Wind Turbine Blade Breakage Monitoring With Deep Autoencoders

Monitoring wind turbine blade breakages based on supervisory control and data acquisition (SCADA) data is investigated in this research. A preliminary data analysis is performed to demonstrate that existing SCADA features are unable to present irregular patterns prior to occurrences of blade breakages. A deep autoencoder (DA) model is introduced to derive an indicator of impending blade breakages, the reconstruction error (RE), from SCADA data. The DA model is a neural network of multiple hidden layers organized symmetrically. In training DA models, the restricted Boltzmann machine is applied to initialize weights and biases. The back-propagation method is subsequently employed to further optimize the network structure. Through examining SCADA data, we observe that the trend of RE will shift by the blade breakage. To effectively detect RE shifts through online monitoring, the exponentially weighted moving average control chart is deployed. The effectiveness of the proposed monitoring approach is validated by blade breakage cases collected from wind farms located in China. The computational results prove the capability of the proposed monitoring approach in identifying impending blade breakages.

[1]  B. Frey,et al.  Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning , 2015, Nature Biotechnology.

[2]  Ahmed Elgamal,et al.  Experimental and Numerical Seismic Response of a 65 kW Wind Turbine , 2009 .

[3]  Meik Schlechtingen,et al.  Comparative analysis of neural network and regression based condition monitoring approaches for wind turbine fault detection , 2011 .

[4]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[5]  Jui-Sheng Chou,et al.  Failure analysis of wind turbine blade under critical wind loads , 2013 .

[6]  Yongqian Liu,et al.  Smart Monitoring of Wind Turbines Using Neural Networks , 2009 .

[7]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[8]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[9]  Jae-Kyung Lee,et al.  Transformation algorithm of wind turbine blade moment signals for blade condition monitoring , 2015 .

[10]  David Lange,et al.  Overview of problems and solutions in fire protection engineering of wind turbines , 2014 .

[11]  Daniel J. Inman,et al.  Impedance-based structural health monitoring of wind turbine blades , 2007, SPIE Smart Structures and Materials + Nondestructive Evaluation and Health Monitoring.

[12]  George C. Runger,et al.  Designing a Multivariate EWMA Control Chart , 1997 .

[13]  Zijun Zhang,et al.  Fault Analysis and Condition Monitoring of the Wind Turbine Gearbox , 2012, IEEE Transactions on Energy Conversion.

[14]  Sedef Kent,et al.  Wind turbine signal modelling approach for pulse Doppler radars and applications , 2015 .

[15]  Gyuhae Park,et al.  Full-scale fatigue tests of CX-100 wind turbine blades. Part I: testing , 2012, Smart Structures and Materials + Nondestructive Evaluation and Health Monitoring.

[16]  Christian Igel,et al.  An Introduction to Restricted Boltzmann Machines , 2012, CIARP.

[17]  F. Oyague,et al.  Wind Energy's New Role in Supplying the World's Energy: What Role will Structural Health Monitoring Play? , 2009 .

[18]  Lina Bertling Tjernberg,et al.  An Artificial Neural Network Approach for Early Fault Detection of Gearbox Bearings , 2015, IEEE Transactions on Smart Grid.

[19]  Raimund Rolfes,et al.  Monitoring a 5 MW offshore wind energy converter—Condition parameters and triangulation based extraction of modal parameters , 2013 .

[20]  Mark D. Powell.  Wind Measurement and Archival under the Automated Surface Observing System (ASOS): User Concerns and Opportunity for Improvement , 1993 .

[21]  Michel Kinnaert,et al.  Model‐based fault detection for generator cooling system in wind turbines using SCADA data , 2016 .

[22]  Daniel Rixen,et al.  Challenges in testing and monitoring the in-operation vibration characteristics of wind turbines , 2013 .

[23]  Manolis Kellis,et al.  Deep learning for regulatory genomics , 2015, Nature Biotechnology.

[24]  Yoshua Bengio,et al.  Practical Recommendations for Gradient-Based Training of Deep Architectures , 2012, Neural Networks: Tricks of the Trade.

[25]  Miguel Á. Carreira-Perpiñán,et al.  On Contrastive Divergence Learning , 2005, AISTATS.

[26]  Léon Bottou,et al.  Large-Scale Machine Learning with Stochastic Gradient Descent , 2010, COMPSTAT.

[27]  Fred Spiring,et al.  Introduction to Statistical Quality Control , 2007, Technometrics.

[28]  H. J. Sutherland,et al.  The application of non-destructive techniques to the testing of a wind turbine blade , 1994 .

[29]  Gunner Chr. Larsen,et al.  Fundamentals for remote structural health monitoring of wind turbine blades - a preproject , 2002 .

[30]  Paulo Cortez,et al.  Modern Optimization with R , 2014, Use R!.

[31]  Huan Long,et al.  Data-Driven Wind Turbine Power Generation Performance Monitoring , 2015, IEEE Transactions on Industrial Electronics.

[32]  Peter Tavner,et al.  Wind turbine downtime and its importance for offshore deployment. , 2011 .

[33]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.