The paper considers concepts of design and optimization of intellectual network of direct electric heating. Queuing techniques together with classical approaches describing thermal processes have been applied to describe and analyze processes taking place within the network.
Intellectual network of direct electric heating is a system of electric heaters located within different areas of an object being heated; the heaters are connected to a common energy source by means of specific automation devices that is “smart” connectors. While sharing information on a real time basis, latter ones generate a subset of local heat sources; total power of the heat sources is not more than randomly time-controlled power limit meant for heating purposes. Structure of the subset varies in such a way to provide preset temperature distribution within area heated with the help of the devices. The network experiences reconfiguration automatically as soon as either power surplus or power deficit occurs.
Following out of the ordinary result has been obtained: selective connection of certain quantity of heaters which total power is equal to heating power limit is more efficient. In the context of the considered example of 3-room apartment heating where power limit is from 2 to 4kW and power of heaters is 1500W only 2 of 4 heaters can be connected simultaneously. In this regard average heating power output is 35% more than the power output by four heaters with equivalent power 750W each connected without waiting in queue.