Rotating machinery anomaly detection using data reconstruction generative adversarial networks with vibration energy analysis

Li, Jun; Liu, Yongbao; Wang, Qiang; Xing, Zhikai; Zeng, Fanlin

doi:10.1063/5.0085354

Cited by 6 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…in the operating regime. Measuring the random vibrations of structures and improving or even optimising them based on the measurements has become one of the most commonly used methods in the study of phenomena affected by vibrations [54][55][56][57][58].…”

Section: Identification Of Defects Deficiencies or Operating Errorsmentioning

confidence: 99%

The Random Vibrations of the Active Body of the Cultivators

et al. 2023

View full text Add to dashboard Cite

The article continues the exposition of the results obtained in researching an agricultural machine for processing soil, designed for research with applications including exploitation. The MCLS (complex machine for soil tillage) was designed to research the working processes of the instruments intended for soil processing. The MCLS cultivator is a modulated machine (it can work for three working widths: 1, 2, and 4 m, with tractors of different powers) that is designed to use a wide range of working bodies. The experimental data obtained with the structure with a working width of 1 m and the results of their processing within the framework of the theory of random vibrations are presented in this article. The experimental results are analysed as random vibrations of the supports of the active working bodies. As a result, the main characteristics of random vibrations are exposed: the distribution function, the average value, the autocorrelation, and the frequency spectrum. These general results regarding random vibrations are used for several critical applications in the design, execution, and exploitation of some subassemblies and assemblies of agricultural machines of this type. The main applications include estimating the probability of the occurrence of dangerous load peaks, counting and selecting the load peaks that produce fatigue accumulation in the material of the supports of the working bodies, identifying some design deficiencies or defects in the work regime, and estimating the effects of vibrations on the quality of soil processing. All of the outcomes are composed of applications in MCLS research and exploitation. The applications pursue well-known objectives of modelling the working processes of agricultural machines: safety at work, increasing the quality of work, optimising energy consumption, and increasing productivity, all in a broad context to obtain a compromise situation. The material and the method are based on experimental data acquisition, processing, and interpretation.

show abstract

Section: Identification Of Defects Deficiencies or Operating Errorsmentioning

confidence: 99%

The Random Vibrations of the Active Body of the Cultivators

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Anomaly detection can be regarded as a one-class classification (OCC) task, where the objective is to learn a single data category and identify whether new data belong to this category [9,10]. Performing anomaly detection without any fault samples available in the training set is indeed a form of one-class classification task [11]. As a machine learning approach, neural networks have been widely applied in the industry.…”

Section: Introductionmentioning

confidence: 99%

Gas Turbine Anomaly Detection under Time-Varying Operation Conditions Based on Spectra Alignment and Self-Adaptive Normalization

Miao,

Feng,

Xiao

et al. 2024

Sensors

View full text Add to dashboard Cite

Gas turbine vibration data may exhibit considerable differences under time-varying conditions, which poses challenges for neural network anomaly detection. We first propose a framework for a gas turbine vibration frequency spectra process under time-varying operation conditions, assisting neural networks’ ability to capture weak information. The framework involves scaling spectra for aligning all frequency components related to rotational speed and normalizing frequency amplitude in a self-adaptive way. Degressive beta variational autoencoder is employed for learning spectra characteristics and anomaly detection, while a multi-category anomaly index is proposed to accommodate various operating conditions. Finally, a dataset of blade Foreign Object Damage (FOD) fault occurring under time-varying operating conditions was used to validate the framework and anomaly detection. The results demonstrate that the proposed method can effectively reduce the spectra differences under time-varying conditions, and also detect FOD fault during operation, which are challenging to identify using conventional methods.

show abstract

“…If the network is trained on data for a normal state, the probability output space of the discriminator calculated for abnormal data is then used to predict an abnormal state. GANs have been used for anomaly detection in time series data [20], image and tabular data [21], and fault detection in unbalanced datasets [22,23].…”

Section: Introductionmentioning

confidence: 99%

“…They used sparse stacked autoencoders for feature extraction and supervised learning, which resulted in 97% accuracy. Researchers have also used GANs as anomaly detectors [23] where the bearing and gear teeth were under investigation. To detect an abnormal state, the anomaly score was used on the discriminator output, which resulted in accuracy from 95% to 98% and an F1-score from 97% to 99%, depending on the test case.…”

Section: Introductionmentioning

confidence: 99%

Gear Fault Detection Method Based on Convex Hull Clustering of Autoencoder’s Latent Space

Batsch,

Kiczek

2024

Applied Sciences

View full text Add to dashboard Cite

This paper presents a method of pitting failure detection in toothed gears based on the reconstruction of the gear case vibrational signal. The effectiveness of the proposed method was tested in an experiment on a power circulation test stand. The autoencoder deep neural network architecture, semi-supervised training, and validation, along with the latent data convex hull-based clustering, are presented. The proposed method offers high efficiency (0.99 F1-measure) in gear state prediction (100% in failure detection, 98.9% in normal state prediction) and provides more capabilities in terms of generalization in comparison with linear machine learning techniques such as principal component analysis and nonlinear like the generative adversarial network. Moreover, it is distinguished by high sensitivity while also being able to detect even slight surface damage (initial pitting). These findings will be of particular relevance to a range of scientists and practitioners working with gear drives who are willing to implement machine learning in signal processing and diagnosis.

show abstract

Rotating machinery anomaly detection using data reconstruction generative adversarial networks with vibration energy analysis

Cited by 6 publications

References 42 publications

The Random Vibrations of the Active Body of the Cultivators

The Random Vibrations of the Active Body of the Cultivators

Gas Turbine Anomaly Detection under Time-Varying Operation Conditions Based on Spectra Alignment and Self-Adaptive Normalization

Gear Fault Detection Method Based on Convex Hull Clustering of Autoencoder’s Latent Space

Contact Info

Product

Resources

About