Energy infrastructure is a major factor in sustainable economic development, world and social stability in the world, especially during periods of economic development. The transformation of energy and transport infrastructure affects the achievement of the goals of the transition to a low-carbon economy. However, due to excessive energy consumption, the physical wear and tear of infrastructure is growing at a high rate. Creating a secure infrastructure for the future is essential to achieving sustainable development and ensuring energy efficient consumption of energy. The use of digital and information technologies has a positive effect on the development of energy efficient management of energy industry facilities and will create prerequisites for the growth of investments in the modernization of infrastructure and its re-equipment. Therefore, an important role in the management of the life cycle of the energy infrastructure is the management of the reliability and energy efficiency of the equipment of electromechanical and electrical complexes of consumers. In the report, based on the analysis of modern structures and algorithms for controlling the electric drive, the most promising areas in the field of digital technologies were selected, which will have a significant impact on the energy efficiency and reliability of the electromechanical systems of oil producing enterprises. The existing methods of monitoring and predicting the technical conditions of electromechanical equipment are analyzed and the expediency of creating an intelligent system for diagnosing electromechanical systems is substantiated. To control the level of energy efficiency, it is proposed to assess the technical condition and energy characteristics of operating modes based on the analysis of electrical parameters. The results of experimental studies and simulation modeling are presented. A technique for determining the type of damage and classifying the operating modes of a group of electromechanical consumers based on the analysis of electrical parameters data using artificial intelligence tools for classifying and searching for the level of damage is considered.