This article presents an intermittent heating controller which is designed
for calculating the supply temperature of hot-water-radiator heating system
s. This controller uses a new heating law which has been obtained by invert
ing a simplified model for calculating the internal temperature of a buildi
ng in relation to the external temperature, as well as the supply temperatu
re of the distribution circuit and the temperature of the structure. The la
tter is used to determine the thermal state of the building, which is an im
portant parameter in the evaluation of energy needs when a building is heat
ed in an intermittent made. During non-occupancy periods, the parameters of
the model are automatically modified by a recursive least-squares routine.
A second law optimizes the heating restart time during non-occupancy perio
ds; this law also incorporates a learning mechanism.
The controller's performance is analysed, and the results are compared with
those of a controller using a classical heating law which does not take in
to account the thermal state of a building. Numerical simulations demonstra
te that the new heating controller leads to performances comparable with th
ose obtained by classical methods, but for a higher degree of intermittence
. This means that the comfort obtained is similar, but that the energy cons
umption is lower. The new heating law also proves to be advantageous in oth
er ways; it provides faster learning, much better lower temperature control
and a reduction in instances of overheating during inter-season periods, w
hen little heating is needed.