Using neural networks to estimate wind turbine power generation

Abstract

This paper uses data collected at Central and South West Services Fort Davis wind farm to develop a neural network based prediction of power produced by each turbine. The power generated by electric wind turbines changes rapidly because of the continuous fluctuation of wind speed and direction. It is important for the power industry to have the capability to perform this prediction for diagnostic purposes-lower-than-expected wind power may be an early indicator of a need for maintenance. In this paper, characteristics of wind power generation are first evaluated in order to establish the relative importance for the neural network. A four input neural network is developed and its performance is shown to be superior to the single parameter traditional model approach.

A Novel Policy for Locational Marginal Price Calculation in Distribution Systems Based on Loss Reduction Allocation Using Game Theory

“…Different approaches have been proposed for prediction of different intermittent phenomena, like wind, in power system such as Kalman Filter [22], Time Series [23], and ANN-based approaches [24]- [26]. However, the common point about all of these tools is uncertainty and error in prediction.…”

Short-term wind speed prediction based on robust Kalman filtering: An experimental comparison

“…The difference between the model in Eq. (5), and the model in Eq. (1), for the sKF, is the inclusion of the scalar weight w k in the conditional probability of the observation z k given the state h k .…”

“…Physical-based methods use parametric models based on a physical description of the atmosphere, and are intended mainly for long-term forecasts [3]. Statistical-based methods include artificial neural networks [4][5][6][7][8], and time series models [9][10][11]. These techniques use data for wind speed forecasting, without resorting to a model of the real system [12].…”

Green IT scheduling for data center powered with renewable energy

“…The prediction is available only for t within a given time window window prediction from the current instant. The prediction function is considered as a black box, and numerous existing works may be used to compute it based on source characteristics and weather forecast, as much for solar panels [30,31] than for wind turbines [32,33]. The grid model allows to represent variations of the electricity price over time, such as on-peak/off-peak pricing.…”