The scientists of the Perm Polytechnic have developed an intelligent module for managing the local heat supply system. Neuralati will help accurately and quickly calculate the temperature of the coolant at the exit of the boiler room. The technology allows you to maintain it in terms of consumers, avoid unreasonable overheating of the coolant and save funds on heating. The development has no analogues in Russia yet.
Now the control units are quite widely used, which automatically support the specified temperature at the outlet of the boiler room. The required values defines the operator, mainly focusing on the thermometer and the available feedback. Our development involves control using such neural networks, which are used in the calculations not only the current value of the ambient temperature, but also a reasonable forecast. This allows you to pre-evaluate the temperature of the carrier and avoid delay, says the associate professor of the department of computing mathematics, mechanics and biomechanics of the Perm Polytech, Candidate of Technical Sciences Vladimir Onistkiv.
For teaching neuralo, scientists used a large amount of statistical data. It includes synchronized coolant temperatures at various points of the thermal network and ambient temperature.
Scientists have tried the intelligent module by typing it in a software and hardware automated Aurora control system. The thermal balance in housing and communal services, which has developed and uses one of the companies of the Perm Territory. As a result, the complex allows you to automatically adjust the temperature of the coolant at the outlet of the boiler room, given the forecast for changing weather conditions.
To ensure comfortable thermal conditions in consumer homes, heat supply organizations must constantly monitor the temperature state of the network. But this service is still unavailable for most thermal companies, so they insure their risks, maintaining higher thermal carrier temperature. As a result, residents are often forced to overpay for utilities, explains the researcher.
According to scientists, the use of neural network in the process of controlling the heat network allows you to save fuel and prevent its overrun. With sudden weather changes, this effect becomes especially significant. Gas savings can reach 10-15%, depending on the outer air temperature and the overall state of the heat network.
Multilayer neural networks and deep learning networks are able to predict the necessary boiler temperature, given the weather forecast and features of the coolant movement.
In the process of creating an intelligent module, scientists analyzed various types of neural networks. The final architecture consists of 224 neurons, ordered in three layers. The calculated temperature of the coolant at the outlet of the boiler room provides those temperature values at the entrance to the house that the standards are required.