PROJECT TITLE :

A Short-Term Wind Power Forecasting Approach With Adjustment of Numerical Weather Prediction Input by Data Mining

ABSTRACT:

This paper proposes a novel short-term wind power forecasting approach by mining the dangerous knowledge of numerical weather prediction (NWP). Today's short-term wind power forecast (WPF) highly depends on the NWP, that contributes the foremost in the WPF error. This paper 1st introduces a bad knowledge analyzer to fully study the relationship between the WPF error with many new extracted options from the raw NWP. Second, a hierarchical structure is proposed, that is composed of a K-suggests that clustering-primarily based bad information detection module and a neural network (NN)-primarily based forecasting module. Within the NN module, the WPF is absolutely adjusted based mostly on the output of the unhealthy data analyzer. Simulations are performed comparing with two different totally different ways. It proves that the proposed approach can improve the short-term wind power forecasting by effectively identifying and adjusting the errors from NWP.