Modeling Wind Power Forecasting Error Using Piecewise Exponential Distribution

Abstract

The forecast error of wind power is challenging for system operators, and the uncertainty of forecast can be more instructive than forecast values. A piecewise exponential distribution model has been proposed for analysis of short term wind power forecast errors. In case study, comparisons of curve fitting are made between conventional error distributions and the proposed model, which demonstrates the flexibility and accuracy of the latter one.