Project Summary
Although the project has relevance to the challenges on whole system integration and heat, its scope is most clearly associated with the challenge for data and digitalisation:
- Automated pressure management software, and the use of near real time data and machine-learning techniques, will contribute to better coordination, planning and network optimisation
- Increased injection of biomethane and hydrogen into the network will enable progress towards net zero and enable strategic outcomes from other challenges e.g., decarbonisation of heat
The project directly addresses as its primary focus points 7 and 9 in the challenge scope definition:
- Point 7: this project will use novel sensor technology to improve visibility of the condition of network infrastructure and make data-driven decisions about that infrastructure.
- Point 9: this project will use data, combined with machine-learning and AI techniques, to improve the forecasting abilities of both demand on the network, and required maintenance and interventions.
The principal innovation underscoring the project is use of data-driven techniques based on AI and machine-learning to address each Opportunity Area (OA). These would constitute novel methods which, combined with modular dashboards that integrate the solutions with data analytics, will help SGN continue its positive journey to delivering a digitalised network.
During Discovery, the project evolved by researching network user needs, identifying underlying motivations and enabling deeper understanding of the opportunities. This allowed the refinement of problem statements, outline AI solutions, and impact and complexity assessments. During Alpha, progress will continue by refining benefits cases and undertaking "bench-testing" of solutions ahead of Beta field trials.
The solutions address the challenges in several ways:
- ML/AI models optimise the pressure in the Low Pressure (LP) network to reduce leakage.
- In the Medium Pressure (MP) networks, the models optimise the injection of biomethane enabling progress towards net zero.
- Data from the Utonomy system is used for network anomaly detection leading to improved maintenance. This data is combined with other sources of data to predict the distribution of reported escapes.
- A dashboard is used to display relevant data and KPIs. This enables operators to make faster and more effective interventions.
Utonomy will be the main project partner for Alpha. The Utonomy engineering team has capabilities in electronics design for hazardous areas, data science and machine learning, industrial IoT and digital communications technologies, cyber security, and cloud-hosted software applications. Utonomy has already collaborated successfully with SGN and Wales & West Utilities on developing, trialling and proving 'remote pressure control & management' technology. Utonomy has carried out initial field trials with SGN of its Intelligent Gas Grid Control software concept, developed via an Innovate UK funded project completed in March 2022.
Utonomy will use Faculty Science Limited as lead subcontractor; who are uniquely placed to deliver state-of-the-art AI solutions from teams formed from over 200 professionals comprising both technical and commercial experts. In delivering AI solutions, in-house developed AI Engines allow specialised techniques to be applied to customer problems and to optimise performance
The solutions will be primarily used by two sets of users; the network maintenance team, responsible for managing pressures and undertaking maintenance, and the network planning team, responsible for overall network planning, performance and analysis.
The maintenance team will use the solution to adjust governor pressures remotely, and automatically, to minimise leakage and optimise biomethane feed-in. They will also use the data to diagnose and resolve network or asset faults more quickly. The planning team will use the solution to track KPIs such as leakage reduction or biomethane injection. They will also take decisions based on data and analysis provided by the solutions.
Innovation Justification
Some gas distribution networks today are still manually operated with governors adjusted seasonally. This means that they are often set conservatively high, creating more pressure in the network than is required which in turn causing avoidable leakage. Manual adjustment is also a logistical problem, tying up skilled technicians, and it can also be difficult to judge when to make a setting change.
Another problem with networks is they were designed in the 1970's for North Sea gas, which had a small number of entry points and flowed in one direction. Today, there are multiple entry points as further biomethane plants are connected to lower pressure tiers in the network, which will grow even more with the arrival of green hydrogen. Therefore, intelligent and smarter network control is necessary to effectively manage the new operating challenges facing networks.
Networks are continually under pressure to increase efficiencies; improved access to data and analysis will enable faults on the network to be diagnosed and resolved more efficiently.
Better analysis of the causes of reported gas escapes should also enable the frequency to be reduced and labour to be more efficiently deployed.
AI/ML models have not previously been used for pressure management, making this project highly innovative. Gas Technology Institute (GTI) of the US has done a global market search on behalf of US utilities and has not discovered anything else comparable. SGN tendered the pressure management NIA in 2018 which also did not reveal any alternative solutions.
This project is important to the energy sector, which needs to:
- Maximise injection of biomethane to support network goal to reach net zero.
- Reduce methane emissions: There is greater emphasis following COP26 and the methane pledge.
- Reduce costs of distribution which will contribute to lower energy bills.
The problem of optimising the pressure in the network is expected to be solved by developing two ML models. The first forecasts demand profile over the following 24 hours. The second predicts network extremities' pressure over the same period, given a level of demand and the governor settings. Using both models together it's possible to determine the optimum settings of the governors feeding the network. Biomethane application is similar; natural gas feeding into the MP network needs to be continuously adjusted to ensure Security of Supply is guaranteed and biomethane feed-in is prioritised. The escape prediction application is based on two models; an escape count model and a causal escapes model.
Knowledge missing from previous projects is as follows:
- Risk assessment of pressure management. Low point monitoring and alarm facility.
- Testing pressure management on wider variety of networks.
- Testing anomaly detection on live networks.
- Using intelligent control on existing biomethane projects.
Not carrying out the project would have the following impacts on the GB gas network:
- The network would run at higher pressures leading to higher methane emissions.
- Biomethane injection would continue to be restricted leading to increased flaring.
