基于背景噪声降噪的管道阀门内漏量化检测

发布时间：2019-04-04 14:24

【摘要】：由于天然气管道阀门内漏声发射检测环境复杂、噪声干扰严重,极大地降低了阀门内漏流量的检测精度。为此,提出一种基于背景噪声的小波包软阈值降噪处理方法,并通过对降噪处理后的声发射信号采用基于支持向量回归(Support Vector Regression,SVR)方法进行输气管道阀门内漏流量的量化回归预测。结果表明:采用基于背景噪声的小波包软阈值降噪方法能够获取较为纯净的内漏源信号,降噪后获得内漏声发射信号信噪比为6.11。通过对小波包降噪处理后的特征参数进行输气管道阀门内漏流量回归预测,结果优于未进行降噪处理的预测结果,且软阈值降噪预测结果优于硬阈值预测结果,采用软阈值降噪的预测结果平均绝对比例误差为0.164,有效提高了阀门内漏流量量化回归预测的准确度。
[Abstract]:Because of the complicated environment and the serious noise interference, the detection precision of valve internal leakage rate is greatly reduced by acoustic emission detection environment of natural gas pipeline valve. In this paper, a wavelet packet soft threshold denoising method based on background noise is proposed, and support vector regression (Support Vector Regression,) is applied to the acoustic emission signal after denoising. SVR) method was used to predict the internal leakage rate of gas pipeline valve by quantitative regression. The results show that the wavelet packet soft threshold denoising method based on background noise can obtain relatively pure internal leakage source signal, and the signal-to-noise ratio of internal leakage acoustic emission signal is 6.11 after de-noising. Based on the characteristic parameters of wavelet packet de-noising process, the results are better than those without noise reduction processing, and the soft threshold noise reduction prediction results are better than the hard threshold values prediction results. The average absolute proportional error of prediction results with soft threshold noise reduction is 0.164, which effectively improves the accuracy of quantitative regression prediction of valve internal leakage flow.
【作者单位】：中国石油管道科技研究中心/油气管道输送安全国家工程实验室;中国石油管道公司西安输油气分公司;中国石油大学(北京)机械与储运工程学院;
【基金】：中国石油天然气股份公司科技专项“油气管道关键设备国产化”,2012E-2802
【分类号】：TE973.6

【相似文献】