Sustainable Building Operation Using Simulation - Low Energy Cooling for Sustainable Buildings - page 199

primary energy (Mendes et al ., 1998). The exact value of the minimum solar fraction
required for energy saving depends not only on the performance of the thermal chiller,
but also on other components such as the cooling tower: a thermal cooling systemwith
an energy-efficient cooling tower performs better than a compression chiller at a solar
fraction of 40%; a low-efficiency cooling tower increases the required solar fraction
to 63%. These values were calculated for a thermal chiller COP of 0.7, a compressor
COP of 2.5 and an electricity consumption of the cooling tower between 0.02 and
0.08 kWh el per kWh of cold (Henning, 2004b).
Double effect absorption cycles have considerably higher COPs around 1.1-1.4, but
require significantly higher driving temperatures between 120 and 170 ◦ C (Wardono
and Nelson 1996), so that the energetic and economic performance of the solar thermal
cooling system is not necessarily better (Grossmann, 2002).
Several authors have published detailed analyses of the absorption cycle perfor-
mance for different boundary conditions, which showed the very strong influence of
cooling water temperature, but also of chilled water and generator driving temperature
levels on the COP(Engler et al ., 1997; Kim and Machielser, 2002). Most models re-
solve the internal steady-state energy and mass balances in the machines and derive the
internal temperature levels from the applied external temperatures and heat transfer
coefficients of the evaporator, absorber, generator and condenser heat exchangers. To
simplify the performance calculations, characteristic equations have been developed
(Ziegler, 1998),which are an exact solution of the internal energy balances for one
given design point and which are then used as a simple linear equation for different
boundary conditions. For larger deviations from design conditions or for absorption
chillers with thermally driven bubble pumps, one single equation does not reproduce
accurately the chiller performance (Albers and Ziegler, 2003). The disadvantages of
the characteristic equation can be easily overcome if dynamic simulation tools are
used for the performance analysis and internal enthalpies are calculated at each time
step - an approach which is chosen in this work using the simulation environment
INSEL (Schumacher, 1991). Dynamic models taking into account the thermal mass
of the chillers are also available and can be used for the detailed optimization of
control strategies such as machine start-up (Kim et al ., 2003; Willers et al ., 1999).
However, for an energetic analysis of annual system performance, steady-state models
are sufficiently accurate.
The available steady-state models of the absorption chillers provide a good basis for
planners to dimension the cooling system with its periphery such as fans and pumps,
but they do not give any hints for the dimensioning of the solar collectors or any
indication of the solar thermal contribution to total energy requirements. This annual
system performance depends on the details of the collector, storage and absorption
chiller dimensions and efficiencies for the varying control strategies and building load
conditions. An analysis of the primary energy savings for a given configuration with
renewable energy heat input requires a complete model of the cooling system coupled
to the building load with a time resolution of at least 1 hour. If stratified storage tanks
are modelled, the time resolution has to be even higher (10 minutes or less).