Using fast-responding resources to control frequency in a power system




Peydayesh, Mansoureh

Journal Title

Journal ISSN

Volume Title



Frequency control is one of the major concerns of power system operators. Frequency varies as the result of a supply-demand mismatch. Due to possible destructive outcomes of large frequency variations, several mechanisms are in operation to keep supply and demand in balance. Increasing penetration of non-dispatchable intermittent generation resources may increase power supply volatility, which makes frequency control more challenging. Emerging utility-scale storage technologies with reasonable cost have participated in electricity markets in recent years. Because of fast-ramping capabilities of these resources, one of their attractive applications is providing frequency regulation service. However, the amount of energy they can produce or consume is limited due to their restricted storage capabilities. Thus, in spite of their fast response to a deployment signal, their duration of response is bounded. In this thesis, we focus on using fast-responding resources to control frequency in power systems. In this research, the first question is if the participation of these resources in the regulation market have any adverse effect on the frequency control performance of the system. If the answer is yes, the next question is what is the best strategy to not only prevent the negative consequences but also improve the benefits of using fast-responding resources for frequency control. For this research, the system of Electric Reliability Council of Texas (ERCOT) is selected. All power system studies related to frequency control require an appropriate dynamic model. In this dissertation, a simplified model is constructed, which represents the ERCOT system frequency response during a short period of time after a contingency. The model is validated and tuned against system frequency measured by Phasor Measurement Units. Especially in situations of not having information about system individual units, this simplified model is highly advantageous. However, to study system frequency during normal conditions, a more comprehensive model is essential. Thus, we develop ERCOT Frequency Modeling and Analysis Tool (EFMAT), which has the required level of details and accuracy to simulate system frequency. All proposed approaches of modeling and parameter tuning in this research are also applicable to other power systems. In order to answer our research questions, we start with investigation of ERCOT Fast-Responding Regulation Service (FRRS). For selected historic days, conventional regulation providers are replaced by a storage system providing FRRS. For various capacities of the storage system, frequency is simulated using EFMAT and a system frequency control performance index is calculated. Comparing calculated index of different simulations can reveal the effect of FRRS capacity on the system performance. The simulations are repeated for several FRRS deployment strategies similar to the strategies of other North America power markets along with our proposed modifications. Three different storage systems are assumed in the simulations: one with unlimited stored energy, one with 6 minutes energy duration, and one with 15 minutes energy duration. Finally, FRRS optimal capacity and equivalency ratio between FRRS and conventional regulation are defined and calculated for the best deployment strategy.


LCSH Subject Headings