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

在线阅读下载全文

作  者:孙学超 张友源 朱今祥 王平 尹达 SUN Xuechao;ZHANG Youyuan;ZHU Jinxiang;WANG Ping;YIN Da(The 22nd Research Institute of China Electronics Technology Group Corporation,Qingdao 266000,China)

机构地区:[1]中国电子科技集团公司第二十二研究所,山东青岛266000

出  处:《人民珠江》2025年第2期67-74,共8页Pearl River

基  金:基础性军工科研院所稳定支持项目(A132303219)。

摘  要:PCCP(Prestressed Concrete Cylinder Pipe)断丝电磁检测技术是维护PCCP管线工程安全的一种重要的技术手段,电磁检测技术虽然检测准确率较高、应用较广,但仍存在数据处理烦杂、人工耗时高的问题,难以在实际工程中大规模应用。为了解决传统PCCP管道断丝检测设备存在识别效率低和人工成本高的问题,提出一种基于树莓派和深度学习的断丝检测系统。以树莓派作为主控系统的核心,进行数据采集,再调用提前在PC平台上训练出的LSTM(Long Short-term Memory)网络模型,利用LSTM模型强大的特征提取能力对采集数据进行处理,实时给出断丝检测结果,成功克服了传统方法的局限性,实现了对断丝问题的高效准确识别。试验结果表明,该检测系统对测试集数据检测准确率达到80%,为PCCP管道断丝检测工程化应用提供可用性的解决方案。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 

分 类 号:TV1[水利工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象