
Research Article
Prediction Method of Crack Depth of Concrete Building Components Based on Ultrasonic Signal
@INPROCEEDINGS{10.1007/978-3-031-28787-9_47, author={Kangyan Zeng and Yan Zheng and Jiayuan Xie and Caixia Zuo}, title={Prediction Method of Crack Depth of Concrete Building Components Based on Ultrasonic Signal}, proceedings={Advanced Hybrid Information Processing. 6th EAI International Conference, ADHIP 2022, Changsha, China, September 29-30, 2022, Proceedings, Part I}, proceedings_a={ADHIP}, year={2023}, month={3}, keywords={Ultrasonic signal Concrete Building components Crack analysis Crack depth Depth prediction}, doi={10.1007/978-3-031-28787-9_47} }
- Kangyan Zeng
Yan Zheng
Jiayuan Xie
Caixia Zuo
Year: 2023
Prediction Method of Crack Depth of Concrete Building Components Based on Ultrasonic Signal
ADHIP
Springer
DOI: 10.1007/978-3-031-28787-9_47
Abstract
Cracks in concrete components have a serious impact on the safety of building structures. On the one hand, with the increase of service life, cracks will reduce the safety of building structures, and with the deepening of cracks, the service life of building structures will be reduced. Therefore, it is very necessary to predict cracks in concrete components. In the crack prediction of concrete building components, the predicted results deviate from the actual value due to the deviation of the measured strain value of concrete. Based on ultrasonic signal, a method for predicting crack depth of concrete building components is proposed. In the process of concrete ultrasonic transmission, the number and length of micro cracks will increase and expand due to stress concentration. According to this phenomenon, the finite element simulation of concrete building components is carried out to obtain the damage model. The characteristics of concrete cracks are extracted based on ultrasonic signals, and the law of acoustic frequency changing with time can also reflect the state of medium stress. The extracted crack signal features of concrete building components are input into CNN model for prediction and recognition. The test results show that the prediction method of crack depth of concrete building components based on ultrasonic signal can improve the accuracy of prediction results and has high engineering application value.