Machine Learning for Wind Turbine Blades Maintenance Management

Wind turbine blades undergo high operational loads, experience variable environmental conditions, and are susceptible to failures due to defects, fatigue, and weather induced damage. These large-scale composite structures are essentially enclosed acoustic cavities and currently have limited, if any, structural health monitoring in practice. A novel acoustics-based structural sensing and health monitoring technique is developed, requiring efficient algorithms for operational damage detection of cavity structures. This paper describes a systematic approach used in the identification of a competent machine learning algorithm as well as a set of statistical features for acoustics-based damage detection of enclosed cavities, such as wind turbine blades. Logistic regression (LR) and support vector machine (SVM) methods are identified and used with optimal feature selection for decision making using binary classification. A laboratory-scale wind turbine with hollow composite blades was built for damage detection studies. This test rig allows for testing of stationary or rotating blades (each fit with an internally located speaker and microphone), of which time and frequency domain information can be collected to establish baseline characteristics. The test rig can then be used to observe any deviations from the baseline characteristics. An external microphone attached to the tower will also be utilized to monitor blade damage while blades are internally ensonified by wireless speakers. An initial test campaign with healthy and damaged blade specimens is carried out to arrive at certain conclusions on the detectability and feature extraction capabilities required for damage detection.

Download Full-text

A Comparative Study for Condition Monitoring on Wind Turbine Blade using Vibration Signals through Statistical Features: a Lazy Learning Approach

International Journal of Engineering & Technology ◽

10.14419/ijet.v7i4.10.20833 ◽

2018 ◽

Vol 7 (4.10) ◽

pp. 190 ◽

Cited By ~ 2

Author(s):

A. Joshuva ◽

V. Sugumaran

Keyword(s):

Machine Learning ◽

Wind Turbine ◽

Phase 1 ◽

Turbine Blades ◽

Computational Time ◽

Learning Approach ◽

Statistical Features ◽

Wind Turbine Blades ◽

Vibration Signals ◽

Machine Learning Approach

This study is to identify whether the wind turbine blades are in good or faulty conditions. If faulty, then the objective to find which fault condition are the blades subjected to. The problem identification is carried out by machine learning approach using vibration signals through statistical features. In this study, a three bladed wind turbine was chosen and faults like blade cracks, hub-blade loose connection, blade bend, pitch angle twist and blade erosion were considered. Here, the study is carried out in three phases namely, feature extraction, feature selection and feature classification. In phase 1, the required statistical features are extracted from the vibration signals which obtained from the wind turbine through accelerometer. In phase 2, the most dominating or the relevant feature is selected from the extracted features using J48 decision tree algorithm. In phase 3, the selected features are classified using machine learning classifiers namely, K-star (KS), locally weighted learning (LWL), nearest neighbour (NN), k-nearest neighbours (kNN), instance based K-nearest using log and Gaussian weight kernels (IBKLG) and lazy Bayesian rules classifier (LBRC). The results were compared with respect to the classification accuracy and the computational time of the classifier.

Download Full-text

Interpretable machine learning in damage detection using Shapley Additive Explanations

10.31224/osf.io/96yf5 ◽

2021 ◽

Author(s):

Artur Movsessian ◽

David Garcia Cava ◽

Dmitri Tcherniak

Keyword(s):

Machine Learning ◽

Damage Detection ◽

Wind Turbine ◽

Prediction Models ◽

Turbine Blades ◽

Wind Turbine Blades ◽

Lumped Mass ◽

Damage Indices ◽

Wide Range ◽

The Difference

In recent years, Machine Learning (ML) techniques have gained popularity in Structural Health Monitoring (SHM). These have been particularly used for damage detection in a wide range of engineering applications such as wind turbine blades. The outcomes of previous research studies in this area have demonstrated the capabilities of ML for robust damage detection. However, the primary challenge facing ML in SHM is the lack of interpretability of the prediction models hindering the broader implementation of these techniques. For this purpose, this study integrates the novel Shapley Additive exPlanations (SHAP) method into a ML-based damage detection process as a tool for introducing interpretability and, thus, build evidence for reliable decision-making in SHM applications. The SHAP method is based on coalitional game theory and adds global and local interpretability to ML-based models by computing the marginal contribution of each feature. The contribution is used to understand the nature of damage indices (DIs). The applicability of the SHAP method is first demonstrated on a simple lumped mass-spring-damper system with simulated temperature variabilities. Later, the SHAP method has been evaluated on data from an in-operation V27 wind turbine with artificially introduced damage in one of its blades. The results show the relationship between the environmental and operational variabilities (EOVs) and their direct influence on the damage indices. This ultimately helps to understand the difference between false positives caused by EOVs and true positives resulting from damage in the structure.

