Use of survival analysis and simulation to improve maintenance planning of high voltage instrument transformers in the Dutch transmission system

TL;DR

This paper applies survival analysis and simulation techniques to model the lifetime of high voltage instrument transformers, providing data-driven insights for maintenance planning across different voltage levels in the Dutch transmission system.

Contribution

It introduces a novel approach combining survival analysis with simulation to optimize maintenance strategies for high voltage transformers using real failure data.

Findings

Failure probability increases with in-service age

Voltage level influences remaining lifetime

Simulated scenarios optimize maintenance costs and reduce unavailability

Abstract

This paper describes the use of survival analysis and simulation to model the lifetime of high voltage instrument transformers in the Dutch transmission sys-tem. To represent asset aging, the non-parametric Kaplan-Meier method is used to enable the fitting of Weibull distribution. Such an approach is implemented on three different voltage levels, namely 110kV, 150kV, and 220/380kV. Real failure and inspection data is used to achieve a realistic failure model of the instrument trans-formers. Failure and maintenance data occurring between 1989 and 2021 have been used for this study. In spite of missing and low-quality data, a rich failure database could still be prepared. This study also offers insights into factors (i.e., voltage level, in-service age) influencing the remaining life from both graphical survival function and parametric Weibull distribution analysis. Based on the derived statistics, future possible maintenance planning scenarios are simulated under a complex system modelling framework in a digital twin enabled platform. Eventually, the scenarios are evaluated in terms of replacement costs (CAPEX), inspection hours, and unavailability hours.