Application of New Forecasting Models Based on Artificial Intelligence, New Materials and New Mechanisms to Improve the Grounding System of Supports

Objective: to improve the grounding system for AC contact network supports by developing a device for remote monitoring of the state of spark gaps and gas-discharge protection devices, as well as substantiating the use of intelligent AI agents to process diagnostic data and transition from reactive to predictive maintenance. Methods: comparative analysis of existing control methods. A constructive solution based on a short- circuit indicator is proposed, in combination with unmanned aerial vehicles and intelligent video analytics tools. Results: the proposed solution provides contactless remote diagnostics without decommissioning the equipment and does not require significant design changes. Practical significance: continuous monitoring, early detection of hidden defects and degradation processes, increased electrical safety, optimization of operating costs and reduction of the burden on the personnel of the power supply service.