Processing and Visualisation Methodologies for Lightning and Outage Related Big Data in an Effort to Improve Maintenance and Operations within an Electrical Power Utility

Abstract

This paper presents a methodology that aims to reconcile the presence of lightning within the vicinity of a power line and the unexpected circuit breaker operation or insulation failure of electrical power plant assets, based on geospatial and temporal correlation. In this process, it can be inferred that repeated geospatial correlations indicate poorly performing lightning protection. This information set forms a basis that can facilitate prioritisation of maintenance routines and money spent on network improvement for specific locations of the network. The ultimate output of the methodology is a map which indicates poorly performing segments of a power line in relation to lightning activity exposure. The method reduces unmanageably large (~108) multi-dimensional datasets into a user-friendly 3D geospatial representation that can be readily interpreted and actioned. The method produces convincing correlations that can form the basis for preliminary investigation and adapted to many potential uses.