Enhancing voltage regulation in medium voltage distribution grids in Ghana using photovoltaic generation
Datei | Beschreibung | Größe | Format | |
---|---|---|---|---|
Thesis_Desmond_Ampofo.pdf | 5.61 MB | Adobe PDF | Anzeigen |
Autor/Autorin: | Ampofo, Desmond Okwabi | BetreuerIn: | Myrzik, Johanna Rehtanz, Christian |
1. GutachterIn: | Michels, Kai | Weitere Gutachter:innen: | Frimpong, Emmanuel Asuming | Zusammenfassung: | It is estimated that the photovoltaic (PV) generation capacity in distribution grids in Ghana will increase to 200 MW by the year 2030. According to the number of licenses issued by the Energy Commission of Ghana, 80% of this capacity will be located in the northern part of the country. This will have a significant impact on voltage regulation as two main issues arise. First, the distribution grid in the northern and middle parts of the country experiences low voltages especially during the peak demand period as a result of the transmission of power over long distances from the southern part of the country, where about 90% of the total generation capacity is located. The expected integrations of PVs mostly in the northern part of the country will therefore improve the voltage profiles but only during the solar generation period as it reduces the power transfer from the south to the north. The second issue is that the northern parts of the country account for about 10% of the national load demand. Thus, the expected high increase in PV units will potentially lead to a voltage rise on the grid especially as the peak solar hours coincide with the low demand period on a typical day in Ghana. Mitigating these issues requires the active participation of PV units in voltage regulation, which the national grid code allows in the form of local reactive power control methods. Each reactive power control method has a varied impact on the network performance. A comparative assessment of the Q(U) and cosφ(P) local voltage control methods based on genetic algorithm (GA) optimal power flow is therefore carried out using probabilistic models for both load (demand) and generation to account for their stochastic nature. Simulation results reveal the effectiveness of the Q(U) control in reducing power loss on the grid, improving voltage profile, and minimizing the tap movements of the on-load tap changer and voltage regulator as compared to the cosφ(P) voltage control method when implemented on a medium voltage distribution network. Notwithstanding, its performance depends on its control parameters and how it can change its reactive power compensation with the changing load and generation conditions. Therefore, this research proposes an innovative concept to change the reactive power compensation to the changing grid conditions by incorporating an adaptive functionality made up of voltage stability and sensitivity blocks into the Q(U) control method. Different load and generation scenarios are used to demonstrate the effectiveness and robustness of this proposed concept to improve the voltage profiles, especially during both peak demand and solar hour periods by keeping the voltages within the regulatory grid code range using only locally measured variables (voltage and DG active and reactive power) at the PCC. This scheme was furthermore compared with Q(U) control operating with fixed control parameters and fully centralised voltage control. In all the comparisons, the proposed concept performed better than the Q(U) control with fixed control parameters and achieved comparable performance with the fully centralised scheme which uses communication facilities. |
Schlagwort: | Medium voltage distribution network; Ghana; local voltage control; photovoltaic; Q(U) control | Veröffentlichungsdatum: | 25-Aug-2022 | Dokumenttyp: | Dissertation | DOI: | 10.26092/elib/1807 | URN: | urn:nbn:de:gbv:46-elib62455 | Institution: | Universität Bremen | Fachbereich: | Fachbereich 01: Physik/Elektrotechnik (FB 01) |
Enthalten in den Sammlungen: | Dissertationen |
Seitenansichten
250
checked on 26.11.2024
Download(s)
272
checked on 26.11.2024
Google ScholarTM
Prüfe
Diese Ressource wurde unter folgender Copyright-Bestimmung veröffentlicht: Lizenz von Creative Commons