Zongxu Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zongxu Liu has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Ke Man
College of Civil Engineering, North China University of Technology, Beijing 100144, China
2 shared publications - Liwen Wu
School of Civil Engineering, North China University of Technology, Beijing 100144, China
2 shared publications - Ruilin Liu
School of Civil Engineering, North China University of Technology, Beijing 100144, China
2 shared publications - Xiaoli Liu
State Key Laboratory of Hydroscience and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, China
2 shared publications - Zhifei Song
School of Civil Engineering, North China University of Technology, Beijing 100144, China
2 shared publications - Zixiang Cao
School of Civil Engineering, North China University of Technology, Beijing 100144, China
2 shared publications
Latest Publications
- Open Access
Water Leakage and Crack Identification in Tunnels Based on Transfer-Learning and Convolutional Neural Networks
In order to solve the problems of long artificial time consumption, the inability to standardize the degree of damage, and the difficulty of maintaining data in traditional tunnel disease detection methods, this paper proposes the use of Residual Network (ResNet) models for tunnel water leakage and crack detection. ResNet proposes a residual learning framework to ease the training of networks that are deeper than those previously used. Furthermore, ResNet explicitly reformulates the layers as learning the residual functions of the reference layer inputs, rather than learning the unreferenced functions. The ResNet model is built on the Tensorflow Deep Learning (DL) framework and transfer-learning is used to optimize the model. The ResNet-V1 can be obtained by pre-training in ImageNet. The fully connected layers of the ResNet-V1 were modified to four classifications of tunnel disease. Then, the SoftMax function is used to recognize the tunnel diseases. Four network structures have been chosen, i.e., ResNet34 and ResNet50, with and without Transfer-learning, respectively. Those models were selected for testing and training on the sample dataset, and these four network structures were compared and analyzed using five types of evaluation indicators, which are the confusion matrix, accuracy, precision, recall ratio and F1. In identifying tunnel cracks and water leakage, the accuracy of ResNet50 and ResNet34 using the transfer-learning were 96.30% and 91.29%, and the accuracy of ResNet50 was 5.01% higher than that of ResNet34; for the network structure without the transfer-learning, the accuracy of ResNet50 was 90.36% and ResNet34’s accuracy was 87.87%. These data show that the accuracy of ResNet50 is higher than that of ResNet34 with or without the transfer-learning, and the deep structure framework is superior in the identification of tunnel diseases; secondly, comparing the network structures with and without the transfer-learning, it can be found that using the Transfer-Learning can improve the ResNet network’s accuracy for tunnel disease identification. The experiments and reliability analysis demonstrate the intelligent tunnel disease identification method proposed in this paper, and its good robustness and generalization performance. This method can be used for the rapid identification of cracks and water leakage in a tunnel survey, construction and maintenance, which has practical engineering implications for tunnel disease detection.
Open AccessWater Leakage and Crack Identification in Tunnels Based on Transfer-Learning and Convolutional Neural Networks
In order to solve the problems of long artificial time consumption, the inability to standardize the degree of damage, and the difficulty of maintaining data in traditional tunnel disease detection methods, this paper proposes the use of Residual Network (ResNet) models for tunnel water leakage and crack detection. ResNet proposes a residual learning framework to ease the training of networks that are deeper than those previously used. Furthermore, ResNet explicitly reformulates the layers as learning the residual functions of the reference layer inputs, rather than learning the unreferenced functions. The ResNet model is built on the Tensorflow Deep Learning (DL) framework and transfer-learning is used to optimize the model. The ResNet-V1 can be obtained by pre-training in ImageNet. The fully connected layers of the ResNet-V1 were modified to four classifications of tunnel disease. Then, the SoftMax function is used to recognize the tunnel diseases. Four network structures have been chosen, i.e., ResNet34 and ResNet50, with and without Transfer-learning, respectively. Those models were selected for testing and training on the sample dataset, and these four network structures were compared and analyzed using five types of evaluation indicators, which are the confusion matrix, accuracy, precision, recall ratio and F1. In identifying tunnel cracks and water leakage, the accuracy of ResNet50 and ResNet34 using the transfer-learning were 96.30% and 91.29%, and the accuracy of ResNet50 was 5.01% higher than that of ResNet34; for the network structure without the transfer-learning, the accuracy of ResNet50 was 90.36% and ResNet34’s accuracy was 87.87%. These data show that the accuracy of ResNet50 is higher than that of ResNet34 with or without the transfer-learning, and the deep structure framework is superior in the identification of tunnel diseases; secondly, comparing the network structures with and without the transfer-learning, it can be found that using the Transfer-Learning can improve the ResNet network’s accuracy for tunnel disease identification. The experiments and reliability analysis demonstrate the intelligent tunnel disease identification method proposed in this paper, and its good robustness and generalization performance. This method can be used for the rapid identification of cracks and water leakage in a tunnel survey, construction and maintenance, which has practical engineering implications for tunnel disease detection.
- Open Access
Dynamic Compression Characteristics and Failure Mechanism of Water-Saturated Granite
For Fangshan granite in Beijing, the static compression and dynamic compression tests have been carried out separately under natural air drying and water saturation. It was found that the dynamic compressive strength of water-saturated granite is higher than that of air-dried granite, which is contrary to the result that the strength of water-saturated rock is lower than that of air-dried granite under static load. Furthermore, under the medium strain rate condition, when the strain rate is 85 s−1, the dynamic strength of natural air-dried granite could be increased by nearly 0.5 times compared with its static state. The dynamic strength of water-saturated granite could be increased by nearly 1–2 times compared with its static strength, which shows that water-saturated granite has stronger strain rate sensitivity than natural air-dried granite. Meanwhile, under impact loading, from the perspective of water-bearing granite the Bernoulli effect of fluid, the adhesion effect of free water and the Stefan effect of fluid in water-saturated granite were revealed, and found to be the essential reasons affecting the dynamic strength of water-saturated granite. The dynamic strength in different water-bearing states in the range of medium strain rate could then be analyzed in depth, providing a certain reference value for the strength design of water-bearing rock engineering.
Open AccessDynamic Compression Characteristics and Failure Mechanism of Water-Saturated Granite
For Fangshan granite in Beijing, the static compression and dynamic compression tests have been carried out separately under natural air drying and water saturation. It was found that the dynamic compressive strength of water-saturated granite is higher than that of air-dried granite, which is contrary to the result that the strength of water-saturated rock is lower than that of air-dried granite under static load. Furthermore, under the medium strain rate condition, when the strain rate is 85 s−1, the dynamic strength of natural air-dried granite could be increased by nearly 0.5 times compared with its static state. The dynamic strength of water-saturated granite could be increased by nearly 1–2 times compared with its static strength, which shows that water-saturated granite has stronger strain rate sensitivity than natural air-dried granite. Meanwhile, under impact loading, from the perspective of water-bearing granite the Bernoulli effect of fluid, the adhesion effect of free water and the Stefan effect of fluid in water-saturated granite were revealed, and found to be the essential reasons affecting the dynamic strength of water-saturated granite. The dynamic strength in different water-bearing states in the range of medium strain rate could then be analyzed in depth, providing a certain reference value for the strength design of water-bearing rock engineering.