Bolt detection system and implementation method thereof

Abstract

The present invention discloses a bolt detection system and an implementation method thereof. The system comprises a training learning module and a detection module. The method comprises the steps of:collecting bolt knocking sound by employing a bolt detection system, performing feature extraction of the collected bolt knocking sound, and performing training learning of the extracted features according to the machine learning algorithm; and collecting knocking sound of a bolt to be detected by employing the bolt detection system, and performing defect detection of the collected knocking soundof the bolt to be detected according to a result of training learning. The whole work process does not need human intervention so as to reduce the labor cost and improve the detection reliability; and moreover, the bolt detection system and the implementation method thereof can perform feature extraction and training learning of a lot of knocking sound of bolts to achieve quantification of bolt defect indexes, comprehensively identify the defects of the bolts so as to further improve the detection reliability. The bolt detection system and the implementation method thereof are widely appliedto the field of connection piece detection.

Description

technical field [0001] The invention relates to the field of connector detection, in particular to a bolt detection system and an implementation method thereof. Background technique [0002] As we all know, nodes play a pivotal role in civil structures, and damage to nodes can directly lead to damage to the entire structure, with extremely serious consequences. As the main connecting member of steel structure nodes, bolts should be paid attention to by civil engineers. In recent years, there have been several incidents of high-strength bolts breaking and falling in nuclear power plants, such as gantry frames, pumping station roof trusses, and anti-shock brackets. Potential adverse effects on the safety of the structure and a serious threat to the personal safety of those beneath it. Therefore, preventive inspection of high-strength bolts at the joints of large steel structures is required. [0003] At present, maintenance personnel can only perform preventive inspection on...

AI Technical Summary

Problems solved by technology

[0003] At present, maintenance personnel can only perform preventive inspection on bolts by observation method and small hammer knocking method, but this manual detection method has the following problems: 1. Large-scale steel structures are often high in height and huge in size, and construction work requires Building and dismantling a large number of scaffolds consumes huge financial, manpower and time costs, and there are high safety risks. At the same time, the huge cost consumption makes it difficult to carry out such maintenance, and there are potential safety hazards

2. The observation method can only find whether there is corrosion or obvious deformation on the surface of high-strength bolts, and internal defects are difficult to find and the reliability is low.

3. The small hammer knocking method judges whether the bolt is loose by the sound of knocking the bolt, which relies more on the engineering experience of the construction workers and lacks quantitative indicators

Images