Risk-adjusted Cost Ratios for Quantifying Improvements in Wind Power Forecasting

The stochastic nature of wind energy generation imposes technical and economical challenges on the power system. The fast growth of wind energy installations makes the improvement in wind power forecasting (WPF) critical and important. WPF improvements are achieved through different weather-related models and numerical/statistical models. Different statistical representations of the relative error are used to quantify the improvement in WPF. However, these metrics do not necessarily reflect the impact of the improvement on actual power system operations. This paper develops metrics based on modified risk-adjusted cost ratios to quantify the impact of WPF improvements on the actual operation cost. We use historical data of hourly wind power and load from ERCOT, a statistical model for improving WPF from our previous work, and a synthetic model of the Texas electric grid to illustrate the application of the proposed metrics.