Deep Learning Model Facilitates Selection of Wind-Farm Sites

published: 2019-10-11 0:30 | editor: | category: Analysis

A research team at Pennsylvania State University (PSU) of the U.S. has developed a brand new deep-learning model, which is capable of locating the optimal site of wind farms for developers with its 24-hour power output forecast.

The deep learning model employs a computing technique known as Analog Ensemble (AnEn). The technique was originally developed by the National Center for Atmospheric Research (NCAR) for the deep-learning technology of its energy-generation model. The computing technology can be used to calculate wind energy probability and forecast wind power output based on the data of the previous years.

The forecast data for the wind’s strength can be very useful to the development of future wind farms, as it can help the farm’s developers to gain a firmer understanding on its power output and investment returns. In the long run, it can also help the grid administrators to flexibly adjust the outputs of their generators to maintain the balance and stability of the grid. 

The implementation of the deep learning model has already been proven to be quite effective for determining the sites of various wind-farm projects.

announcements add announcements     mail print
Share