Prony analysis and power system stability-some recent theoretical and applications research

Prony analysis has attracted interest as a tool for determining the modes of power system oscillation from network generator swing curves. This paper first revisits the theoretical aspects of the Prony analysis method. A state-space analysis route is followed and some additional insight into the basis for the method reported. Estimation aspects are then investigated. A proposed application of Kalman filtering as an online algorithm for using several swing curves to generate good estimates of power system modes is made. Using simple low order systems some validation case studies are reported. These studies are designed to give some idea of the type of difficulties that can befall the application of Prony analysis. The issues of sample size selection, data de-trending, model order selection and the use of noisy data are examined. Future research directions are also mentioned