Transformer Remnant Life Estimation and Asset Management model based on Insulation Stress Assessment

Abstract

The performance and useful life expectancy of a power transformer generally depends on the condition of oil-impregnated paper insulation system that has a finite life span. End of technical life of a power transformer is reached when some of the insulation diagnostic characteristics drop to a level where further operation is unacceptable. Thus, recurrent evaluation of the transformer insulation system diagnostics enhances better estimation of transformer residual life. However, being a function of voluminous conflicting attributes, the design feature, changing operating conditions, and different maintenance strategies, a flawless affirmation of the criteria governing the transformer remnant life estimation is a very complex issue. Power transformer operational condition that can be mirrored by criticality of ageing and fault stress levels has substantial impact on its residual life estimation. This paper explores the possibility of developing a remnant life estimation and asset management decision model based on diagnostics of transformer insulation characteristics. The proposed model is developed upon the integration of the fuzzy logic diagnostic tool and the fuzzy logic remnant life mapping model. The diagnostic details are signified by fuzzy rules for an accurate assessment to reach a defined outcome in terms of fault stress level and energy of faulting, ageing stresses and transformer condition. A multicriteria analysis was used in developing the model whereby the outcome of the model is achieved by including the combined effect of individual sub outcomes using fuzzification of all the employed attributes. The developed remnant life model is validated using data collected from several in-service mineral oil immersed transformers. Results show that the proposed model provides an approximate but practical means of remnant life estimation and asset management decision.