Interview with Charis Demoulias and Milos Cvetkovic
Charis Demoulias (M’96–SM’11) was born in 1961. He received the Diploma and Ph.D. degrees in electrical engineering from the Aristotle University of Thessaloniki, Thessaloniki, Greece, in 1984 and 1991, respectively. He is currently an Associate Professor with the Electrical Machines Laboratory, Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki. His research interests include the fields of power electronics, harmonics, electric motion systems, and renewable energy sources.
Miloš Cvetković received the B.Sc. degree in electrical engineering from the University of Belgrade, Serbia, in 2008, and the M.Sc. and Ph.D. degrees in electrical and computer engineering from Carnegie Mellon University, Pittsburgh, PA, USA, in 2011 and 2013, respectively. From 2014 to 2016, he was a Postdoctoral Researcher with the Massachusetts Institute of Technology, Cambridge, MA, USA. He is currently an Assistant Professor with the Electrical Sustainable Energy Department, Delft University of Technology, The Netherlands. His research interest includes the development of co-simulations for energy grids, and modeling for control and optimization of the electricity grids.
In this interview, Charis and Miloš answer questions from their webinar, Ancillary Services offered by DRES connected in distribution grids, originally presented on March 26, 2020. For more details regarding these questions, please view this webinar on-demand on the IEEE SG Resource Center.
On slide 9, why did the red curve (Inertial) show a spike at the time of disturbance?
Because we assume an inertial reaction that corresponds to true inertia. At the first instant of the frequency deviation the Rate-of-Change-of-Frequency (RoCoF) is very high. The inertial response is proportional to inertial time constant H and the RoCoF, therefore, for high RoCoFs a spike-like active power variation may be expected.
In slide 14, point 2 suggests that DRES should operate with a headroom. In this case, will it not mean DER is not being operated with MPPT? Is there any other way to increase the headroom by still operating at MPTT?
Yes, it means that the DER operates below its Maximum Power Point (MPP). In the hypothetical case that the DER operated at the MPP, a storage system would be required to provide the additional power for participating in the Primary Frequency Regulation. The latter solution is adopted by the TSOs (Transmission System Operators) of weak power systems because they cannot force the DERs to operate with a headroom. The TSOs of large and stiff power systems engage more conventional Synchronous Generators (SGs) to act as frequency containment reserves (FCRs) in order to compensate the power imbalances incurred by the volatility of Distributed Renewable Energy Sources (DRES). Obviously, this is a centralized approach. The same is true for the use of large-scale Battery storage systems employed in weak transmission systems. Concluding, there is no other way to increase the headroom and keep on operating at the MPP.
In slide 14, can you please comment on the ramp-rate limits and the impact it has on the converter control design?
Limiting the electrical power of the DRES means that there is an instantaneous imbalance between the active power provided by the primary source of the DRES and the electrical outputs. Therefore, a storage system is needed to compensate such imbalances. Regarding the converter control design, the following modification is necessary: the setpoint of the reference value of the electrical power is determined by following a given ramp-rate control. This is different from the current situation, where the electrical power is equal to the power of the primary source (wind, PV) and this is reflected in the stability of the DC-link voltage of the power converter. In the ramp-rate control method the DC-link voltage is controlled by the Fast Storage System (FSS) – e.g. an ultracapacitor- which, in this way, absorbs or releases the instantaneous imbalances between the input and output active powers.
In slide 21, to inject the "P" to match the frequency change or in accordance to ROCOF, the PV converter will have to be sized much bigger than the actual requirement. Further, will the power still be not limited by the current carrying capability of the converter?
This is generally true. However, the inertial response has a duration of a few seconds – typically 3-10 seconds. Therefore, the oversizing of the PV converter is not proportional to the additional power injected during the inertial response. It is actually determined by the short-term overloading capability of the switches of the PV converter (usually around 2.5-3 pu) and on the operating point of the converter at the time of a frequency event. The worst case appears when the converter operates at its nominal power (thus, the temperature of the switches is close to their maximum allowable) and an inertial response is required. Concluding, a slight oversizing of the converter might be required when a DRES unit is assumed to provide inertial response while operating close to its rated power.
