Sizing Step-up Transformers for Solar Generation based on Load Profiles Predictability

Abstract

This paper shows the conclusions of practicing a more accurate sizing for step-up transformers based on real load profile when applied for solar generation, either thermal or photovoltaic. Due to inherent cyclic load for solar generation, with peaks and valleys along the day and months on a year, the transformer rated power can be better adjusted to an average value in such a way that loss of life is closer to unitary during a complete year period of evaluation; in this way, the transformer is not over-dimensioned and can provide the same reliable and robust service during enough amount of years compared to today's traditional approach. Additionally, the increasing trend in industry to incorporate transformer monitoring with digital data analysis, allows to obtain a real and life load profile, providing real time remaining life expectancy and valuable predictability that is proposed to be used as a base to minimize the auxiliaries power consumption, thus minimizing the thermal dynamic effects on the active part and helping to reduce life ageing as well.