Power system stability analysis and control based on PMU
This paper presents the development of a dynamic equivalent model for a Distribution Network Cell (DNC) based on the UK 11kV distribution system. In a new and innovative approach, the methodology utilises the k-means clustering procedure to approximate the input and output dynamic responses of the network. By grouping together similar dynamic responses, the developed methodology is able to significantly reduce the computation time required to approximate the dynamic response of the network. The accuracy of the procedure is quantified by analysing the deviation of the true response from the approximated cluster centre. The computational efficiency of the approach is compared against parameter estimation without clustering, and the resulting approximation errors and computational improvements highlighted.