Download Full-text

Health Monitoring of Wind Turbine Blades Through Vibration Signal Using Machine Learning Techniques

Proceedings of the International Conference on Computing and Communication Systems - Lecture Notes in Networks and Systems ◽

10.1007/978-981-33-4084-8_22 ◽

2021 ◽

pp. 239-247

Author(s):

Kuldeep Kushwah ◽

Sudarsan Sahoo ◽

A. Joshuva

Keyword(s):

Machine Learning ◽

Wind Turbine ◽

Health Monitoring ◽

Turbine Blades ◽

Vibration Signal ◽

Machine Learning Techniques ◽

Wind Turbine Blades ◽

Learning Techniques

Download Full-text

Machine Learning Analysis of Non-Destructive Evaluation Data from Radar Inspection of Wind Turbine Blades

10.1109/sdpc52933.2021.9563264 ◽

2021 ◽

Author(s):

Wenshuo Tang ◽

Daniel Mitchell ◽

Jamie Blanche ◽

Ranjeetkumar Gupta ◽

David Flynn

Keyword(s):

Machine Learning ◽

Wind Turbine ◽

Turbine Blades ◽

Wind Turbine Blades ◽

Evaluation Data ◽

Non Destructive Evaluation ◽

Non Destructive ◽

Learning Analysis

Download Full-text

Wind Turbine Blade Damage Detection Using Supervised Machine Learning Algorithms

Journal of Vibration and Acoustics ◽

10.1115/1.4036951 ◽

2017 ◽

Vol 139 (6) ◽

Cited By ~ 14

Author(s):

Taylor Regan ◽

Christopher Beale ◽

Murat Inalpolat

Keyword(s):

Machine Learning ◽

Damage Detection ◽

Wind Turbine ◽

Health Monitoring ◽

Learning Algorithms ◽

Turbine Blades ◽

Machine Learning Algorithms ◽

Test Rig ◽

Wind Turbine Blades ◽

Baseline Characteristics

Wind turbine blades undergo high operational loads, experience variable environmental conditions, and are susceptible to failure due to defects, fatigue, and weather-induced damage. These large-scale composite structures are fundamentally enclosed acoustic cavities and currently have limited, if any, structural health monitoring (SHM) in place. A novel acoustics-based structural sensing and health monitoring technique is developed, requiring efficient algorithms for operational damage detection of cavity structures. This paper describes the selection of a set of statistical features for acoustics-based damage detection of enclosed cavities, such as wind turbine blades, as well as a systematic approach used in the identification of competent machine learning algorithms. Logistic regression (LR) and support vector machine (SVM) methods are identified and used with optimal feature selection for decision-making via binary classification algorithms. A laboratory-scale wind turbine with hollow composite blades was built for damage detection studies. This test rig allows for testing of stationary or rotating blades, of which time and frequency domain information can be collected to establish baseline characteristics. The test rig can then be used to observe any deviations from the baseline characteristics. An external microphone attached to the tower will be utilized to monitor blade health while blades are internally ensonified by wireless speakers. An initial test campaign with healthy and damaged blade specimens is carried out to arrive at several conclusions on the detectability and feature extraction capabilities required for damage detection.

Download Full-text

Damage detection in operational wind turbine blades using a new approach based on machine learning

Renewable Energy ◽

10.1016/j.renene.2020.12.119 ◽

2020 ◽

Author(s):

Kartik Chandrasekhar ◽

Nevena Stevanovic ◽

Elizabeth J. Cross ◽

Nikolaos Dervilis ◽

Keith Worden

Keyword(s):

Machine Learning ◽

Damage Detection ◽

Wind Turbine ◽

Turbine Blades ◽

Wind Turbine Blades ◽

New Approach

Download Full-text

Interpretable Machine Learning in Damage Detection Using Shapley Additive Explanations

ASCE-ASME J Risk and Uncert in Engrg Sys Part B Mech Engrg ◽

10.1115/1.4053304 ◽

2021 ◽

Author(s):

Artur Movsessian ◽

David Garcia Cava ◽

Dmitri Tcherniak

Keyword(s):

Machine Learning ◽

Damage Detection ◽

Wind Turbine ◽

Prediction Models ◽

Turbine Blades ◽

Wind Turbine Blades ◽

Lumped Mass ◽

Damage Indices ◽

Wide Range ◽

The Difference

Abstract In recent years, Machine Learning (ML) techniques have gained popularity in Structural Health Monitoring (SHM). These have been particularly used for damage detection in a wide range of engineering applications such as wind turbine blades. The outcomes of previous research studies in this area have demonstrated the capabilities of ML for robust damage detection. However, the primary challenge facing ML in SHM is the lack of interpretability of the prediction models hindering the broader implementation of these techniques. For this purpose, this study integrates the novel Shapley Additive exPlanations (SHAP) method into a ML-based damage detection process as a tool for introducing interpretability and, thus, build evidence for reliable decision-making in SHM applications. The SHAP method is based on coalitional game theory and adds global and local interpretability to ML-based models by computing the marginal contribution of each feature. The contribution is used to understand the nature of damage indices (DIs). The applicability of the SHAP method is first demonstrated on a simple lumped mass-spring-damper system with simulated temperature variabilities. Later, the SHAP method has been evaluated on data from an in-operation V27 wind turbine with artificially introduced damage in one of its blades. The results show the relationship between the environmental and operational variabilities (EOVs) and their direct influence on the damage indices. This ultimately helps to understand the difference between false positives caused by EOVs and true positives resulting from damage in the structure.

Download Full-text