Resumen de Demand response in electric systems: its contribution to regulation reserves and its role in aggregating distributed energy resources
Electric power systems have experienced major changes since policy makers have opted for fostering renewable energy technologies. As a result many countries are experiencing a surge of renewable energy technologies within their generation mix. This development entails a series of challenges in the electric system operation. Renewable generators are distributed across the system and are typically substantially smaller in capacity size than conventional thermal generation units. These units have intermittent generation profiles. As such, variability in electricity production, uncertainty in forecasting and little controllability must all be taken into account. The leitmotiv of the options which exist to adapt to these systems is flexibility. During operation of the electric power system flexibility is required in order to respond to rapidly changing system conditions. Shiftable demand is one source of flexibility with high potential. Given the advances of smart grid technologies, for example smart metering, this potential has been relatively untapped to date. This thesis focusses upon the response of electricity consumers to actual electric system conditions.
To begin I analyse the responsiveness of demands in energy markets in the context of the day-ahead planning phase. I estimate how flexible demands may support the integration of intermittent generation relating to high wind penetration and analyse in detail the factors impacting Demand Response at the system level. Based on the considered case study, one can conclude that in a system with high wind energy production, flexible demands, and in particular those that shift their demand, can be useful to partially level out variations in wind production, which is relevant on days with extremely high wind production. Furthermore, various factors are critical concerning the impact of flexible demands: a high Demand Response potential of considered devices and the time-overlap of the electric load of these devices with adjacent hours to peak and off-peak hours of electricity consumption. Another critical factor is the Demand Response cost borne by or assigned to the consumer. Already low Demand Response cost levels result in strongly reduced net benefits for consumers. This discourages the participation in Demand Response programs, and therefore reduces significantly the positive impact of Demand Response on system outcome.
Afterwards, I analyse two specific issues that may concern the reliability of system operation as well as integration of distributed energy resources. To maintain reliability, reserve must be procured, typically from thermal or hydro plants. Demand Response is able to provide reserve and I quantify the impact on welfare this could have in the Spanish electric system considering the joint existence of Demand Response in both energy and reserve markets. According to results obtained, when demand is allowed to provide reserves, social welfare is increased, operation costs and emissions are reduced, and a higher level of integration of renewable generation is achieved. Considering the parallel application of several DR mechanisms, I find the provision of reserves by demands to be far more efficient than only shifting electricity consumption from a system point of view.
Last, using the concept of Virtual Power Plants, I examine how Demand Response may help in the system-wide integration of distributed energy resources in mainland Spain. Results for different strategies these Virtual Power Plants can pursue are obtained and used to estimate how flexible demands in Virtual Power Plants may support the integration of distributed energy resources. The results of this analysis indicate that the chosen strategy of distributed generation and demand in Virtual Power Plants has little impact on the rest of the electric system in terms of thermal operation costs, emissions or renewable energy spillages, but a high impact on the results relevant for the Virtual Power Plant¿s corresponding strategy: benefit maximising Virtual Power Plants can more than double their benefits and auto-consuming Virtual Power Plants can achieve very high rates of self sufficiency in the supply of local load. These results indicate that Virtual Power Plants should be considered as a viable and interesting option to integrate renewable energies into the electric energy system.
I conclude that Demand Response is an option to enhance electricity system operation and reliability. As a case study I have assessed the specific case of the mainland Spanish electricity system, which has been analysed in depth. Demand Response is interesting for more than just achieving cost reductions. The rise of the active consumer with his knowledge of electricity as a good with time sensitive prices based upon underlying system conditions can only be beneficial to the system.
