CH-24-C028 - Flexibility Potential of Residential Heat Pumps in the Power Grid of the Future

Renewable energy sources such as solar and wind power will play an increasing role in the power grid. While generating low carbon electricity at an attractive cost, their intermittency poses challenges to the balance of our electrical system. During moments of high energy use but low availability of renewable energy, e.g. winter evenings, the most polluting fossil energy generators need to be active in order to fulfill the peak consumption. Conversely, during moments of low energy use but high availability of renewable energy, e.g. summer days, the excess energy will have to be curtailed. With the rise in heat pump adoption, the electrification of heating systems might increase these challenges if we fail to devote adequate attention to the power consumption pattern. Yet, using intelligent control strategies can turn heat pumps from a system liability into a system asset. The success of large-scale industrial flexibility contributions to the energy system demonstrates that intelligent demand-response plays a key role in our current and future energy supply. Extending that role to small-scale distributed assets such as heat pumps is however not yet thoroughly investigated. In this paper, the flexibility potential of heat pumps in the Belgian residential building stock is explored. The flexibility is expressed as the electricity which can be postponed by turning off the heat pumps during a given period of time. User comfort is taken into account, and minimum thresholds for the room temperature are defined. Dynamic building energy simulations are run for buildings with different parameters using RC grey-box models. We investigated dwellings of different typology and size. Multiple insulation levels were applied, as well as different heat emission systems (radiator vs. floor heating) and heating schedules. A cold, average and warm climate year are used for the simulations, as well as different minimum temperatures. Based on data from government agencies, the building stock is divided into categories of building typologies and insulation levels. Using various data sources and making several assumptions, we defined new built rates, renovation rates and heat pump adoption to estimate the building stock in the year 2030 and 2040. The results from the individual building simulations are extrapolated to the building stock level and different climate and comfort scenarios are evaluated. According to our analysis, deactivating all Belgian heat pumps during a December month results in a initial load reduction of 2500 MW (8530 MMBtu/h) for the 2040 scenario. This value gradually reduces in the hours after the flex event, since certain buildings reach their minimum temperature threshold and restart the heat pump. After around 9 hours, a period of load increases takes place since the building stock requires additional heating compared to the baseline scenario. Load increases up to 1000 MW (3412 MMBtu/h) are registered. These values are largely influenced by assumptions regarding the building stock, as well as boundary conditions such as climate and minimum temperature threshold.