Fault Prediction of Fan Gearbox Based on K-Means Clustering and LSTM

3Citations
Citations of this article
6Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

The traditional fault prediction model of wind turbine equipment mainly aims to establishing the degradation characteristic curve of the wind turbine, and draws conclusions from the analysis of the curve. However, in the actual industry, the wind turbine operating data has a high degree of nonlinear complexity and the traditional enterprises pay less attention to this aspect, which leads to partial loss in the fault characterization and serious loss in the label data. This makes the previous data processing more difficult, and it also causes the accuracy of the later prediction results to be inferior. Therefore, this paper uses k-means clustering algorithm to process the original running data into K clusters, and then comprehensively analyzes the cluster results and the provided incomplete fault list to extract useful information. The data is built to fit the Long Short-Term Memory (LSTM) model predictions. Comparing the prediction results with the SVRM method reveals that LSTM has certain advantages in fault prediction for the data in this paper.

Cite

CITATION STYLE

APA

Zheng, H., Dai, Y., & Zhou, Y. (2019). Fault Prediction of Fan Gearbox Based on K-Means Clustering and LSTM. In IOP Conference Series: Materials Science and Engineering (Vol. 631). Institute of Physics Publishing. https://doi.org/10.1088/1757-899X/631/3/032043

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free