Wind Energy Forecast in Complex Sites with a Hybrid Neural Network and CFD based Method

详细信息    查看全文

• 作者：Francesco Castellani ; Massimiliano Burl ; o ; Samad Taghizadeh ; Davide Astolfi ; Emanuele Piccioni
• 关键词：power forecast ; Neural Networks ; SCADA ; wind energy
• 刊名：Energy Procedia
• 出版年：2014
• 年：2014
• 卷：45
• 期：Complete
• 页码：188-197
• 全文大小：2104 K

文摘

The wind is an intermittent renewable energy source and the energy production forecast is a fundamental activity for many reasons (grid regulation, maintenance, etc.).

In this work a hybrid method (based on weather forecast data, neural networks and computational fluid dynamics) and a pure neural network approach are compared in a complex terrain site.

The post processing of real production data has been discovered to be a key activity. Treatment and filtering of data spreading out from the supervisory control and data acquisition system are fundamental both for training and testing methods reliability.