Bridge health state on-chip monitoring method based on lightweight network

Abstract

The invention relates to a bridge health state on-chip monitoring method based on a lightweight network. The method comprises the following steps: firstly, inputting bridge health state feature information into a deep feature extraction network; extracting depth feature information in the input information through a one-dimensional displacement convolutional network, and inputting the depth feature information into a bridge health state recognition network layer; and finally outputting a state recognition result of the bridge. One-dimensional time sequence information collected by a vibration sensor is used as input, and the health state of a bridge is used as output. Compared with a traditional method, the calculation complexity is reduced by utilizing one-dimensional displacement convolution operation and lightweight point convolution, and the lightweight of the network is realized. Moreover, the displacement convolution operation provides a flexible receptive field for the time sequence, so that the bridge recognition accuracy is improved. Through the edge computing technology, the bridge detection real-time performance is improved.

Description

technical field [0001] The invention belongs to the technical field of bridge health monitoring and evaluation, and in particular relates to an on-chip monitoring method of bridge health status based on a lightweight network. Background technique [0002] With the continuous development of social economy and the continuous increase of my country's road traffic volume, the role of bridges in road traffic is becoming more and more important. It is the basis for accelerating the urbanization of our province and the key to ensuring safe and smooth road traffic. In my country, old and old bridges account for nearly 70% of the total number of bridges. With the continuous increase of traffic load, the safety problems of bridge structures come with it, especially the health monitoring of old bridge structures is an urgent problem in the traffic field. The key issue. Due to the limitation of construction level, understanding of structural complexity, and external unpredictable enviro...

AI Technical Summary

Problems solved by technology

This method increases the on-site workload, resulting in low real-time bridge diagnosis and high test costs

[0007] 2. There are many parameters to be measured, a large number of learning samples, and high evaluation costs

Since old bridges can only be monitored by external installations, the number of sensors increases with the increase of bridge spans. In order to obtain long-term massive data, the workload and difficulty of wired cable layout and evacuation increase, and the on-site test cycle lengthen

Images