Wind Power Prediction Models- Case Study with Artificial neural network for prediction
Keywords:wind power data analysis, forecasting models, physical forecasting, statistical wind forecasting, artificial neural network, wind speed
Research into the viability of renewable energy sources has expanded as a result of the rising prices and unfavorable environmental effects of conventional, nonrenewable energy sources. Since the 1990s, wind energy has had the world's fastest rate of growth in terms of electricity production. Wind is also a renewable energy source that is both more ecologically benign than conventional energy
sources and happens naturally. One of the key challenges restricting the use of wind energy as a source of energy in the renewable energy market is reliability. The concepts of wind speed and wind power are interrelated and subject to location- and time-specific variations. Aside from that, there is currently no viable method to store the output of a wind turbine, thus it must be incorporated right
away into the electrical grid. Since utility companies must disclose the amount of energy they will produce in the future in order to satisfy expected energy demands, knowing future wind power is essential for wind energy to be economically viable. As more wind power is introduced to the electricity markets, the ability to accurately estimate wind power becomes increasingly important, as a
1% error in estimating wind parameters can result in an estimated loss of $1,200,000 for a 100 MW wind farm over the life cycle of the farm. Hence the importance of this paper by addressing the different sections related to wind energy forecasting in three comprehensive groups. The first section presents the different forecasting methods for wind energy, while the second section presents a case study for building a neural network model to predict wind power using global climate data. The third section refers to using the results of the second section to predict wind energy in Egypt according to Egyptian weather data.
This work is licensed under a Creative Commons Attribution 4.0 International License.