A Data-driven Method for Estimating Parameter Uncertainty in PMU-based Power Plant Model Validation

Abstract

Event-based, online power plant model validation is an important component in economical maintenance of power grid reliability. It is well-known that any given event will not excite all portions of a power plant model, so those portions cannot be considered validated. This paper introduces a method to provide quantitative information about the uncertainty of model parameters for a given event. This information can be used to identify the portions of a model that should be considered validated and those that will require additional events to validate.