STEADY-STATE ANALYSIS OF POWER NETWORKS BASED ON DROOPCONTROLLED INVERTERS

Carballo Palacio, Minerva

Permalink

https://hdl.handle.net/2142/124892 Copy

Description

Title

STEADY-STATE ANALYSIS OF POWER NETWORKS BASED ON DROOPCONTROLLED INVERTERS

Author(s)

Carballo Palacio, Minerva

Issue Date

2022-05-01

Keyword(s)

Power Flow Problem; Newton-Raphson; Quasi-Newton-Raphson; Voltage Droop Control; Distributed Slack Bus; Grid-Forming Inverters.

Date of Ingest

2024-10-15T15:20:43-05:00

Abstract

Steady-state analysis of bulk power networks is typically performed by solving the so-called power flow problem. This problem can be solved through two different approaches: either by considering a single generation bus to take all the slack (single slack bus) or by considering a distributed slack bus (all generation buses contribute to cover the slack). In this work, I will focus on a distributed slack bus formulation of the power flow problem for the steady-state analysis. The generator buses active power output is modelled using an active power setpoint modulated the power imbalance between total generation and net demand, whereas the gen-buses reactive power injected into the grid is usually considered as an unknown variable. The load buses active (reactive) power injection is represented by the corresponding active (reactive) power demand preceded by a negative sign. Then, the power flow problem is solved for all buses’ angles and load-buses’ voltages, since the genbuses’ voltages are set to a constant setpoint value (normally about 1 p.u.). A gen-bus angle must be fixed to a certain value (usually 0 rad without loss of generality). However, the approach in which we are working on over this project differs significantly from the one explained above. The reactive power injected by the generators is no longer a simple variable but depends on two terms: a reactive power setpoint and a voltage mismatch allocated by another participation factor. Moreover, the gen-buses’ voltages are not fixed to a setpoint anymore, so we need to solve the model for them as well. Therefore, the resulting unknowns finally include all buses’ not only angles but also voltages (apart from the angle which we need to fix to 0 rad), and complexity considerably increases. To address the power flow problem described above, we will be working on the implementation of Newton-Raphson in an adequate software platform. Python will be the chosen programming language to code the mentioned algorithm, as well as several more Quasi-Newton methods based on simultaneous and/or non-simultaneous updates of the state variables.

Type of Resource

text

Genre of Resource

dissertation/thesis

Language

eng

Owning Collections