Mitigating Load Imbalance and Enhancing Distribution Transformer Performance via Fuzzy Inference System

Load imbalance in distribution transformers presents a significant challenge to power system efficiency due to the generation of high neutral currents, which can compromise system stability. To address this issue, this study proposes a fuzzy logic-based approach using a Mamdani-type Fuzzy Inference System (FIS). The input variables, current magnitudes in phases R, S, and T, are represented by trapezoidal membership functions classified into several categories. The output variable is defined as corrective current adjustments for inter-phase load redistribution. The methodology includes transformer data collection exhibiting more than 25% imbalance, fuzzification, application of 512 fuzzy If-Then rules, fuzzy inference using the Mamdani method, and defuzzification via the centroid method. MATLAB simulation results demonstrate a reduction in load imbalance from 28.19% to 3.6%, in accordance with SPLN 017:2014 standards. Additionally, neutral current decreased significantly from 97.94 A to 12.51 A, improving transformer operational efficiency. These findings confirm that fuzzy logic provides an effective, adaptable, and computationally efficient solution for balancing phase loads in distribution transformers, with strong potential for real-time intelligent grid applications.