Fuzzy Logic Based Energy Efficient Transformer Cooling Control

Abstract

Cooling is the most important way that can be used to extend the life of overloaded transformers, especially by reducing the aging rate due to temperature, and to ensure that the transformer can be loaded above its rated power. In addition to increasing the load capacity of the transformer, good cooling is also required to be energy efficient. In this study, an energy efficient transformer cooling control method is presented. The proposed method was developed by revising the differential equations method, which is one of the transformer temperature calculation methods given in the IEC 60076-7:2018 standard. With this method, transformer hot-spot temperatures are calculated. For energy efficient cooling control, transformer winding currents and top-oil temperature are first measured and transformer windings’ loading rates and hot-spot temperatures are calculated. Afterwards, the highest loading rate of the transformer windings and the highest hot-spot temperature value are given to the input of a fuzzy logic controller, and the on-off signals of the cooling fans used to cool the transformer are generated. The fuzzy logic controller performs the fuzzification process with three membership functions for the loading rate and hot-spot temperature, and performs the defuzzification using the Takagi-Sugeno model. The method has been investigated by experiments performed in the laboratory environment using short-circuit experiments on an ONAF cooled transformer with a power of 50 kVA. Test results showed that as the hot-spot temperature and loading rate of the transformer increases, more cooling fans are activated, thus the loading capacity of the transformer is significantly increased energy efficiently.