An improved risk management model for a fleet power transformer through reliability asset condition and probability of failure using weibull analysis

However, Recent CIGRE reliability studies indicate that failures are random in nature and there is no correlation of the probability of failure with age. In response to this, a failure probability model was developed using the two-parameter Weibull distribution. The scale parameter (), which represents the characteristic life, and the shape parameter (β) were derived from historical failure data of CELSIA's transformer fleet through linear regression using the LSM (Least Squares Method). This analysis incorporated transformers that failed after being repaired. Notably, the "time" variable in the traditional

Weibull distribution was replaced by a weighted performance linear function, which was adjusted based on periodic test data, including DGA (Dissolved Gas Analysis), physical and chemical tests, furan analysis and TFC (Through Fault Current). The resulting failure probability is quantified on a scale from 0 to 100, varying across transformer groups, and is determined based on a reliability index (derived from the scale and shape parameters) and a health index. To assess the consequences of transformer failures, five factors were considered: environmental impact, operational flexibility, safety and health risks to personnel, maintenance costs, and the potential impact on the company's public image. The final output of this analysis is a risk map that facilitates the prioritization of transformers based on their relative importance, associated risk, and performance. The map categorizes transformers into risk levels: very high, high, medium, and low. This prioritization tool aids in the optimization of maintenance activities, enabling the establishment of a risk-based maintenance framework that enhances the reliability and availability of the power transformer fleet.