Data-driven Assessment and Control of Smart Power Distribution Systems

• Author / Creator

  Bagheri, Pooya

• Modern energy policies and emerging information technologies continuously encourage for innovative strategies in efficient and reliable operation of the energy grids. In electrical power systems, the existing forms of advanced control strategies depend heavily on computer simulations running based on circuit models. This becomes a challenge for operation of distribution feeders where accurate and complete circuit models are not easily attainable. This thesis introduces a data-driven approach in assessment and control of distribution grids. The key idea is to replace the conventional role played by circuit models with the statistical data-analytic techniques. Such innovation eliminates the difficulties and uncertainties generally associated with power system models at the distribution level.A statistical framework is developed to realize model-free methods in operation and assessment of a distribution system. A comprehensive theoretical analysis is presented to support efficacy of this framework. The framework serves as a foundation to the proposed model-free Volt-Var Control (VVC) and Conservation Voltage Reduction (CVR) methods. The assessment and control schemes presented for VVC operates using the data from an Advanced Metering Infrastructure (AMI), while, the CVR one can perform with limited data from meters at the feeding substation, voltage regulators and critical buses of the system. The data-driven approach is also applied to power quality assessment of a distribution system in terms of voltage sag occurrence. Using statistical estimations, the present formulation of Voltage Sag State Estimation (VSSE) methods is generalized to cover both meshed and radial feeder configurations. This new perspective also allows for consideration of Distributed Energy Resource (DER) impact on the voltage sag profile. Several simulation studies are presented to confirm feasibility and effectiveness of the proposed data-driven techniques. The results show that model-free VVC operates well even without synchro-phasor measurement or when the time resolution of data is as high as 15 minutes. Hence, it is consistent with existing technology of AMI systems. The simulations also confirm sufficiency of the measurement requirement for a model-free CVR as specified by this thesis. The proposed VSSE methods are also demonstrated as effective even when only 25% of a system nodes are equipped with meters. Indeed, the proposed concept can be considered as a tangible application for the available data from modern grid measurement technologies such as AMI systems. By taking advantage of technology advancements in communication and data-analytics, this research can act as a genuine support to ongoing transformation of energy grids toward the so-called Smart Grid vision.

• Subjects / Keywords

  • Volt-Var-Control
  • Data Analytics
  • Smart Power Systems
  • Power System Assessment

• Graduation date

  Spring 2019

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-wxb8-0691

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.