USING ARTIFICIAL INTELLIGENCE FOR MONITORING UNAUTHORIZED CONNECTIONS TO CONSUMER POWER NETWORKS

In 2024, the issue of «electricity theft» remains relevant in Kazakhstan, especially amid the projected electricity shortage, which, according to JSC «KEGOC», may exceed 1.025 GW. This situation forces the government of the Republic of Kazakhstan to consider measures such as importing electricity from the Russian Federation or introducing consumption restrictions for domestic users. With increasing tariffs and resource shortages, illegal connections place an additional burden on the energy system, leading to financial losses for energy companies and reducing the reliability of power supply. This article examines a monitoring system for electricity consumption aimed at detecting «unauthorized connections» and excessive power usage. An AI-based automated management system analyzes electricity consumption in real-time, comparing it with the nominal average value stored in a database. If the consumption exceeds the set limit by 10%, the program triggers an alarm by sending notifications via SMS to the subscriber’s phone, the Telegram messenger, and activating an alarm signal on the control panel or management system. If the consumer fails to respond within 10 minutes, the system automatically disconnects power on all phases, preventing possible overloads and unauthorized consumption. AI-based automated monitoring systems can track electricity consumption in real-time, detect anomalies, and quickly respond to unauthorized connections. This capability helps minimize financial losses, enhance control over power grids, and improve power supply reliability. AI-driven information technologies are becoming key tools in preventing electricity theft and unauthorized electricity usage, including in the Republic of Kazakhstan. Keywords: artificial intelligence, unauthorized access, energy consumption, automation of management, monitoring.