Novel Hotspot Temperature Prediction of Oil-Immersed Distribution Transformers: An Experimental Case Study

Abstract

Distribution transformers (DTs) are deemed as fundamental and high-priced equipment of power grids and their failure influences grid reliability. Continuity of DTs operation highly depends on its insulation conditions and following that to the hotspot temperature (HST). In this paper, a novel formula is proposed for HST prediction of DTs by considering the effective electrical and mechanical parameters on heat dissipation capacity. As well as, a complete and accurate 3D modeling based on computational fluid dynamic (CFD) is presented to validate the proposed novel formula for HST prediction. In this modeling, the conservator, which plays a key role in the HST value, is accurately modeled. Optical fiber sensors (OFSs) are utilized in the studied 500 kVA DT to verify the accuracy of the proposed HST prediction formula during the experimental temperature rise test. Experimental results show that the proposed formula is highly accurate and has an acceptable correlation with the empirical values. The root mean square error (RMSE), mean error (ME) and average error percentage (AEP) of the proposed formula are 0.3°C, 0.3°C and 0.37%, respectively, in HST transient values and 0.3°C, 0.3°C and 0.29% in HST steady state values, which substantiate the precision and proficiency of the proposed HST prediction formula rather than IEC and IEEE equations. Finally, a thermal camera is employed to verify the results of CFD-based 3D modeling in top-oil temperature (TOT), bottom-oil temperature (BOT) and conservator oil temperature (COT) during the experimental tests. According to the measurement results, temperatures of CFD-based 3D simulation and thermal camera in the aforementioned three points are in good agreement with each other and AEP is less than 1.4%, which indicates the accuracy and efficiency of the proposed 3D modeling.

Topics & Concepts