利用叩击和浅层机器学习算法进行水下螺栓连接件松动检测的探索性研究  

An exploratory study of underwater bolted connection looseness detection using percussion and a shallow machine learning algorithm

作　　者：何思宏 陈吉安 陈峥 宋钢兵 Sihong He;Ji'an Chen;Zheng Chen;Gangbing Song(Department of Mechanical Engineering,University of Houston,Houston,TX,77204,USA)

机构地区：[1]Department of Mechanical Engineering,University of Houston,Houston,TX,77204,USA

出　　处：《Acta Mechanica Sinica》2023年第4期14-20,共7页力学学报（英文版）

基　　金：supported by Texas Commission on Environmental Quality through Subsea Systems Institute Award(Grant No.582-15-57593);This project was paid for[in part]with federal funding from the Department of the Treasury through the State of Texas under the Resources and Ecosystems Sustainability,Tourist Opportunities,and Revived Economies of the Gulf Coast States Act of 2012(RESTORE Act).

摘　　要：水下螺栓连接件松动可能导致灾难性后果,因此,对其进行定期检测至关重要.本文探索性地提出了一种通过叩击检测水下螺栓连接件松动情况的方法.当螺栓连接件预紧力降低时,通过敲击其产生的声音会发生相应变化.该方法利用功率谱密度进行特征选择,根据叩击引起的声音频率特征变化,采用浅层机器学习方法K近邻算法来识别相应的松动状态.实验结果证明了该方法的有效性.As the looseness of underwater bolted connections may cause catastrophic consequences,their regular inspection is vital.This paper proposes an exploratory approach to detecting the looseness condition of underwater bolted connections by percussion.The sound produced by tapping a bolted connection will alter when the preload on the connection reduces.Using the power spectrum density for feature selection,the proposed approach employs the frequency feature change of impact-induced sounds and implements the KNN(K-nearest neighbors)algorithm,a shallow machine learning method,to identify the corresponding looseness status.Experiments demonstrate effective performances of the proposed method.

关 键 词：螺栓连接件 K近邻算法 机器学习算法 功率谱密度 频率特征 机器学习方法 定期检测 预紧力 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]