The development of active-adaptive electrical networks with an intelligent control system involves the creation of energy information complexes with the possibility of continuous monitoring and remote control of the operating modes of all its components in order to optimize network parameters. This necessitates the need to adapt educational programs and training technologies to train specialists with skills in related fields. The paper describes a developed research complex with elements of artificial intelligence for performing research and studying intelligent systems and means of controlling the modes of power supply systems based on a set of educational equipment using real and virtual objects of electric power systems with adjustable parameters. Pychram Community Edition was chosen as the integrated development environment. The software part of the complex has been developed, including a digital twin of the stand, an executive part and a neural network model. The neural network allows you to optimize the parameters of an active-adaptive electrical network losses during electricity transmission.