Abstract

In Germany it is becoming more challenging to maintain the dynamical stability of the power grid due to the increasing share of intermittent renewables like photovoltaics. A possible solution suggested in literature is to improve this dynamical grid stability using Decentral Smart Grid Control (DSGC), where an electric vehicle (EV)
... read more
is used as flexible load device. By using a control strategy these EVs can control their power consumption from- and power injection into the grid to dampen undesirable frequency fluctuations, whenever they are plugged in. In this study it has been investigated how three different control strategies affect the dynamical grid stability when a MV grid is perturbed by power output fluctuations from PV. Contrary to existing literature on DSGC, this study used a system model developed in the CoNDyNet project, which explicitly models the network structure of the grid and includes power output fluctuations from PV with a temporal resolution of seconds. DSGC was integrated into this model by using agent-based modelling, so EVs were included as autonomous agents with DSGC availability parameters based on German driving patterns. The performance of the control strategies was assessed by calculating the average probability that a node in the grid experienced frequency deviations of 200 mHz or more and by comparing this probability to the exceedance probability in a baseline scenario without DSGC. It was found that all three control strategies improved the dynamical stability in the grid, since they strongly reduced the amount of undesirable frequency deviations. It was also shown that different control strategies affect the dynamical grid stability in a different manner and result in different degrees of stability improvement. The simplest control strategy, called the threshold strategy, performed best by prescribing the EV to charge or inject with 3.7 kW when the frequency deviated 150 mHz or more. This threshold strategy reduced 98.7% of undesirable frequency deviations, which was slightly more than the in literature frequently described band gap strategy that caused a reduction of 96.4% and the band gap+ that resulted in a 93.4% reduction. The use of the threshold and band gap+ strategy also resulted in many switching events between the EV’s standby and charging/injecting mode, while the band gap strategy needed only one charging/injecting event to stabilize the frequency deviation. Thus, this study showed that the choice for a control strategy in DSGC is important, since different control strategies affect the grid’s dynamical stability differently.
show less