Uncertainty modeling for analysis and design of monitoring systems for dynamic electrical distribution grids

Angioni, Andrea; Ponci, Ferdinanda (Thesis advisor); Nguyen, Phuong (Thesis advisor)

1. Auflage. - Aachen : E.ON Energy Research Center, RWTH Aachen University (2019)
Book, Dissertation / PhD Thesis

In: E.ON Energy Research Center : ACS, Automation of complex power systems 73
Page(s)/Article-Nr.: 1 Online-Ressource (419 Seiten) : Illustrationen, Diagramme

Dissertation, RWTH Aachen University, 2019

Abstract

Electrical distribution grids are undergoing radical changes. The increased penetration of distributed energy resources in the form of controllable loads, storage and generation brings significant alterations to dynamic behavior of the distribution systems. For such transformations to be reliable and effective, they must be supported by upgrading the automation of the grid and, particularly, its monitoring system. The current trend in distribution automation is to use distribution system state estimation to merge heterogeneous measurement quantities and obtain accurate state of the network. In the short to medium term scenario, state estimation will be based on static models of the systems, given the lack of experience of the Distribution System Operator (DSO) on the fast evolving dynamics features of the grid and the high effort required to build and maintain an accurate dynamic model. This thesis addresses this scenario, by modelling the propagation of uncertainty sources in presence of static state estimation applied to dynamic systems. The dynamic model of the distribution grid, defined in state-space representation, includes inverters, conventional loads and lines, aggregated in a unique system. Consequently, in addition to the uncertainty sources associated to steady-state conditions, such as measurement devices and pseudo-measurements uncertainties, it is possible to include uncertainty sources related to dynamic distribution grids, namely variations in times of loads, voltage and solar irradiance. The formulation for propagation of uncertainty sources is applied on an extensive impact analysis, aimed at showing the quantitative effect of grid parameters such as topology and line parameters, number and parameters of power inverters on total state estimation uncertainty. The same approach is applied on the monitoring parameters, such as measurement uncertainty, reporting rate of state estimation, communication delay and time synchronization error. The impact study serves as input for the analysis and design process that constitutes the last contribution of this thesis. Indeed, it is possible to calculate the expected total uncertainty of a monitoring system applied to a specified distribution network, or vice-versa, it is possible to discern indications on monitoring parameters for a determined distribution grid once the uncertainty objective are specified. In order to guarantee usability and repeatability of the thesis contribution, the state estimation based monitoring system is mapped on standard use case and Smart Grid Automation Model (SGAM) architecture. To demonstrate applicability in real scenarios, the use case and the architecture are particularized on two real distribution grids, respectively in low and medium voltage level, to test analysis and design of monitoring systems.

Institutions

• E.ON Energy Research Center [080052]
• Monitoring and Distributed Control for Power Systems Teaching and Research Area [616520]