Long-term sustainable operation of hybrid geothermal systems through optimal control

Abstract

Hybrid geothermal systems such as hybrid GEOTABS typically comprise a geothermal heat pump that supplies the main building thermal energy needs, complemented by a fast-reacting supplementary production and/or emission system for the peak building thermal loads. Optimal predictive controllers such as Model Predictive Control (MPC) are desired for these complex systems due to their optimized and automated energy savings potential (while providing the same or better thermal comfort) thanks to system integration and their anticipative action. However, the predictions of these controllers are typically limited to a few days. Consequently, the controller is unaware whether abusive energy injection/extraction into/from the soil will deplete the source over the years. This paper investigates in which cases the long-term dynamics of the borefield ought to be included in the MPC formulation. A simulation model of a hybrid GEOTABS system is constructed. Different borefield sizes, ground imbalance loads, and electricity/gas ratios are evaluated. The model control inputs are optimized to minimize the energy use in 5 years through (i) a reference Optimal Control Problem (OCP) for the 5 years, solved in hourly timesteps and (ii) an MPC control with a prediction horizon of 1 week. The obtained results reveal that MPC can be up to 20% far from the true optimal, especially in the cases where the borefield is undersized and there is a large cost gap between the different energy systems.