A Method for Selecting Critical Transmission Grid Usage Cases

Abstract

When performing transmission network analysis, the computational expense of underlying network calculations is a basic problem. For instance, sensitivity analysis in network planning will be inapplicable due to the large number of components, the hourly resolution of time steps and the complexity of N-1 calculations, if considering all data. Anyway, only the most critical network loadings are relevant for the outcome of analysis. For this reason, this paper proposes a method for selecting those time steps that cover the most critical network loadings. At first, the critical range of each line is been defined. Than the scope of critical data is reduced by K-means clustering. Finally a global searching method is adapted to find the most important data. In this way, subsequent network calculations can be applied with acceptable effort and still high accuracy. A case study on the real German transmission network is carried out to validate the method.