Abstract

The use of Health Index (HI) has helped in driving electrical utilities decision making for strategic investment planning including operation, and maintenance (O&M) programmes. This approach increases the ability of the business to implement robust investment decision objectives that makes physical assets safe, productive, efficient, and cost effective. Asset Management (AM) framework is defined by British Standards PAS55 and ISO55001/2 to ensure electrical network operators are delivering best quality of services by operating at high performance, low cost while managing unexpected risks [38, 39].
Data monitoring and recording have improved but still robust master dataset is either limited or not available due to a range of factors including huge costs for capturing live data, the lack of monitoring tools, limited or no data collection and, data uncertainty challenges. there are a small number of electrical utilities [31] around the globe who capture data using recent technologies that work in line with information best practice which serve large fleet of Power Transformers (PTs).
Specific tools such as Artificial Intelligence (AI) has helped to address problems associated with limited or missing data uncertainty issues, data capturing process, and therefore impact the accuracy of a HI calculation. Recently, innovative systems became an alternative approach in structuring big data to support condition assessment and condition monitoring tools which are reliant on various data sources.
This paper discusses HI scoring for transformer condition assessment using conventional methods that can add value to the AM practice. This includes defining HI model requirements. Power transformer health index data interpretation analysis will be considered using international standard: Institute of Electrical Electronics & Engineers (IEEE) C57.104 transactions on industrial informatics. Preliminary analysis for data management using Python Programming Language (PPL) is considered. The authors hope to provide an up-to-date review of the current literature to allow academia and industry to question and challenge a preconceived perception that large datasets are difficult to obtain and review to support the development of new decision-making investment activities.

Keywords: Power Transformer, Asset Management, Health Index, Artificial Intelligent, Electrical Utility, Decision Making, Strategic Investment Planning, Python Programming Language.