Control system of intelligent buildings based on convolutional neural networks

In recent years, with the development of ”Internet of things” appeared another concept – smart home environment, which is defined as the physical infrastructure to operate of ambient intelligence. For the effective operation of such facilities requires to develop of new methods and approaches for adaptive control. In this paper, we analyzed the parameters characterizing the state of the control object, identify the necessary set of measured parameters obtained from the sensors, and an informative set of input parameters. We design method of training of neural network.