Day-ahead scheduling for model predictive power generation based on interval prediction of photovoltaics — generalization to multiple generators

Abstract

The purpose of this study is to obtain a day-ahead schedule of model predictive power generation for multiple generators considering the uncertainty of PV power and demand prediction. Expressing the prediction uncertainty as interval prediction, we calculate the regulating capacity of multiple generators that is sufficient to tolerate a given amount of prediction uncertainty. The regulating capacity is decided by solving an interval valued allocation problem. We give a solution by taking an approach based on the monotonicity analysis.