Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
 MOSCOW +7(495)-142-86-81

A HYBRID REINFORCEMENT LEARNING, FUZZY LOGIC, AND EVOLUTIONARY MODEL FOR TRANSFORMER TEMPERATURE FORECASTING

Chernikov Aleksandr Vladimirovich  (postgraduate student, Federal State Budgetary Educational Institution of Higher Education "MSTU "STANKIN")

This paper presents a novel hybrid approach, FELAR, which integrates tabular Q-learning, socially-oriented fuzzy logic, and genetic hyperparameter optimization to accurately predict the oil winding temperature of power transformers. Using the ETTh1 benchmark dataset, we demonstrate that the proposed architecture achieves performance comparable to or better than state-of-the-art transformer models. The hybridization enables effective adaptation to thermal inertia and seasonal variations without sacrificing computational simplicity, allowing for online fine-tuning on edge devices without GPU acceleration. Experimental analysis of the error distribution and forecast-vs-actual scatter plots highlights FELAR’s robustness to right-skewed errors and overnight temperature dips, suggesting avenues for further calibration of fuzzy rule weights and evolution of strategy parameters.

Keywords:FELAR, ETTh1, Q-learning, fuzzy logic, genetic optimization, transformer temperature forecasting.
 

Read the full article …

Citation link:
Chernikov A. V. A HYBRID REINFORCEMENT LEARNING, FUZZY LOGIC, AND EVOLUTIONARY MODEL FOR TRANSFORMER TEMPERATURE FORECASTING // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2025. -№06/2. -С. 238-240 DOI 10.37882/2223-2966.2025.06-2.45
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"

Terms Publications

Issues of the journals (Archive)

ECONOMICS AND LAW
HUMANITIES
NATURAL AND TECHNICAL SCIENCES
COGNITION

Серия - Natural and Technical Sciences