Posters
Siblings:
ProgrammeSpeakersPostersContent PartnersElectrification StageMarkets TheatreStudent DayPresenters dashboardCome meet the poster presenters to ask them questions and discuss their work
Check the programme for our poster viewing moments. For more details on each poster, click on the poster titles to read the abstract.

PO247: Performance benchmarking of an Artificial Neural Network model for the classification of wind turbine operation mode
Minh-Thang DO, Head of Energy Division, Meteodyn
Abstract
In order to analyse the performance of wind turbine, Meteodyn has developed an algorithm to automatically classify datapoints into different operation modes. This algorithm is based on the technical knowledge of the wind turbine and the statistical methods. However, the limit of this approach is that it cannot work well with a small amount of data, and therefore difficultly being applied in real-time application or to new wind turbine recently committed. In the scope of this research, a novel classification algorithm, based on Artificial Neural Network is proposed. This algorithm learns from the behaviour of wind turbine in the past (training dataset) to classify the datapoint in the future (test dataset). The test case shows a high level of precision of this model, not only to classify the future data from the wind turbine used to train the model but also for other wind turbines in the wind farm and even wind turbines from other wind farms. This model could be used to classify SCADA in real-time and could be integrate easily into current Real-Time monitoring platforms on the market.
Follow the event on: