Betriebsplanung für aktive Verteilnetze

Pfeifer, Pascal Philipp; Moser, Albert (Thesis advisor); Vennegeerts, Hendrik (Thesis advisor)

1. Auflage. - Aachen : printproduction M. Wolff GmbH (2023)
Book, Dissertation / PhD Thesis

In: Aachener Beiträge zur Energieversorgung 228
Page(s)/Article-Nr.: X, 183 Seiten : Illustrationen

Dissertation, RWTH Aachen University, 2023

Abstract

The growth of renewable energy sources, along with power consumers such as heat pumps and electric vehicles, has amplified the utilization of the power distribution system. This surge necessitates distribution system operators to initiate countermeasures for maintaining system security. Recent regulatory and technical shifts have stimulated the exploration and implementation of operational remedial actions, like modulating the load or power production of system users. These strategies present an alternative to the conventional approach of system capacity expansion. We refer to power distribution systems that can be monitored and controlled in real-time during system operation as smart grids. The initiation and coordination of remedial actions among various system operators necessitate lead times, requiring decisions to be made in the time domain of operational planning and therefore based on grid utilization forecasts. High forecast errors of grid utilization imply that operational planning must account for forecast uncertainties and their potential impact on congestion of grid elements. Hence, an optimized determination of remedial actions must consider these uncertainties. However, the processes required to execute the aforementioned tasks are not entirely developed at present. Therefore, the aim of this thesis is to propose an effective design for future operational planning for smart grids. This will be achieved by developing process designs and required optimization procedures that account for the uncertainties in grid utilization. The first scientific contribution of this thesis introduces an innovative operational planning process for medium and low voltage power systems, adequately addressing uncertainties in grid utilization forecasts. This process begins by determining probabilistic system usage using kernel density estimation, which is parameterized based on historical forecasts and forecasting errors. The second step in this process involves an innovative analytical method which suitably approximates the relationship between probabilistic system usage and probabilistic system variables of current and voltage, based on complex power flow equations. By integrating the resulting probability density functions of voltages and currents, critical currents and voltages can be derived, considering a user-selected risk preference measure, also known as the confidence level. These key parameters serve as the input for a novel power flow optimization method in the third step of the process, which is based on a successive optimization approach. The effects of the remedial actions on the system parameters of current and voltage are linearly or quadratically approximated in each iteration. The second scientific contribution of this thesis consists of investigations on selected power distribution grids, which determine the added value of such probabilistic approaches in operational planning processes compared to existing deterministic approaches. Detailed investigations reveal that the chosen optimization approach always selects suitable remedial actions from the set of available actions. Planning lead time and selected confidence level significantly influence the amount of remedial actions. The simulation of the entire process underscores a conflicting objective between early implementation of remedial actions, total amount of required remedial actions, and the demand for short-term emergency remedial actions. This confliction objective can be influenced by parameterization and process design.

Institutions

• FGH - Forschungsgemeinschaft für Elektrische Anlagen und Stromwirtschaft e.V. [051400]
• Chair of Transmission Grids and Energy Economics [614110]