The use of affine arithmetic for thermal state estimation of substation distribution transformers

Thermal protection of mineral‐oil‐filled substation distribution transformers is of critical importance in power systems. The failure of such a transformer is a matter of significant concern for electrical utilities, not only for the consequent severe economic losses, but also because the utility response to a customer during outage condition is one of the major factors in determining the overall customer attitude towards the utility. Therefore, it is essential to estimate the thermal state of transformers during load cycling and, in presence of overload conditions, to evaluate the need to reduce the load current or to install another transformer bay. A method of solving the transformer's thermal model, considering explicitly the source of uncertainty affecting its parameters, is required. In this paper, such an activity is developed by an interval‐based approach, which provides the calculation of the inner and outer solution in the hot‐spot temperature or top‐oil temperature estimation process, keeping track of correlation between uncertain quantities.