Improved Ensemble Learning for Wind Turbine Main Bearing Fault Diagnosis

Mattia Beretta,Jordi Cusidó,Yolanda Vidal,Jose Sepulveda,Olga Porro

doi:10.3390/app11167523

Abstract

The goal of this paper is to develop, implement, and validate a methodology for wind turbines’ main bearing fault prediction based on an ensemble of an artificial neural network (normality model designed at turbine level) and an isolation forest (anomaly detection model designed at wind park level) algorithms trained only on SCADA data. The normal behavior and the anomalous samples of the wind turbines are identified and several interpretable indicators are proposed based on the predictions of these algorithms, to provide the wind park operators with understandable information with enough time to plan operations ahead and avoid unexpected costs. The stated methodology is validated in a real underproduction wind park composed by 18 wind turbines.

Highlights

Global demand for energy has achieved an unprecedented level with the growing world population and the rising industrialization in developing countries
The goal of this paper is to develop, implement, and validate a methodology for wind turbines’ main bearing fault prediction based on an ensemble of an artificial neural network and an isolation forest algorithms trained only on supervisory control and data acquisition (SCADA) data
In this work an ensemble method for main bearing fault diagnosis has been proposed, implemented, and validated on a real under-production wind park composed by 18 wind turbines (WTs)

Summary

Introduction

Global demand for energy has achieved an unprecedented level with the growing world population and the rising industrialization in developing countries. It is noteworthy that wind energy levelized cost of energy (LCOE) fell 39% between 2010 and 2019 [3]. From this LCOE, the operation and maintenance (O&M) accounts for 28.5% in land-based wind projects, and up to 34% in offshore wind projects [4]. Predictive, or condition-based maintenance (CBM), provides operators with an advanced warning before the actual fault occurs, allowing them to plan ahead and schedule repairs to coincide with weather or production windows to reduce costs and turbine downtime. The latest CBM developments tend to use expensive tailored sensors for condition monitoring (CM), which is not economically viable for turbines already under operation and even less in case they are close to reach the end of their lifespan

Objectives

Methods

Results

Conclusion