Prediction of power system balancing requirement and tail event

This paper presents a methodology for the prediction of power system balancing requirement and the probability of tail event (large imbalance between generation and load). Maintaining sufficient balancing reserves to match the difference between hourly generation schedule and real-time variable load and intermittent resources becomes more and more challenging with the increasing penetration of intermittent energy sources. The presented methodology uses yearly distributions and hourly distributions of balancing requirement and tail events to provide a high level look at the issue and warn system operators of those hours when problems are most likely to occur. For real-time prediction, a Bayes net model is constructed to model the statistical relationships between system imbalance and forecast errors, generation schedule control errors and other influential factors. The methodology will be able to provide reference information to system operators in determining the sufficiency of system balancing reserve and taking appropriate control actions.