go to top scroll for more


Projects: Projects for Investigator
Reference Number NIA_NGET0197
Title Development of fittings analysis model
Status Completed
Energy Categories Other Power and Storage Technologies(Electricity transmission and distribution) 100%;
Research Types Applied Research and Development 100%
Science and Technology Fields ENGINEERING AND TECHNOLOGY (Electrical and Electronic Engineering) 100%
UKERC Cross Cutting Characterisation Not Cross-cutting 100%
Principal Investigator Project Contact
No email address given
National Grid Electricity Transmission
Award Type Network Innovation Allowance
Funding Source Ofgem
Start Date 01 January 2017
End Date 01 January 2018
Duration 12 months
Total Grant Value £300,000
Industrial Sectors Power
Region London
Programme Network Innovation Allowance
Investigators Principal Investigator Project Contact , National Grid Electricity Transmission (100.000%)
Web Site http://www.smarternetworks.org/project/NIA_NGET0197
Objectives The objectives of the project is to develop data driven models and draw new insights into the overhead line fittings system, in order to deliver values to both National Grid asset management business and the UK electricity consumers. The project will develop methodologies of enhanced level of utilisation of the condition information recorded by National Grid at the span level and component level, which is a valuable dataset of both research interest and long term planning. This project will also examine the impact of the new understandings to the asset management practices in terms of cost, risk, performance, regulation, as well as process complexities. As a research project, this project will take an exploratory approach using modelling and data science. The following criteria are expected to be met at the end of the project: Increased utilisation of the more granulised and up-to-date information from the innovative condition assessment methods for overhead line fittings. Increased/quantitative understandings of the overhead line fittings system, including end of life, condition/AHI classification, and future degradation behaviours. Increased transparency and consistency in the asset health assignment process and investment strategies. Quantitative models developed to understand the above and also can be used for National Grid’s asset management practices. Optimised solutions based on studies on the interactions/optioneerings among different investment strategies, by taking into consideration of key investment drivers, and engaging with key stakeholders.
Abstract Since year 2013, National Grid has adopted advanced condition assessment techniques and enhanced database capacity in order to condition monitor the fittings located on overhead lines (OHL), on a span-by-span basis. This enables more granulised asset condition assessment that then drives replacement schemes on a case-by-case and span-by-span basis, instead of the previous, more traditional circuit-by-circuit basis. While savings can be derived in terms of targeting on the worst condition spans and components, the use of an innovative fittings strategy introduces a number of challenges to the practice of asset management (AM). the understanding of future deterioration for the granulised condition information is still being developed, and this impacts on long term planning. complexities introduced into the process and performance measures, particularly in the interaction between other, more traditional OHL asset management strategies such as OHL conductor deterioration and replacements. new data-science technology and methodologies that feed drive the new strategy need to be trialled in the form of impact analysis within a controlled working environment prior to implementation. To understand the condition of assets and its future deterioration is a complicate issue. On the one hand, electricity transmission assets are subjecting to a large number of condition indicators that need to be categorised into a small (manageable) number of asset health index scores. On the other hand, those condition indicators change with time (heterogeneously) due to natural deterioration and asset management activities such as maintenance and repair. Data driven asset management models have been one of the most important and fruitful ways to answer those questions. The research proposed in the project will provide a framework of combining state-of-art data science, innovative condition assessment information, subject matter knowledge, engineering understandings, and expertise opinions to inform the risk management and investment management practices. This is particularly meaningful for the transmission and distribution network owners as most of the investment projects take long time to plan and deliver while collecting condition information can be expensive and time consuming. For the RIIO regulatory framework applied to the UK transmission and distribution networks, the asset health index (AHI) is an important factor that regulates the non-load related Capex. Reducing the uncertainties and inconsistencies around the AHI, and enhancing the ability of forecast deterioration are crucial in terms of achieving efficiencies while delivering outputs to the consumers. The project will use a methodology of a data-driven, bottom-up approach, guided by engineering knowledge and expertise from subject matter experts (SME). It will be benefit from a variety of innovative data analysis techniques, including statistical methods, mathematical modelling and machine learning. Additionally, this work builds on the experience and knowledge from the previous research project (OHL Condition Assessment (10359 - NIA_NGET0140)). The methods are detailed in the following bullet points. Creation of relational database To support the requirements of the data analysis and modelling aspects of this work, a relational database will be created to capture all data sources used by National Grid for fittings asset condition. This will include additional data cleansing and validation, using both manual process and state-of-art text matching techniques. End of life behaviours modelling To understand the end of life behaviour of fittings, existing condition information can be used to determine what asset condition (types/number of defects, visual assessment etc. ) would be considered indicative of end of life, and then determine the age profile of assets which are found to be in this condition. Depending on the type, amount, and distribution of the data provided, the analysis may take the form of correlation analysis and visualisations, dimensionality reduction, and unsupervised or semi-supervised clustering, and then validated by the SMEs Data analysis and quantification To review, model and validate the asset health index (AHI) process, it is proposed to analyse the relative contribution of each element to the overall AHI score, and (if necessary) quantify any top-down influences which may arise following engineering and management review. Any manual interventions in the process will be quantified by using machine learning techniques to train a classification model. The goal of the analysis is to validate the AHI assignment process in a data-driven fashion, working backwards from the end of life criteria identified in bullet point b) will also be used as inputs and validations. The relationship between span level AHI and circuit level AHI will be also reviewed, to understand how much variation can be expected within a single circuit. Review and update the deterioration model To compare the AHI process and the deterioration model, the model created in the previous phase of work will be used to simulate an asset portfolio. This enables a like-for-like comparison and solutions to resolve any differences will be explored. Additionally, this work will also explore detailed degradation modelling strategies and approaches published by other infrastructure asset owners such as Network Rail, London Underground, UK Power Networks and United Utilities. Those modelling approaches (such as P-f type models) or data techniques (such as Bayesian or machine learning models) will be reviewed and tested to understand the applicability to the transmission assets. Case studies and implementation plan This project is also to study the interactions in portfolio costs, risk and process complexity between different asset replacement strategies for OHL assets. To ensure a comprehensive comparison of the different fittings investment strategies, as well as their interactions with the conductor’s investment strategies, this project will collect various comparison metrics via workshops with relevant National Grid’s teams, e.g. planning engineers, investment management team, field engineers, etc. These workshops will help collect important information such as: 1) span-level cost data, for replacement parts as well as labour; 2)planning systems, processes and their complexity; 3) system access and land access, on a span-by-span level; 4) additional business and asset risks if not covered in existing risk framework. The project will then set up a model which will take above into account, as well as fittings AHI, conductor AHI, and any other factors identified by National Grid stakeholders, and produce scenarios and guidelines for choosing the most economical and appropriate asset investment strategy. Using the model developed above, a cost, and risk and benefits study can be performed regarding different investment strategies and assess the short term and long term business impacts.Note : Project Documents may be available via the ENA Smarter Networks Portal using the Website link above
Publications (none)
Final Report (none)
Added to Database 09/08/18