We are happy to announce that our recent paper entitled: “An A Priori Analytical Method for the Determination of Operating Reserve Requirements”, has been accepted at the International Journal of Electrical Power & Energy Systems (IJEPES). This study comes as a result of collaboration between three universities; Masdar Institute, Dartmouth, and MIT. The work is authored by Aramazd Muzhikyan (Masdar Institute), Prof. Amro M. Farid (Dartmouth) and Prof. Kamal Youcef-Toumi (MIT).
As renewable energy becomes an ever present resource in power systems, so called “operating reserves” become increasingly important instruments for reliable power grid operations. One can think of operating reserves as additional generation capacity scheduled to compensate for real-time power supply and demand imbalances due to the existing uncertainties in forecasting not just demand but also renewable energy. On the one hand, the amount of operating reserves should be sufficient to successfully mitigate the real-time imbalances and maintain power system reliable operations. On the other hand, operating reserves are a costly commodity and they should not exceed the minimum required amount to avoid unnecessary expense. This makes accurate assessment of the operating reserve requirements vital for reliable, economic, and environmentally friendly operation of the power grid.
Currently, the necessary amount of the operating reserves is assessed based upon the power system operator experiences and the assumption that the circumstances of power system operations remain relatively unchanged. However, growing integration of renewable energy sources (RES), implementation of demand side management and transportation electrification alter the overall structure and the dynamics of the power grid. High penetration of RES brings new levels of variability and uncertainty to the grid which challenges the established practices of power system operations and the operating reserve requirement assessment methods. This newly published article provides closed-form analytical formulae that tells grid planners how much reserves to procure as they plan for more renewable energy without sacrificing economics or reliability.
While RES integration can potentially reduce the grid’s CO2 emissions and operating costs, it also brings new challenges that power grid operators need to address in order to maintain reliable operations. Wind power, for example, is known to have high intermittency; that is, the output power of a wind turbine may vary uncontrollably in a wide range. This, combined with comparably low wind forecasting accuracy, requires careful scheduling of traditional power plants and their operating reserves. Integration of solar power, on the other hand, has its own challenges. As shown in the figure below, the net load profile (the power demand minus the solar generation) of a system with integrated solar generation has a distinctive profile. It is often called the “Duck Curve” for its resemblance to the side-profile of a duck. The figure presents the net load profiles of the California Independent System Operator (CAISO) for the day of March 31 for forecasted from 2014 to 2020. The “belly” of the curve corresponds to the day time when the solar generation is at its maximum and is expected to grow with new solar power installations. With an estimated demand of 22,000MW in the year 2020, the solar generation accounts for 10,000MW or 45%; leaving only 12,000MW for the traditional generation. This situation increases the risk of overgeneration and solar generation curtailment. Another challenge is the steep jump of the net load around 6pm as solar generation wanes with the sunset and demand picks up for evening home life. Such severe variations of the net load require more careful consideration of the ramping capabilities of the scheduled generation.
The CAISO duck chart (source: P. Denholm, M. O’Connell, G. Brinkman, and J. Jorgenson, “Overgeneration from Solar Energy in California: A Field Guide to the Duck Chart,” National Renewable Energy Laboratory, Nov. 2015)
This publication has developed analytical formulae for calculation of the requirements for each type of operating reserves; namely, load following, ramping and regulation. The derivations show that the operating reserve requirements are effectively defined by a set of dimensionless parameters related to the RES characteristics and the operations of the power grid. Those parameters are the penetration level, renewable energy capacity factor, variability, day-ahead and short-term forecast errors of the integrated RES, and the power grid day-ahead scheduling and real-time balancing time steps. Such analytical expressions reveal how the requirements of each type of reserve will change when, for instance, more renewable energy is integrated, renewable energy forecasting accuracy is improved, and the day-ahead scheduling time step is reduced. This study show that higher RES variability significantly increases the requirements of all three types of reserves. Also, while the impact of the RES forecast error on the ramping reserve requirement is negligible, its impact on the load following and regulation reserve requirements can dominate that of the variability. On the other hand, reducing the day-ahead scheduling time step can mitigate the impact of the variability on the load following reserve requirement while having negligible impact on the ramping and regulation reserve requirements. Also, changing the balancing time step has no noticeable impact on the load following reserve requirement, it has opposing impacts on the ramping and regulation reserve requirements. Reducing the balancing time step reduces the regulation reserve requirement but increases the ramping reserve requirement.
These formulae can be used for renewable energy integration studies, such as those conducted in NE-ISO and PJM-ISO, to assess the required amount of reserves for the planned RES installation. They can also be adapted by the state and federal standards organizations to establish reserve procurement standards that reflect the evolution of the power grid.
In depth materials on LIINES smart power grid research can be found on the LIINES website.