by Tech Science Press in Structural Durability & Health Monitoring
- Vol. 19 (5) , 1241-1263
Seismic Vibration Control of Wind Turbine Towers with Bidirectional Tuned Bellow Liquid Column Damper
To address the vibration issues of wind turbine towers, this paper proposes a bidirectional tuned bellow liquid column damper (BTBLCD). The configuration of the proposed BTBLCD is first described in detail, and its energy dis... | Find, read and cite all the research you need on Tech Science Press
- 1 January 2025
- Published by Tech Science Press in Structural Durability & Health Monitoring
- Vol. 19 (5) , 1241-1263
Open AccessSeismic Vibration Control of Wind Turbine Towers with Bidirectional Tuned Bellow Liquid Column Damper
To address the vibration issues of wind turbine towers, this paper proposes a bidirectional tuned bellow liquid column damper (BTBLCD). The configuration of the proposed BTBLCD is first described in detail, and its energy dis... | Find, read and cite all the research you need on Tech Science Press