Improvements in Wind Speed Forecasting
Posted in Energy Inventions | Wind Farms | Wind Power
Sources of alternative energy are still in infancy stage and taking nascent steps towards the future. So there are ample scopes for scientists to improve upon the various aspects of existing models of alternative sources of energy. There is so much untapped, and so much to explore in this sector. That’s why we are daily flooded with information in the alternative energy field. But sometimes we hear about the improvement in the existing models itself. Researchers from the University of Alcala (UAH) and the Complutense University in Madrid (UCM) have devised a new method for predicting the wind speed of wind farm aerogenerators.
Why wind speed forecasting is necessary? Why should one be accurate while forecasting the wind speed? We know that an electricity grid can develop disturbances in power supply if the balance between demand and supply is disturbed. Power generation from wind is entirely dependent on wind speed and is not easily dispatchable. Various factors affect the wind speed such as season, temperature variation, pressure variation etc. So accurate forecast of wind speed is not easy. But when wind farms are contributing considerably in the energy mix of the grid then it becomes necessary to know that how much power will be produced by the wind farm. They have to behave like conventional power generator units. These forecasts are used to schedule the operations of other plants, and are also used for trading purposes.
How this neural network was developed? Researchers have taken the help from Global Forecasting System from the US National Centers for Environmental Prediction. They provide the data of entire planet earth with a resolution of approximately 100 kilometres. The best thing is one can access all the data for free on the Internet. But researchers went a step further and for more detailed predictions they integrated the ‘fifth generation mesoscale model’ (MM5), from the US National Center of Atmospheric Research. It has a resolution of 15×15 kilometres.
Sancho Salcedo, an engineer at the Escuela Politécnica Superior and co-author of the study, published online in the journal Renewable Energy explained, “This information is still not enough to predict the wind speed of one particular aerogenerador, which is why we applied artificial neural networks.” These neural networks are automatic information learning and processing systems. While doing their work the neural networks imitate the mechanisms of animal nervous systems. Neural networks utilize the temperature, atmospheric pressure and wind speed data already fed to them by forecasting models and data collected from the aerogenerators. All these data are used to acclimatize the systems so that they can predict the wind speed in the time range of one and forty eight hours. Wind farms are bound by law to provide these forecasts to Red Eléctrica Española, the company that delivers electricity and runs the Spanish electricity system.
Salcedo states that the method can be applied immediately: “If the wind speed of one aerogenerator can be predicted, then we can estimate how much energy it will produce. Therefore, by summing the predictions for each ‘aero’, we can forecast the production of an entire wind farm.” They have already applied this method at the wind farm in Fuentasanta, in Albacete. The trial was very successful.
This neural network for wind speed forecasting can save millions of Euros. They have detected an improvement of 2% in predictions as compared to the existing models. But this improvement is really significant if we see it in totality because it will lead to the amount of energy production that can save millions of euros. Scientists are trying to improve the method. They want to incorporate several global forecasting models that will result in several sets of observations. These observations will be applied to banks of neural networks to achieve a more accurate prediction of aerogenerator wind speeds. It will naturally lead to more accurate forecasting of wind speed.