Finding human behaviour in hybrid heat pump power usage

Abstract

Climate change is causing countries to look for change and improvement in different sectors, including the heating industry. In the Netherlands, this is manifested, among other things, in a transition from the use of gas to electricity. Heat pumps could be a part of the solution to this problem, and the Dutch government has therefore initiated subsidies. However, not every house is suitable for a heat pump. For this reason, this thesis investigated the performance of hybrid heat pumps by searching for human behavioural patterns. The study demonstrates that by making use of predictive modelling, it is possible to forecast power usage trends. By doing so,we produced residuals with reduced influence of temperature, thereby enabling us to use Fourier analysis to look for human behavioural patterns. Although we found patterns that yielded indications, we could not give conclusive evidence that we found human behavioural patterns. However, we observed differences in how different heat pumps behaved under similar conditions. These findings suggest that hybrid heat pumps may not be suitable for every house included in this study. Further investigation is necessary to prove if our findings were signs of human behavioural patterns and how the differences in behaviour effects hybrid heat pump performance.