Master's Thesis Hannes Engel


Development of a procedure for dynamic evaluations of heat pump systems

The number of domestic heatings that include electrically driven heat pumps is on the rise
over the last years. For the decision whether the investment in a heat pump system is economically
wise accurate predictions for the efficiency of the heat pump system are essential.
The performance of heat pump systems is affected by a number of influencing factors, most
importantly the temperature lift. Current testing procedures take a static approach on the
evaluation of heat pump systems. Dynamic effects like start-up behaviour of the heat pump
are not taken into consideration.
This thesis investigates approaches for a dynamic testing procedure for heat pump systems
on a Hardware in the Loop test bench. The idea is to test the heat pump system for typical
boundary conditions. The heat pump system is installed inside of a conditioned laboratory
and connected to the test bench. A simulation model of a virtual building, including internal
loads and realistic weather conditions, is used to emulate the heat demand of a dwelling,
which is covered by the heat pump system. Subsequently, the heat pump system is tested for
a number of typical test days. The aim is to receive results for the efficiency and the provided
heat as accurate as possible, whilst keeping the testing procedure as short as possible. In order
to achieve this it is essential that the testing procedure emulates the dwelling for weather
conditions that are representative for the whole period of the operating time of the heat pump
system. Different approaches for the preparation of typical weather data are investigated.
For the assessment of possible methods to provide typical weather conditions for the testing
procedure a simulation model of a dwelling is used. The virtual building is equipped with
a heat pump system, emulated by a grid-based simulation model. Subsequently, reference
simulations of the virtual building are carried out for the complete period of operating time
of the heat pump system. Thereafter, the testing procedures are applied to the simulation
model and simulated as well. The simulation results received from the reference simulation
and the procedure simulation are compared. The deviation between the two simulations is
used as an indicator for the suitability of the investigated testing procedure. Possible testing procedures are investigated for four different heat pump systems, including air source and
solar combisystems, both with and without supply of domestic hot water. Promising testing
procedures then are applied to a ground-coupled heat pump system.
The simulation results for the efficiency of the heat pump system show that the Typical Days
according to Least Squares, Typical Days according to k-medoids Clustering and the Simulated
Selection Method are able to provide results with a relative deviation of less than 4 %,
between reference simulation and simulation of the testing procedure. The simulations do
not consider the affect of changing user behaviour between seasons and between weekdays
and weekends. Additionally the influence of different insulation standards it not investigated,
nor is the influence of different weather conditions for the reference simulations.