Diagnosis of Blade Icing Using Multiple Intelligent Algorithms

Abstract

The icing problem of wind turbine blades in northern China has a serious impact on the normal and safe operation of the unit. In order to effectively predict the icing conditions of wind turbine blades, a deep fully connected neural network optimized by machine learning (ML) algorithms based on big data from the wind farm is proposed to diagnose the icing conditions of wind turbine blades. This study first uses the random forest model to reduce the features of the supervisory control and data acquisition (SCADA) data that affect blade icing, and then uses the K-nearest neighbor (KNN) algorithm to enhance the active power feature. The features after the random forest reduction and the active power mean square error (MSE) feature enhanced by the KNN algorithm are combined and used as the input of the fully connected neural network (FCNN) to perform and an empirical analysis for the diagnosis of blade icing. The simulation results show that the proposed model has better diagnostic accuracy than the ordinary back propagation (BP) neural network and other methods.

A Comprehensive Analysis of Wind Turbine Blade Damage

“…The aim is to achieve the most cost-effective operating mode in terms of electricity consumed for heating versus the maximisation of electricity production from the wind turbine. To this end, several algorithms have been developed, based on a cluster of measurements received through an appropriate SCADA system and the exploitation of neural networks and artificial intelligence [123,124].…”

Digital Twins for Wind Energy Conversion Systems: A Literature Review of Potential Modelling Techniques Focused on Model Fidelity and Computational Load

“…In recent years, several technologies have been developed for anti-icing and de-icing, i.e., electro-thermal heating with carbon heating mats inside the blades, electro-impulse de-icing (EIDI) [47] and passive anti-icing paints and coatings. Similarly to detecting and modelling blade ageing, the combination of SCADA data and data driven machine learning algorithms allows detecting the presence of ice [48]. In [49], a technique was presented for ice detection that relies on processing RGB camera images with convolutional neural networks.…”

Physical De-Icing Techniques for Wind Turbine Blades

“…The impacts of vegetation cover on ice accretion in Northern Sweden have been examined as well [12]. Big data on icing conditions based on the operation of wind farms in northern territories are used to create intelligent algorithms for wind turbine operation with minimal icing [13]. A probabilistic machine learning method was applied to icing-related production loss forecasts for wind energy in cold climates [14].…”

“…optimizing the operating regimes and surface shape to reduce the adhesion of ice by using the intelligent algorithms for wind turbine operations [11,13];…”

“…The authors of [11] note that one option that has not received much attention is airfoil pitch control. Nonetheless, several studies with intelligent algorithms for wind turbine operations [13,14] have shown that the optimal airfoil angle of attack can reduce icing by using the pitch differently during normal and icing operation modes. The work [11] presents an advanced intelligent automatic control system by the pitch variation in the icing condition.…”