Sustainable Building Operation Using Simulation - Low Energy Cooling for Sustainable Buildings - page 226

15%. More important are the different internal loads, which can increase the required
collector surface area by nearly a factor of 2. The best solar thermal efficiency was
45% for a high full load of nearly 2000 hours. For the location in Madrid, 80% solar
fractions are possible for surface areas between 2 and 4m 2 per kilowatt of cooling
power, the high values occurring for larger full load hours. For each megawatt hour
of cooling energy demand, between 1.6 and 3.5m 2 of collector surface are required
for the Spanish site and between 4.6 and 6.2m 2 for the German installation. The total
system costs for commercially available solar cooling systems are between D 180 and
270 per MWh, again depending on the cooling load file and the chosen control strategy.
The total costs are dominated by the costs for the solar thermal system and the chiller
itself. For a more moderate climate with low cooling energy demand, the costs can
rise as high as D 680 per MWh. The work shows that dynamic system simulations are
necessary to determine the correct solar thermal system size and to reach a given solar
fraction of the total energy requirement.
6.2 Online Simulation of Buildings
After considering the use of simulation tools for the planning of solar cooling plants,
the issue of online simulation and control of buildings and plants will be discussed next.
This is especially important in large, complex buildings, which often are equippedwith
building management systems. Today, building management systems (BMS) are com-
monly designed to control the technical building equipment in order to reach comfort-
able climatic conditions. This setpoint-orientated control strategy does not normally
contain any active supervisory instruments to control the energy consumption of the
building. As a consequence, no error messages will appear as long as the setpoints are
reached, sometimes even if in the worst case the cooling and heating systems are work-
ing against each other. Furthermore, standard BMS control algorithms are only able
to detect abrupt changes in the conditions of, for example, HVAC systems, but they do
not offer detailed fault detection and diagnostic information and are unable to detect
gradual degradations in system performance (Buswell et al. , 2003). On the other hand,
dynamic simulation tools have so far only been used during the planning phase for
the building design and the dimensioning of technical equipment like the heating and
cooling system. Control algorithms can then be well developed when the simulation
results and the planned equipment are considered, but the implementation is later nor-
mally done by the system engineers of the chosen BMS company. Due to information
losses, interpretation and implementation errors are nearly unavoidable, which leads
to suboptimal system control. To reduce such problems new strategies are required to
implement directly well-tested control algorithms from simulation programs into the
building control system. During building operation the simulation tools can then be
used for instance in the energy management system for online simulation and control,
and to check the control actions and measurement data against the simulation results.