Inertial response does not exist in the wholesale market today as Ancillary Services for synchronous generators, what is your proposal for those who generator (hundreds of them per region) that are already providing inertia? Are you saying the new ancillary services will be applicable to all generators?
Yes. In case an inertial response market is established, then for reasons of fairness, all the units (including the conventional SGs) should be properly compensated. However, similar measurement and quantification methods should be applied for the SGs as well. It is reminded that currently no such methods exist.
For inertia response is it only supercapacitors that can be used or other types of energy storage as well?
We believe that supercapacitors are the best candidates since they can charge/discharge very fast, in other words, they have very high power density. This is exactly what is required for true inertial response.
Is the non-linear FCR curve a more advanced version of the standard linear droop P-f characteristics?
We have heard of non-linear FCR curves, but our knowledge is limited in this topic. Therefore, we feel unable to provide a justified answer for their advantages in application on DRES units.
What do the current regulations or Standards require from the DRES with respect to the Ancillary Services mentioned in this presentation?
Recent Grid Codes and Standards provide specifications for some of the services mentioned in the presentation, however, they are treated from the point of view of system support functions, meaning that the DRES are not remunerated to provide the services but they have to provide them as mandatory functions. More details on this issue will be presented in Part 2 of the Webinar.
Since already Synchronous Generators participate in Ancillary Services Markets as Frequency Containment Reserves (FCRs), why is it important to measure the participation of DRES as Frequency Containment Reserves? Isn’t there an established way to measure it in a similar way like Synchronous Generators (SGs)?
Although there are established methods to measure the contribution of large SGs as FCRs, it is necessary to develop new methods for the DRES because their contribution requires the synchronized measurement of their Maximum Power Point (MPP) power and because their headroom is time-varying, since their MPP is time-varying too. In addition, methods for the measurement and quantification of the aggregated Primary Frequency Response of a DRES portfolio consisting of numerous small DRES within the distribution system needs to be developed. This is important because FCRs appear in the ancillary service markets of the transmission system, while the small-power DRES are currently not “visible” to the TSOs.
Wind Turbines are required to limit their ramp-rates in some weak transmissions systems; are they paid for that?
No, they are not. This service is treated as a system support function by the respective TSOs – at least currently.
Are the dispatchable DRES connected to the grid via Synchronous Generators remunerated to provide Ancillary Services like central conventional Synchronous Generators connected to the transmission system?
No, they are not. They are simply remunerated like converter-interfaced DRES to inject their power/energy into the distribution system. They are not participating in any ancillary service market, because they are not “visible” by the respective TSO and because there are no ancillary services markets at the distribution system level.
In the technical literature for the DRES integration there exist many methods for: inertial response, primary frequency response, fault-ride through, harmonic mitigation, reactive power provision for voltage regulation. What is new in the UVSG? Do you implement known methods, or new ones are also being developed?
For some of the foreseen Ancillary Services, like inertial response, fault-ride through, contribution to faults and harmonic mitigation, we developed new methods. However, the most innovative part -in our opinion- is the integration of all the foreseen Ancillary Services in a unified control model of the converters with specific structure and priorities.
What are the time-frames for these Ancillary Services? Which of them are considered as dynamic and which as steady-state?
Inertial Response is a dynamic Ancillary Service lasting a few seconds. The same is also true for ramp-rate limitation performed by Fast Storage Systems, the contribution of fault currents and fault-ride-through. Primary Frequency Response corresponds to slower dynamics since it lasts several seconds (up to 30-40s). The exchange of reactive power for voltage regulation and the harmonic mitigation may be considered as quasi-steady-state ancillary services, since the time intervals involved are in the order of minutes. Exception is the dynamic reactive power support which is a rather new ancillary service (provision of reactive currents within few seconds in a voltage instability event).
To view past interviews, please visit the IEEE Smart Grid Resource Center.