Civil Engineering Reference
In-Depth Information
of energy consumption, utilisation and waste. The output of these models provides
valuable information for the simulation of a single building. Central to this set is
the stochastic model of occupant presence. The stochastic models of occupant
behaviour regarding the use of the lighting system imposes no loss of generality as
it can always be defined to other type of devices, represent the variety of occupant
behaviours and their randomness over time.
The obtained results demonstrate that it is important to consider the behaviour
analysis of occupants within the building, as different occupancy patterns result in
different patterns of energy consumption. Moreover, the obtained results are
aligned to hypothesis of applying stochastic models to represent either building
occupant behaviour or the space occupancy patterns.
The proposed ECM for predicting building energy consumption produces
reliable predictions. Additionally, it can accurately capture space utilisation and
predict different scenarios and support reliable decision making for investing in
potential areas for energy savings. An important aspect being refined in further
developments is the ECM simulation. The simulation results show better results
with the dynamic approach. However, this algorithm is based on a search method
and it does not guarantee the best prediction. Additionally, as the size of the
building increases, the dynamic algorithm becomes computationally complex.
Another important improvement is extending this ECM to account for multiple
devices in the building's spaces.
Although the proposed methodology seeks to improve and support retrofitting
projects in existing buildings, the results can be extrapolated to support new
building, e.g., buildings in a design or project stage. That is, buildings with the
same activities, in the same geographic area and occupied by people that share the
same cultural background are expected to have similar energy usage patterns. In
this case, previous obtained retrofitting project results can be used at an early stage
to promote the installation of energy efficient technology and improving the energy
efficiency of new buildings.
Acknowledgments The presented work has been partly supported by the European Commission
through ICT Project EnPROVE: Energy consumption prediction with building usage measure-
ments for software-based decision support (G. A. FP7-248061).
References
Abreu JM, Ferrão P, Pereira FC (2012) Using pattern recognition to identify habitual behavior in
residential electricity consumption. Energy Buildings 49:479-487
Spinar R et al (2008) Efficient building management with ip-based wireless sensor network. Cork
Atallah L, Yang G (2009) The use of pervasive sensing for behaviour profiling—a survey.
Pervasive Mobile Comput 5:447-464
Borgeson S, Brager G (2008) Occupant control of windows: accounting for human behaviour in
building simulation. University of California, Berkeley
