人民珠江2025，Vol.46Issue(2)：67-74,8.DOI:10.3969/j.issn.1001-9235.2025.02.008
基于树莓派和深度学习的PCCP管道断丝检测系统
Broken Wire Detection System for PCCPs Based on Raspberry Pi and Deep Learning
摘要
Abstract
Broken wire electromagnetic detection technology for prestressed concrete cylinder pipes(PCCPs)is an important technical means to maintain the safety of PCCP engineering.Although electromagnetic detection technology has a high detection accuracy rate and wide application,it still faces the problems of complicated data processing and high labor time,which limits its large-scale application in actual engineering.In order to solve the problems of low identification efficiency and high labor cost of traditional broken wire detection equipment for PCCPs,a broken wire detection system based on raspberry pi and deep learning was proposed.The raspberry pi was used as the core of the main control system to collect data,and then the long short-term memory(LSTM)network model trained in advance on the PC platform was imported.The powerful feature extraction capability of the LSTM model was used to process the collected data,and the broken wire detection results were given in real time,successfully overcoming the limitations of traditional methods and realizing efficient and accurate identification of broken wires.The test results show that the detection accuracy of the system on the test set data reaches 80%,which provides a feasible solution for the engineering application of broken wire detection for PCCPs.关键词
预应力钢筒混凝土管/断丝检测/深度学习/树莓派Key words
PCCP/broken wire detection/deep learning/raspberry pi分类
建筑与水利引用本文复制引用
孙学超,张友源,朱今祥,王平,尹达..基于树莓派和深度学习的PCCP管道断丝检测系统[J].人民珠江,2025,46(2):67-74,8.基金项目
基础性军工科研院所稳定支持项目(A132303219) （A132303219）
