Fault diagnosis of gas turbine generator bearings using enhanced valuable sample strategy and convolutional neural network

Abstract

Gas turbine bearings operate continuously under complex and harsh conditions such as high temperatures, high pressures and high speeds. Bearing fault monitoring data often exhibits anomalies, noise, missing values, and strong coupling and non-linearity due to real-world random factors. In addition, the traditional convolutional neural network is still limited by the scarcity of labelled samples in real-world conditions and cannot fully extract fault features. To address the complexities of strongly coupled fault data, inconsistent data quality, and the shortage of authentic labeled samples, this paper presents a novel method for bearing fault diagnosis. This method utilizes an enhanced valuable sample strategy in conjunction with convolutional neural networks, integrating data enhancement and active learning for preprocessing to create valuable training sets for the networks. At the outset, random dropout and scaling jitter operations are applied to the original data, effectively removing anomalies and gaps in the fault signals. This process simultaneously emulates variations in the frequency spectrum and amplitude of vibration signals under real operating conditions, thereby enhancing the quality of gas turbine bearing fault data. Subsequently, the method incorporates active learning techniques, iteratively selecting unlabeled data samples with the highest information value, and annotating them using support vector machines. This strategy creates a valuable training dataset that further enhances the fault diagnosis capabilities of the model. In the subsequent phase, a three-dimensional convolutional neural network is employed to extract fault feature information efficiently from the fault data, culminating in precise fault classification. In order to validate the effectiveness and superiority of the algorithm, validation and comparison analyses were carried out using the CWRU and XJTU open datasets. To evaluate the practical effectiveness, a gas turbine main bearing dataset collected in a real environment, named the BaiChuan dataset, is used for the engineering validation of the proposed method.

Bearing fault diagnosis method based on angular domain resampling, relative position matrix and transfer learning

“…equipment. According to the type of signals used, commonly used bearing fault diagnosis methods are classified as vibration analysis [4,5], current analysis [6][7][8], sound analysis [9], and other methods [10,11]. Bearing troubleshooting based on vibration signals is a widely used method because vibration signals can reflect the dynamic behavior and health of the equipment [12].…”

A lightweight multi-feature fusion vision transformer bearing fault diagnosis method with strong local sensing ability in complex environments

“…CNN has been widely used in bearing fault diagnosis, as it can extract rich feature information through the stacking of multiple convolutional layers. For example, Xu et al [9] proposed a bearing fault diagnosis method based on an enhanced valuable sample strategy combined with CNN for gas turbine generators. However, due to the limited perceptual ability of CNN for long sequential data, the data information is usually lost and blurred.…”

DDAGCN: an unsupervised cross-domain identification method for tie rod bolt loosening in a rod-fastening rotor system under different working conditions

“…Guo et al [41] generated fault data through a generative adversarial network to expand the fault dataset and help CNN extract a large amount of feature information, thus achieving fault classification. Xu et al [42] introduced an enhanced valuable sample strategy and combined it with active learning to provide an effective labeled dataset to improve the feature extraction capability of CNN. Although CNN has great potential in feature extraction, it lacks the extraction capability of feature geometry structure information.…”