250934Enhance publishing丨Unsupervised Anomaly Detection of Hydro-Turbine Generator Acoustics by Integrating Pre-Trained Audio Large Model and Density Estimation

Declaration: If this open-source content is used in papers, books, academic reports, or other works, please cite the following publication (refer to the original link for the latest citation format):WU Ting, WEN Shulin, YAN Zhaoli, FU Gaoyuan, LI Linfeng, LIU Xudu, CHENG Xiaobin, YANG Jun. Unsupervised Anomaly Detection of Hydro-Turbine Generator Acoustics by Integrating Pre-Trained Audio Large Model and Density Estimation[J]. Journal of Electronics & Information Technology. doi: 10.11999/JEIT250934DOI: 10.11999/JEIT250934Original Article: [Unsupervised Acoustic Anomaly Detection for Hydro-turbine Generator Units Fusing Pre-trained Large Audio Models and Density Estimation]Corresponding Author: WU Ting, [e-mail] Source Date: December 03, 2025Funding: China Yangtze Power Co., Ltd. Project (Z152302048)Open Source ContentUnsupervised Acoustic Anomaly Detection for Hydro-turbine Generator Units Fusing Pre-trained Large Audio Models and Density Estimation - Reproduction CodeAs the core power equipment of hydropower stations, the safe and stable operation of hydro-turbine generator units is of great significance to the entire station. In recent years, non-contact acoustic measurement has received widespread attention as an effective detection method. However, it is difficult to collect abnormal acoustic signals during the actual operation of hydro-turbine generator units, which restricts the application of traditional anomaly detection methods and supervised learning-based classification strategies in this field. To address these challenges, this paper proposes an unsupervised acoustic anomaly detection method for hydro-turbine generators that fuses pre-trained large audio models with density estimation k-nearest neighbors (k-NN). First, the validity of general audio features extracted by pre-trained audio models in anomaly detection is verified. Subsequently, a parameter fine-tuning strategy integrating attention statistical pooling and warm-up is designed to achieve model transfer optimization. In the inference stage, a density estimation k-NN is designed to realize a robust distance metric. Experimental results show that the proposed method achieves a multi-metric harmonic mean of 98.7% in the wind tunnel environment and up to 99.9% in the slip ring chamber, providing a feasible and excellent solution for acoustic anomaly detection in hydropower stations.