- A number of biomethane plants would not be built because they couldn't get connection.
- The network would have higher operational costs which would be passed on to consumers.
In the long-term, the value of the solutions is to enable the network to transition to net zero. In the short-term it is increased efficiency and a reduction in methane emissions.
If the project were funded under BAU or within the price control, it would take significantly longer, and the solutions would arrive too late to enable effective transition to net zero.
Benefits
The most significant expected benefits are:
- Reduction in methane emissions.
- Increase feed-in capacity of biomethane and hydrogen.
- Predict and reduce gas escapes.
- Predict the occurrence of faults on key network assets.
- Summarise the health and ongoing performance of the network.
- Enabling more effective KPI/dashboard management for networks.
- Improve Repex efficiency (e.g., more insertion and less open cut).
- Meet increased demand through pressure management rather than capex.
Net benefits to consumers, as follows:
- Reduced operating costs will be passed on to consumers in the next price review (GD3).
- Customer surveys indicate a high proportion of gas customers would welcome
- reductions in methane emissions and effective progression towards net zero.
- Improved customer service - if network problems are fixed more quickly.
Discovery Phase considered eight Opportunity Areas (OAs). The benefits and impacts of each of the OAs was analysed. An impact framework was developed which identified four key benefits sources for consumers, society, government, and the environment.
- Cost savings: direct savings (for example, reduced gas purchases) and improved performance against outcomes framework targets.
- Carbon reduction: minimisation of emissions, gas escapes and enabling increased injection of renewable gas sources.
- Customer experience: via reductions to unplanned interruptions, and proactive maintenance scheduling.
- Operational efficiencies: via augmentation of periodic network activities with AI solutions.
A prioritisation approach was developed which considered both potential impact and complexity, which assessed factors including feasibility, development effort, data availability, production suitability and the ease of translating decision-making insights into operational outcomes. These complexity metrics will ultimately define costs, in addition to those directly related to this project, of deploying any resulting solutions.
In Discovery, potential benefits have been assessed through a combination of stakeholder input and high-level statistics. Given the wide potential set of use cases, the Discovery Phase has therefore focused on assessing all initial OAs that maximise the value of AI with remote pressure control technology.
The project will create associated benefits as follows:
End consumer
- Lower gas bills through reduced distribution costs.
- Faster resolution of network problems.
- Less disruption through fewer gas escapes.
Economic benefits for supply chain, broader industry, and the UK economy
- Create high skilled jobs in Utonomy (SME).
- Develop UK AI capability with application to energy networks.
- There is significant interest in this technology in US and Europe which will lead to increased exports.
- Will increase demand for Utonomy hardware (pressure management) benefitting supply chain.
- Increased number of biomethane plants will create jobs in rural communities.
- Will increase revenue of biomethane plants stimulating further investment.
- Impact on government priorities
Digitalisation of the grid is necessary if it is to be repurposed in the future to carry Hydrogen.
Environmental impacts
- Reduction in methane emissions.
- Enables use of renewable gas.
Expected regional or wider energy supply resilience benefits
- Reduces imports of gas by using more biomethane.
Impacts on consumers of the whole energy system
- Reduces costs of distribution through increased efficiency and reduced unaccounted for gas.
- Using locally produce gas-biomethane reduces reliance on gas imports.
Impacts and benefits
The benefits assessment developed within the Alpha Phase sought to qualify, and where possible, quantify the scope for benefits that could result from the implementation of solutions that leverage AI combined with remote pressure control technology. The overarching objective of this activity has been to provide evidence to justify further investment in the development and implementation of these combined solutions, within the prospective Beta Phase.
At a high level, the steps to achieve this at Alpha have been:
- Identify the potential channels by which benefits can be delivered by AI combined with remote pressure control technology: these include both ‘internal’ benefits to SGN including cost savings and operational efficiency, but also’ external’ benefits relating to customers and the environment.
- Develop a methodology to assess the relative strength of these channels for priority OAs, including a calculation framework to estimate the extent of such benefits.
- Issue a data request for SGN to provide relevant statistics to inform the assessment; this has included a sample of pressure and governor information for a sample of 5 profiled and 5 non-profiled, seasonally adjusted networks.
- Quantify the scope for benefits across these benefits channels for OA1 and OA3, employing a wide set of data sources and assumptions, including SGN operational data, characteristics of the Utonomy technological solution, and information from the public domain (e.g. wider economic variables).
The summary of the outputs from the benefits assessment was as follows:
Cost savings: Implementation of the calculation framework is estimated to deliver leakage reductions of c.6-11% across the sample of seasonally-adjusted, non-profiled networks provided.
Reduction to LP network leakage: across the 5 non-profiled networks, the weighted average potential reduction in leakage (that is the average percentage reduction in leakage, weighted by the absolute level of leakage present in these networks) is estimated as 7.03%.
It is estimated that rolling out the Utonomy solution across the full Southern Network could deliver 12.93 GWh of leakage savings (c.16,000 tCO2e) per year. This would represent a 4.8% saving relative to the total Southern Leakage reported in 2021/22, and 3.0% of Southern Shrinkage (i.e. when including MP leakage, services, AGIs, interference, own use gas and theft).
Scaling the leakage reduction benefits across all of SGN’s networks would equate to c.21,000 tCO2e (17GWh) per year, with a value saved of c£1.6m, and scaling further across GB networks would equate to c.75,000 tCO2e (61GWh) per year, with a value saved of c£5.7m.
