The reliability of electrical networks relies heavily on service continuity and data quality from distribution stations. However, measurement errors and data inaccuracies complicate anomaly detection using conventional methods. This study applies fuzzy logic to analyze the monthly energy balance (2019–2023) of the Kasangulu electrical substation. The methodology involves fuzzifying electrical quantities (delivered energy, energy sold, and efficiency) and designing a Mamdani-based fuzzy inference system. The results demonstrate that the fuzzy logic approach effectively handles data uncertainty and measurement inaccuracies, providing more precise anomaly detection than deterministic techniques. Consequently, this model serves as an efficient decision-making tool for network monitoring and predictive maintenance at the Kasangulu substation.
Copyrights © 2027