Project Summary

The original CReDo project produced failure models from elicitation interviews with operatives familiar with the asset and then translating these into a mathematical model. Whilst this generated useful failure models the process was found not to be scalable due to the time and specialist expertise needed to conduct them.

To address this the project will develop a user-friendly tool (e.g. configurable questionnaire or interface) which asset specialists will interact with to generate probabilistic models of asset failure from climate risks, enhancing the digital twin's capabilities and scope for application. This tool will enable asset engineers to better quantify and understand climate risks and networks robustness, enabling more targeted investment to build resilience and robustness and protect supply for customers.

During the Discovery Phase, we will investigate the suitability of different extreme weather conditions (wind or heat) for this work and then focus on the selected condition in the Alpha Phase.

The project addresses Innovation Challenge 3: Improving Energy System Resilience and Robustness, scope 2.

The network innovation lies in the transformative capabilities for DNOs to understand the risks and modalities of asset failure, which currently exists only as tacit knowledge held by asset specialists. Providing quantitative failure risks enables resilience and robustness to be incorporated as a key measure in network operation and planning. DNOs will therefore be able to make more informed decisions for capital investments and asset planning, strengthening their infrastructure to ensure smooth adoption of multi-energy systems and a robust energy transition towards net zero.

This consortium is uniquely well placed to deliver this project:

1.The partners involved delivered the original CReDo, which included the first creation of Bayesian failure models from elicitation interview.

2.UK Research and Innovation - Science & Technology Facilities Council (STFC) through its Hartree Centre and DAFNI platform, will provide the project with crucial data science and software engineering expertise, and secure hosting.

3.UK Power Networks (UKPN) will support the project from a Distribution Network Operator perspective, supplying relevant data and engineering expertise.

4.Computational Modelling Cambridge Ltd (CMCL) will use their experience developing semantic knowledge graphs to represent the critical assets, the impact of asset failure and the cascade of failures throughout the system

5. Connected Places Catapult (CPC) lead the current phase development of CReDo and bring deep experience in identifying market failures and convening stakeholders to solve them.

Innovation Justification

Extreme weather events driven by climate change increase the risks of electricity supply disruption causing a cascade of failure across the infrastructure system. DNOs have extensive historical recorded data for faults and incidents that have occurred during extreme weather conditions, as well as specific asset performance information held with experienced engineers. The latter is not digitised and would play an important role in identifying risks and mitigations. The frequency and extremity of weather events is expected to increase, therefore there is a need for elicitation to bridge this gap and aid in pre-emptive mitigation.

CReDo pioneered a new approach to identifying the risks of cascading impacts across energy, water, and telecoms networks, which was demonstrated against flood-risk. This project would build on that foundation by adding in models of risk related to another extreme weather phenomenon.

The project is novel in two ways. Firstly -- building asset failure risk models for a new climate risk. Secondly, through its intent to build a tool to allow other risk models (for other assets and risks) to be constructed more easily.

This tool would generate asset-specific failure risk models through a combination of elicitation interviews with asset specialists and other relevant data where available.

The implementation of the resultant risk models in the CReDo digital twin would provide better predictability, robustness, and quantification of uncertainties under a variety of extreme weather conditions.

This consortium brings together the skills needed to develop the tool (CPC, UKRI, CMCL) with the network management and maintenance expertise of UK Power Networks.

UK Power Networks seeks SIF funding for CReDo+ because:

• The current phase of CReDo require development does not include asset failure models, with funding and development focused on the data sharing architecture to ensure scalable of the approach across all networks whilst maintaining security and data confidentiality.
• The vision to develop a connected digital twin across all infrastructure networks is ambitious and requires significant innovation, technical development, and coordination. Developing power network failure models is a crucial part of achieving this long-term vision.
• A complementary funding application is being made by Anglian Water to the Ofwat Catalyst fund to develop modes of failure under extreme weather conditions for the water sector. A combination of SIF and Catalyst funding would propel CReDo into the next phase and closer to the stage where energy networks are able to take this innovative tool in-house.

Project Benefits

We expect that a key part of the Discovery Phase is to identify and quantify the benefits further. As noted by Emma Howard Boyd, Chair of the Environment Agency, "global understanding of the costs avoided, and revenue generated by climate resilience is anecdotal and patchy" and this project will support identifying this value.

Financial -- future reductions in the cost of operating the network:

Repairs and replacement - Further investigation is needed to have an indicative quantification for the baseline network operating costs per annum allocated for repairs and replacement required due to extreme weather conditions. As a counterfactual, incident reports are measured for network interruptions due to extreme weather, through CReDo+ this number would be tracked to realise improvements through using the tool normalised to the number of events per annum.

Faults related to extreme weather in 2022:

• Solar Heat : ~280
• Wind : 3,700

Customer Interruptions (CI) and Customer Minutes Lost (CMLs) -- DNOs are incentivised to reduce CIs and CMLs since assets more resilient to extreme weather will the improved quality of supply for customers. Once CReDo+ is integrated into BAU to predict asset failures and inform resilience upgrade requirements, CI and CML can be tracked on an annual basis relative to historic baseline figures.

Risk management -- better warning systems in case of unprecedented events.

Better risk modelling will enable us to provide appropriate forewarning to our customers of risks to supply disruption (for example in the case of extreme storms such as Storm Arwen), helping them prepare better.

Social -- increased energy security and social value from energy transition.

By increasing resilience and robustness of the energy system, the Social Return on Investment (SROI) is improved. Businesses suffer disruptions to operations and hence economic impacts when damage occurs to their utility supplies. By minimising downtime through tailored resilience upgrade investments across utility asset portfolios, social and economic value will be generated. A more robust system will also support the transition to cleaner low carbon technologies and enable DNOs to help customers transition to net zero.