GRID STATE EVALUATION OF A LVDC TRACTION NETWORK: METHODS FOR THE ANALYSIS OF FORECAST DATA

The paper focuses on the methodical grid state evaluation of a 660 V low-voltage direct current (LVDC) traction network, which primarily serves as a power source for battery-trolleybuses. The replacement of conventional trolleybuses alongside with the integration of stationary storage systems, photovoltaic systems and electric vehicle charging stations within the LVDC traction network is accompanied by the development of a monitoring and control system. The grid state is subject to high volatility due to the dynamically changing power consumption of the battery-trolleybuses. This results in highly volatile current and voltage profiles, making a forecast-based monitoring and control system most suitable. In an initial step, the grid is divided into several zones. The forecast data processing firstly performs an evaluation of the predicted current and voltage time series depending on the permitted limits and time frames in order to detect current overloads and voltage limit violations. The detected problem situations are then sorted according to their criticality for reasons of prioritisation. Finally, depending on the location, time and other physical aspects, the problem situations are combined and summarised as appropriate since related voltage and current problems can often be linked to a specific cause and identified as a single problem situation.