Dynamic State Estimation of Nonlinear Differential Algebraic Equation Models of Power Networks

TL;DR

This paper presents a novel coupled dynamic state estimation method for nonlinear power network models that effectively handles uncertainties and is computationally efficient, improving upon traditional decoupled approaches.

Contribution

It introduces the first coupled estimation algorithm for NDAE power models that is simple, uncertainty-aware, and less computationally demanding.

Findings

Successfully estimates algebraic and generator states simultaneously

Handles various uncertainties including noise and renewable sources

Reduces computational complexity compared to existing methods

Abstract

This paper investigates the joint problems of dynamic state estimation of algebraic variables (voltage and phase angle) and generator states (rotor angle and frequency) of nonlinear differential algebraic equation (NDAE) power network models, under uncertainty. Traditionally, these two problems have been decoupled due to complexity of handling NDAE models. In particular, this paper offers the first attempt to solve the aforementioned problem in a coupled approach where the algebraic and generator states estimates are simultaneously computed. The proposed estimation algorithm herein is endowed with the following properties: (i) it is fairly simple to implement and based on well-understood Lyapunov theory; (ii) considers various sources of uncertainty from generator control inputs, loads, renewables, process and measurement noise; (iii) models phasor measurement unit installations at arbitrary buses; and (iv) is computationally less intensive than the decoupled approach in the literature.