Predictive repair scheduling of wind turbine drive‐train components based on machine learning

Abstract

We devise a methodology to predict failures in wind turbine drive‐train components and quantify its utility. The methodology consists of two main steps. The first step is the set up of a predictive model for shutdown events, which is able to raise an alarm in advance of the fault‐induced shutdown. The model is trained on data for shutdown events retrieved from the alarm log of an offshore wind farm. Here, it is assumed that the timely prediction of low‐severity events, typically caused by abnormal component operation, allows for an intervention that can prevent premature component failures. The prediction models are based on statistical classification using only supervisory control and data acquisition (SCADA) data. In the second step, the shutdown prediction model is combined with a cost model to provide an estimate of the benefits associated with implementing the predictive maintenance system. This is achieved by computing the maximum net utility attainable as a function of the model performance and efficiency of intervention carried out by the user. Results show that the system can be expected to be cost‐effective under specific conditions. A discussion about potential improvements of the approach is provided, along with suggestions for further research in this area.

Deep reinforcement learning driven inspection and maintenance planning under incomplete information and constraints

“…In all baselines, if a component fails, it is immediately replaced. Such decision rules can be optimized by evaluation of possible policies through simulations, based on an underlying Bayesian network, or an actual physics-based model, or a meta-model fitted on data [9,10,82].…”

Automobile Maintenance Modelling Using gcForest

“…Furthermore, Extreme value analysis and maximum difference analysis were adopted to identify operational factors which are relevant to component failure. Colone et al (2019) [18] proposed a machine learning approach to predict failures in wind turbine drive-train components and quantify its utility. Firstly, Naïve Bayes' network and artificial neural network was utilized to predict the failure event.…”

Predictive Maintenance in SCADA-Based Industries: A literature review

“…Gigoni et al (2019) proposed predictive maintenance on wind turbines with SCADA data. Almost similar to Gigoni et al (2019), the SCADA data has been proposed (Colone et al, 2019;Leahy et al, 2018). The difference with Colone et al (2019) and Leahy et al (2018) proposed classification techniques, De La Hermosa González-Carrato et al (2013) proposed a new algorithm for conducting predictive maintenance on wind turbines.…”

“…Almost similar to Gigoni et al (2019), the SCADA data has been proposed (Colone et al, 2019;Leahy et al, 2018). The difference with Colone et al (2019) and Leahy et al (2018) proposed classification techniques, De La Hermosa González-Carrato et al (2013) proposed a new algorithm for conducting predictive maintenance on wind turbines. Still, in research on wind turbines, Bangalore & Tjernberg (2015) studied detection of gearbox bearing fault in wind turbines with a neural network approach, which was a continuation of research about a prediction of wind turbine disturbance with neural network algorithms and SCADA data (Bangalore & Tjernberg, 2013).…”

“…Furthermore, the research article will be mapped based on the learning style and algorithm used. Fifteen papers used a supervised learning style, seven papers Colone et al (2019) used an unsupervised learning style, one paper used a combination of supervised and unsupervised, and four papers did not show a specific learning style. Detail mapping is shown in Table 4.…”