An uncertainty reduction technique for short-term transmission expansion planning based on line benefits

Abstract

In this article we provide a novel uncertainty reduction technique to deal with uncertainties associated to Renewable Energy Sources (RES) in Transmission Expansion Planning (TEP). For the sake of simplicity, we assume the inherent uncertainty of wind and solar output are already captured in the variability of their hourly load profile. Each hourly operational state (or snapshot) of the year is assumed to represent both a realization of the uncertain RES outputs and the given time period of the year. Therefore, we handle uncertainty reduction by means of snapshot selection. Instead of taking into account all the possible operational states and their associated optimal power flow, we want to select a reduced group of them that are representative of all the ones that should have an influence on investment decisions. This reduced group of operational states should be selected to lead to the same investment decisions as if we were considering all snapshots in the target year. The reduction achieve in the size of the TEP problem should allow the user to compute an accurate enough TEP solution in a much smaller amount of time, or, alternatively, compute the expansion of the network considering more accurately other aspects of the TEP problem. Original operational states are compared in the space of candidate line marginal benefits which are relevant drivers for line investment decisions. Marginal benefits of reinforcements are computed from the nodal prices resulting from the dispatch. Principal Component Analysis (PCA) is applied to cope with the high dimension of the line marginal benefit space. Lastly, a clustering algorithm is used to group operational states with similar marginal benefits together. Our algorithm has been tested on a modified version of the standard IEEE 24 bus system.