Advancements in converter-based frequency stability : recommendations for industrial applications

Student thesis: Doctoral Thesis

Abstract

The burning of fossil fuels and related carbon emissions are driving the ongoing climate crisis. A critical path to fully decarbonise the power system is to enable low-carbon converter-interfaced devices to take on responsibility for the generation of electrical energy. However, a low-carbon electrical power system also requires the converters to provide the features that fossil-fuel-powered synchronous machines (SMs) conventionally provide to stabilise the electrical grid. The electrical frequency is one key system parameter that needs to be stabilised. Converter-based frequency stabilising solutions have been proposed but the nuance of their operation is not fully understood. Therefore, this thesis aims to address some of the critical hurdles that the solutions must overcome. The thesis initially outlines the technical characteristics that are required to provide inertial and droop responses. Academic and industrial data are assessed to identify the technologies that are techno-economically suited to provide the support. The impact of different controller choice for droop provision is assessed. Previous works have suggested that certain droop controllers are equivalent but often only consider the steady state. Models of Synchronverter and Grid-forming (GFM) Droop controlled ideal energy storage systems are assessed to identify the equivalence of the controllers’ frequency support. A tuning guide developed earlier in the thesis enables the controllers to provide equivalent inertial and droop responses but the dynamics of each controller are shown to be different. The impact of the GFM Droop’s cascaded controllers and its parametric tuning on the frequency support are then assessed and suggestions are made for their tuning. The industrial attempts to quantify useful inertial response are then assessed. Parametric sweeps of example GFM and grid-following (GFL) controllers are carried out to compare their full capability with the industrial specifications. A more detailed power system model is also used to validate the findings of the parametric sweeps and to assess the impact of the controllers’ properties on the system frequency. The study highlights that useful inertial provision is not unique to GFMs, that GFLs should not be subject to blanket disqualifications from inertial support, and that transient phase responses may require more consideration in converter dominated systems. Finally, the ability of system operators (SOs) to measure wind turbine (WT) based inertial support is assessed. Experimental data of a grid-connected wind farm are used to identify the impact that the wind has on the inertial response. A review is carried out to assess the methods that are currently available to measure WT inertial response (including the existing industrial standard). The accuracy of the existing methods are assessed using a model of a WT and its converters, which resolves the dynamics from wind energy source to grid. Two new approaches are proposed that improve the accuracy of WT inertial response measurement.
Date of Award9 Jul 2023
Original languageEnglish
Awarding Institution
  • University Of Strathclyde
SponsorsEPSRC (Engineering and Physical Sciences Research Council) & University of Strathclyde
SupervisorAgusti Egea Alvarez (Supervisor) & Panagiotis Papadopoulos (Supervisor)

Cite this

'