An acoustic signal cavitation detection framework based on XGBoost with adaptive selection feature engineering. (31st March 2022)

• Record Type: Journal Article
• Authors: Sha, Yu; Faber, Johannes; Gou, Shuiping; Liu, Bo; Li, Wei; Schramm, Stefan; Stoecker, Horst; Steckenreiter, Thomas; Vnucec, Domagoj; Wetzstein, Nadine; Widl, Andreas; Zhou, Kai
• Is Part Of: Measurement. Volume 192(2022)
• Publication Date: 2022-03-31
• Subjects: Cavitation detection -- Acoustics signal -- XGBoost -- Non-overlapping sliding window -- Adaptive selection feature engineering; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2022.110897
• Languages: English
• ISSNs: 0263-2241

Abstract

Abstract: Valves are widely used in industrial and domestic pipeline systems. However, during their operation, they may suffer from the occurrence of the cavitation, which can cause loud noise, vibration and damage to the internal components of the valve. Therefore, monitoring the flow status inside valves is significantly beneficial to prevent the additional cost induced by cavitation. In this paper, a novel acoustic signal cavitation detection framework – based on XGBoost with adaptive selection feature engineering – is proposed. Firstly, a data augmentation method with non-overlapping sliding window (NOSW) is developed to solve small-sample problem involved in this study. Then, the each segmented piece of time-domain acoustic signal is transformed by fast Fourier transform (FFT) and its statistical features are extracted to be the input to the adaptive selection feature engineering (ASFE) procedure, where the adaptive feature aggregation and feature crosses are performed. Finally, with the selected features the XGBoost algorithm is trained for cavitation detection and tested on valve acoustic signal data provided by Samson AG (Frankfurt). Our method has achieved state-of-the-art results. The prediction performance on the binary classification (cavitation and no-cavitation) and the four-class classification (cavitation choked flow, constant cavitation, incipient cavitation and no-cavitation) are satisfactory and outperform the traditional XGBoost by 4.67% and 11.11% increase of the accuracy. Highlights: A novel acoustic signal cavitation detection framework — based on XGBoost with adaptive selection feature engineering is proposed. The XGBoost is applied to the cavitation detection on acoustic signals for the first time, especially to control valves. The non-overlapping sliding window data augmentation method is proposed to solve the few-shot learning problem. Three types of statistical features are extracted to capture sensitive features for cavitation detection. The adaptive selection feature engineering, including feature aggregation and feature crosses.