Failure analysis of transformer liquid — solid insulation system under selective environmental conditions using Weibull statistics method

The failure free operation of transformers is a main factor on economic and safety aspects as per the industrial consumer's point of view. Basically, the transformer is one of the prime equipment in power system network because of its unkind operating condition at different circumstances like higher operating temperature, overloading and continuous operation under uneven outdoor environmental conditions. Above said operating conditions lead to unpredicted failure in the transformer and this directly has an effect on reliability of the power system network. This work aims to investigate the failure rate of transformer liquid insulation (LI)–solid insulation (SI) system at various selective environmental conditions, by considering relative humidity of the air as a frequently variable environmental factor. For investigation, mineral oil (MO) and natural ester like sunflower oil (SFO) and rapeseed oil (RSO) are chosen as the LI system and Kraft paper is chosen as the SI system. The humidified samples are aged at 140°C up to 1000h at closed condition. Critical parameters of both LI and SI are periodically measured at an interval of 200h. Weibull distribution statistics method is used to predict the failure rate of the insulation system with respect to time and relative humidity of the air. From the analysis, it is inferred that the degradation rate of the transformer LI–SI increases with an increase in relative humidity of the air. It is also noted that the LI-MO degrades rapidly when compared with other natural esters. •Degradation rate of mineral oil and natural ester oils are analyzed by varying the relative humidity of the air.•Failure rate of oil samples are analyzed by using Weibull statistics method.•Sunflower oil and Rapeseed oil performs better under all conditions compared to Mineral oil and have higher life time.•This helps to improve the reliability of transformer and power distribution network in a greater extent.