Society is going through a change to be more sustainable, and one big step in that process is to convert the power grid from convectional power generation to increased part of renewable energy sources. Renewable energy sources often require power electronics such as converters to connect to the power grid, and power electronics can cause interference in the grid. This development in Sweden and rest of the Nordic countries is driven by the large increase of load based on power electronics. To maintain a stable power grid, it is essential to detect disturbances through measurement techniques adapted to the challenges that may arise. The measurement techniques in the Swedish power grid have historically been based on power quality meters and RMS-based measurements, and this thesis focuses on the disturbances that RMS-based measurements may fail to detect. The objective is to compare the RMS-based measurements techniques available in the power grid with waveform measurement techniques developed to see how they could detect disturbances in the power grid. The first research question pertains to the characteristic frequency bandwidths within which a power system typically operates, and what frequency bandwidths a system could have with a significant amount of power electronics installed. The second research question concerns the measurement techniques. This work consists of four main parts: bandwidth analysis of control systems for voltage source converter, RMS vs waveform detection techniques, interaction between converter and voltage perturbation, and realistic phenomena studies. The bandwidth analysis aims to determine the frequency range expected in future power systems, which in turn informs the minimum detection capabilities required for measurement techniques. This was achieved through calculations involving the control loops of converters, including the inner current control loop, outer control loop, and phase locked loop. Bode diagrams were generated for each control loop, with variables manipulated, and the short circuit ratio adjusted to determine the bandwidth of the control loop. The RMS vs waveform detection techniques phase was central to the study, involving simulations imitating various phenomena that could occur in the power grid. This was accomplished using two ideal sources and examining their interactions. The simulations included frequency perturbations, amplitude perturbations, and added oscillations apparent at sub-synchronous, super-synchronous, and high super-synchronous frequencies. Simulation results were presented for both RMS and waveform based defection techniques, simulating different reporting rates to mimic real measurement tools. The third and fourth parts involved validating the RMS vs waveform detection techniques results using real-case simulations. An HVDC model was employed to simulate the effects of voltage perturbation and observe the coupling over frequency mechanism. The same HVDC model was also used to simulate phenomena occurring in weak grids. Additionally, a wind park model was utilized to simulate the induction generator effect phenomena. The findings reveal that the bandwidth of a power system undergoes significant expansion as the proportion of power electronics increases in the grid. Specifically, analysis of converter indicates a shift from under 50 Hz to several kHz in systems dominated by synchronous generators versus those dominated by power electronics. To effectively detect higher frequency oscillations within the grid, waveform measurement tools become essential, as classical RMS measurement tools are inadequate for capturing oscillations originating from power electronics. As power systems evolve towards dominance by power electronics, it becomes imperative to develop measurement systems capable of accurately detecting high frequency oscillations. The recommendation of this thesis is to invest in waveform measurement tools alongside the already existing RMS-based ones to enhance the detection of disturbances, particularly those with super synchronous frequencies, addressing future power grid challenges.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-205058 |
| Date | January 2024 |
| Creators | Larsson, Ellen, Carlsson, Carl |
| Publisher | Linköpings universitet, Fordonssystem |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0029 seconds