This project aims to employ long-term experiments to validate the new potential markers for silicone oils, which could denote the onsets of asset failure and thus allow corrective actions. The experimental results will be collated, cross-referenced and quantified to produce new diagnostic methods and maintenance approaches for silicone oil filled network assets. Based on these new diagnostics and approaches, this project would recommend the best practices to prolong asset lives and to carry out asset maintenance. This project would also investigate the feasibility of online analyser integration to mitigate the need for silicone oil sampling, which requires network outages. This project also aims to produce effective and reliable estimation methods for remaining asset life to inform asset replacement strategies.
Benefits
This project would seek to produce a set of procedures and metrics that would eventually allow accurate estimation of the remaining lives of silicone oil filled assets.
This project would also investigate the integration of online analysers to such assets so that real time condition monitoring could be implemented and the need for oil sampling, which is often difficult to arrange due to outage requirement, could be negated.
Learnings
Outcomes
A study of the presence of combinations and permutations of limited materials in the silicone oil, materials that normally exist within the asset such as polyethylene, silicone rubber etc. all subjected to thermal stress temperatures from 50-200°C have not introduced any significant changes on the concentration of the PD markers. It is believed that these markers are discharge selective markers.
Recommendations for further work
- Future research should focus on the comprehensive analysis and interpretation of spectral data obtained from Pyro-GC-MS. This will facilitate the identification of additional breakdown products and markers, thereby improving the accuracy of failure predictions and fault severity assessments.
- Building on the findings of this project, further work should be conducted to refine and validate on-line analysis techniques. This will enable real-time monitoring of oil breakdown and facilitate timely interventions to mitigate potential failures.
- The diagnostic methodologies developed in this project should be tested and applied to a wider range of assets that utilise silicone oil as a liquid insulator or contain silicone solid insulations. This will help to establish the versatility and effectiveness of the proposed techniques across different applications.
- Future studies should investigate the applicability of the principles established in this project to other insulating oils. This could lead to the development of alternative diagnostic methods that enhance the understanding of oil breakdown mechanisms across various types of insulating fluid.
Lessons Learnt
Lessons learned to date are as follows:
- The silicone oil specifications indicate a flash point above 300°C. However, it was observed that the oil supplied was gassing heavily at 250°C, which presents a potential fire hazard. This highlights the importance of verifying the actual performance of materials against their specifications to ensure safety and compliance.
- It has been demonstrated through experiments that temperature and humidity result in the jellification of the oil, strongly implying the occurrence of cross-linking. This finding underscores the necessity for further investigation into the relationship between the degree of cross-linking and temperature in future projects, in order to enhance the understanding of the effects of environmental conditions on material properties.
- Thermal studies uncovered 33 potential markers for PD. It was found that thermal changes ranging from 50°C to 200°C did not introduce any significant changes in the concentration of the PD markers. This indicates that the identified markers may remain stable under varying thermal conditions, which is crucial for ongoing monitoring and analysis.
Dissemination
In the interim, findings from the first section of the project were presented by the researchers to a small group of National Grid experts at NG House, and a synopsis of those findings has been submitted to CIGRE 2026.
