Perm Tech scientists have developed an intelligent module to control the local heat supply system. Neural networks will help to calculate the temperature of the coolant at the leave from the boiler room accurate and quick. The technology allows to keep the temperature satisfactorily, and to avoid unreasonable overheating of the coolant and decrease the heating costs. The development has no analogues in Russia yet.
Nowadays, control units are widely used, that makes it possible to maintain the set temperature at the leave from the boiler room automatically. The required values ??are determined by the operator, mainly focusing on the thermometer and available feedback. The development assumes control with the help of such neural networks, which use in calculations not only the current value of the ambient temperature, but also a intelligent forecast. One can estimate the carrier temperature in advance and avoid "lagging".
To "train" the neural networks, scientists used a large amount of statistical data. It includes synchronized coolant temperatures at various points of the heating networks and ambient temperatures.
Scientists have tested an intelligent module by integrating it into the hardware and software automated control system "Aurora. Heat balance in housing services and utilities ", which was developed and used by one of the companies of the Perm Krai. As a result, the unit makes it possible to regulate the temperature of the coolant at the leave from the boiler room automatically, taking into account the forecast of changes in weather conditions.
According to scientists, the use of a neural network in the processes of managing a heat network allows to save fuel and prevent overspending. This effect becomes especially significant when it come to the sharp changes in the weather. Gas conservation can reach up to 10-15%, depending on the outside temperature and the general conditions of the heating systems.
Multi-layer neural network and deep learning networks are able to predict the required boiler temperature, regarding the weather forecast and the characteristics of the movement of the coolant.
During the creating an intelligent module, scientists analyzed various types of neural networks. The resulting architecture consists of 224 neurons arranged in three layers. The calculated temperature of the coolant at the leave from the boiler room provides those values of the temperature at the entrance to the house, which are required by the standards.
