摘要
在管道泄漏检测中,声信号因其独特的传播特性和敏感性成为研究焦点。然而,实际环境中的噪声干扰严重制约了声信号在泄漏检测中的应用效果。因此提出基于对抗式自动编码器的管道泄漏声信号噪声抑制方法。通过专业采集设备获取管道泄漏声信号,为后续处理提供原始数据。利用自动编码器对采集到的声信号进行处理,提取关键特征,为后续噪声抑制提供低维特征表示。通过构建基于生成对抗网络的噪声抑制模型,实现对降维后声信号的噪声抑制。实验结果显示,该方法提高了信号的信噪比,降低了均方误差,证明其在实际应用中的有效性。
In pipeline leakage detection,acoustic signals have become the focus of research due to their unique propagation characteristics and sensitivity.However,the noise interference in the actual environment severely restricts the application effect of sound signals in leak detection.Therefore,a pipeline leakage sound signal noise suppression method based on adversarial autoencoder is proposed.Professional collection equipment is used to obtain pipeline leakage sound signals,providing raw data for subsequent processing.The collected sound signal is processed using an autoencoder to extract key features,providing a low dimensional feature representation for subsequent noise suppression.By constructing a noise suppression model based on generative adversarial networks,noise suppression of the reduced dimensional sound signal is achieved.The experimental results show that this method improves the signal-to-noise ratio of the signal and reduces the mean square error,proving its effectiveness in practical applications.
作者
焦建冲
陆鹏飞
伍志勇
JIAO Jianchong;LU Pengfei;WU Zhiyong(CNOOC Huizhou Petrochemical Co.,Ltd.,Huizhou 516000,China)
出处
《电子设计工程》
2026年第5期32-36,共5页
Electronic Design Engineering
关键词
自动编码器
生成对抗网络
管道泄漏
声信号
噪声抑制
autoencoder
Generating Adversarial Networks
pipeline leakage
sound signal
noise supp-ression
