The Efficient Determination of Electric Grid Modes for Large Systems using the Matrix Pencil Method

Abstract

This work presents a novel modal analysis technique in the field of power systems, named the Iterative Matrix Pencil method. Commonly referred to as the IMP method, this technique exists as a new tool for the quick and accurate assessment of signals from an ever-growing electric grid. With the proliferation of renewables in the system, along with an increase in measurement tools, there is more data than ever that can be used to assess the resiliency and stability of the grid.

This work establishes the mathematical framework that constitutes the IMP method, and dis-cusses preexisting techniques and how the IMP method seeks to fill gaps that the existing techniques currently have. The methodology is also tested rigorously against a variety of test cases and different methodologies. A sensitivity analysis is applied to the technique, as many of the parameters of the technique are user-defined, giving way to various avenues of optimization. The methodology is also tested with various input data types, so as to measure the robustness of the technique.

The IMP method is then compared against existing modal analysis techniques that were developed to serve the same purpose of quick and accurate modal analysis for large-scale systems. A variant of the IMP method, called the Random IMP method, is also considered.

The work then explores some initial results in mode sensitivity, leveraging the IMP method. In particular, the behavior of the modes and their manifestation are considered in two situations where oscillations on grids occur. The first is considering the modes when contingencies occur, and elements are removed from the grid. The other situation is considering when oscillations are forced on the grid, s a result of frequency inputs on various elements of the grid.

Future works are also discussed, as there are a myriad of opportunities and applications for this new technique, along with a variety of other improvements that could come about.