Efficient electricity distribution in rural areas is strongly affected by seasonal agricultural energy demand, particularly in irrigation-intensive regions where electricity consumption increases substantially during summer periods. Conventional transformer operation strategies in such rural grids often fail to adapt to seasonal load variability, leading to unnecessary idle operation, increased technical losses, and reduced infrastructure efficiency. Existing approaches generally rely on static assumptions or simulated data, limiting their ability to represent real irrigation-driven seasonal load asymmetry. To address this issue, this study proposes a data-driven multi-objective seasonal transformer scheduling framework using a bio-inspired Artificial Immune System (AIS) algorithm. The model was developed using two years of empirical hourly SCADA load data and transformer operation records obtained from a real 380/154 kV TEİAŞ transmission substation in Central Anatolia, Türkiye. Hourly SCADA measurements were used for seasonal load characterization and objective-function evaluation, while transformer scheduling decisions were defined at the seasonal operational level. The proposed AIS-based scheduling strategy reduced annual technical energy losses by approximately 5.4 GWh, decreased operational costs by 10.81 million TL (≈360,000 USD), and lowered carbon emissions by about 2270 metric tons of CO2 compared with conventional static transformer operation. The study presents a proof-of-concept framework integrating empirical SCADA measurements with AIS-assisted seasonal transformer scheduling for practical utility-scale operational planning in irrigation-dominated rural electricity networks.
Loading....