Wind farm generation forecast and optimal maintenance schedule model

Abstract

Wind farms must periodically take their turbines offline in order to perform scheduled maintenance repairs. Given that these interruptions impact energy generation and that under Power Purchase Agreements productions shortfalls must be replaced by energy purchases in the spot market, the optimal time to begin maintenance work in a wind farm is a function of both the expected wind speeds and electricity spot prices. In this article, we develop a model to determine the optimal maintenance schedule of a wind farm based on forecasted wind speeds and energy prices. We analyze a window of opportunity in the most likely period of the year and perform weekly updates of expected wind speeds and energy price forecasts. Wind speeds are forecasted with an ARMAX model, where monthly dummies are used as exogenous variables to capture the seasonality of wind speeds, while spot prices are simulated under a standard dual stochastic programing model. The decision to defer maintenance to a future date is modeled in a probabilistic model and also under the real options approach. We test these models with actual data from a wind farm in the Brazilian Northeast and provide comparisons with current practice in order to determine the benefits of the model. The results suggest that this model may provide advantages over a stopping decision that randomly chooses a week to begin maintenance within the opportunity window and is close to the optimal stopping date considering perfect information on future wind speeds and electricity prices.

Operating Model and Estimation of the Insurance Premium for an Energy Futures Clearing House in Brazil

“…We also analyzed the sensitivity of returns in relation to the expected level of liquidity, given the characteristics of the Brazilian market, and we evaluated the option of taking out insurance to protect the clearing house. The insurance premium was calculated using a Monte Carlo simulation, a well-established and widely used method in studies of derivatives and futures markets, used in studies such as those of Irwin et al (1996), Cortazar and Schwartz (1998), Abadie and Chamorro (2009) and Pelajo et al (2019). The analysis of the simulation results and the insurance calculation are new and important information for market agents interested in the creation of a Brazilian electricity futures clearing house.…”

A Combined Fuzzy GMDH Neural Network and Grey Wolf Optimization Application for Wind Turbine Power Production Forecasting Considering SCADA Data

“…The results demonstrated that non-parametric based combined models usually have a better performance than other models. Jonas C. Pelajo et al (2019) developed a model to predict wind speed and energy price to determine the optimal maintenance planning of a real wind farm in the Brazilian Northeast [23]. Osório et al (2014) proposed a combined forecasting model based on mutual information, wavelet transform, particle swarm optimization, and adaptive neuro-fuzzy inference system framework to predict the short-term wind power and electricity market prices [24].…”

Decomposition‐Based Multistep Sea Wind Speed Forecasting Using Stacked Gated Recurrent Unit Improved by Residual Connections

“…e time series methods model the relationship between current wind speeds and historical wind speeds, which are suitable for short-term and medium-term forecast. Most time series methods, such as the autoregressive integrated moving average (ARIMA) [2,13] and autoregressive moving average with exogenous variables (ARMAX) [14], assume that there is a linear relationship between current data and past data or errors. e construction, order identification of these models is easy to understand, but their linear assumptions lead to poor forecast performance on nonlinear data.…”