摘要
本文分析了油气输送管道阀门中内漏形成的主要原因及声发射技术检测原理,以球阀阀芯划痕损伤为例,采集不同压力情况下阀门的声发射数据与阀门内漏数据。通过BP神经网络和Adaboost算法对采集的数据分别进行回归分析,通过对比两种算法在测试集数据上的均方根误差,验证了Adaboost算法在阀门内漏速率反演中的可行性。
The main causes of internal leakage in natural gas pipeline valves and the principle of acoustic emission testing are analyzed. Taking the damage of the valve spool as an example,the acoustic emission data and valve internal leakage data of valves under different pressure conditions are collected. The BP neural network and Adaboost algorithm are used to analyze the collected data respectively. By comparing the root mean square errors of the two algorithms on the test set data,the advantages of the Adaboost algorithm in the inversion of the leakage rate in the valve are demonstrated.
出处
《计量与测试技术》
2018年第12期85-87,共3页
Metrology & Measurement Technique
作者简介
刘亚楠,男,工程师。工作单位:自贡检验检测院。
