Extension of a probabilistic load flow calculation based on an enhanced convolution technique

Traditional algorithms used in grid operation and planning only evaluate one deterministic state. Uncertainties introduced by the increasing utilization of renewable energy sources have to be dealt with when determining the operational state of a grid. From this perspective the probability of certain operational states and of possible bottlenecks is important information to support the grid operator or planner in their daily work. From this special need the field of application for Probabilistic Load Flow methods evolved. Uncertain influences like power plant outages, deviations from the forecasted injected wind power and load have to be considered by their corresponding probability. With the help of probability density functions an integrated consideration of the partly stochastic behaviour of power plants und loads is possible. In this context an extension to a convolution based probabilistic load flow is present in this paper. The extension reduces the already limited inaccuracy introduced by network model simplifications. Aspects like accuracy improvement and computation time in comparison to existing method are covered in detail.