Case Study Overview
The increasing importance of digitalisation in our society is widely recognised, particularly in regards to enhancing the safety, reliability, and efficiency of facilities.
Digitalisation of the Nuclear Sector
The increasing importance of digitalisation in our society is widely recognised, particularly in regards to enhancing the safety, reliability, and efficiency of facilities. The nuclear industry is well aware of this and is actively exploring how digitalisation can provide value and improve competitiveness and safety. Although still in its early stages, the digitalisation of the nuclear industry has received strong support and commitment, with ongoing studies and developments, such as the implementation of digital twin technologies.
When applied to nuclear power plants (NPP), cost benefit analysis has shown that digital twins have the capacity to bring tangible benefits across the full lifecycle of NPP – from design stage to commissioning, operation and maintenance, and through decommissioning [REF attached]. For NNP owners, digital twins can improve operational efficiency, through the controlling and testing of the plant, the diagnosis of plant equipment such as pumps, motors, valves, etc., and the parameter tuning of automatic regulators.
Therefore, digital twins can be used to digitise nuclear engineering work in order to harmonise and optimise processes and methods throughout the plants’ lifetimes, and harness big data to boost the effectiveness of predictive maintenance.
EDF’s Steam Generator Digital Twin
A digital twin of a steam generator is currently under development at EDF Research and Development in France. The steam generator represents a critical component of the pressurised-water reactors (PWRs) with very expensive replacement cost. The new Steam Generator Digital Twin platform enables the operating history and behaviour of each steam generator in the fleet to be monitored.
Each Steam Generator Digital Twin is fed the data collected during operation and during the controls carried out during unit outages. The digital twin is used to run simulations to test scenarios and forecast the future condition of the assets in order to schedule an optimised maintenance. A similar project in the UK relates to steam generators for the UK Advanced Gas-cooled Reactor (AGR) technology.
Purpose of the Steam Generator Digital Twin
The purpose of the Steam Generator Digital Twin relates directly to EDF’s core business objectives: the safe production of secure and reliable energy sources. EDF is an energy producer, operating various technologies simultaneously (such as wind and solar power, as well as nuclear energy). In the UK, AGR type reactors are relatively old, as they were designed in the 1970s and built in the 1980s. Their conceptualised lifetime was approx. 30 years, yet some are still in service today, and all of them having been safely operated significantly beyond their original design.
The purpose of the Steam Generator Digital Twins, as designed by EDF, is both to support safe operations of plants and to justify safe operations beyond the lifetime. They are used to understand how the components behave through safety studies. These components are critical as they cannot be replaced and are essential for generating electricity. Essentially, the digital twins play a key part in technical assessment work which forms the basis of safety cases. In the UK, safety cases are used to justify to The Office for Nuclear Regulation (ONR) and other internal regulatory bodies that the component is safe to operate throughout their lifetime in order to generate revenue.
Overcoming challenges in the nuclear sector
For a digital twin to be useful, it has to address and alleviate particular issues.
Jefri Draup
Lead Research Engineer (Structural Integrity), EDF R&D UK
The nuclear sector has some unique challenges in comparison to others, such as aerospace. In many ways, this slows down the implementation of digital twin technologies. However, there are similarities to atherother industries, such as healthcare, which are equally stringent on regulation.
More specifically, the nuclear and aerospace sectors have many similarities in terms of degradation and ageing of components. This intersection enables collaboration on various scientific topics, which form parts of the digital twins capability.
In many ways, the steam generator digital twin application is very similar to digital twins in other industries. Still, there are a few factors that set it apart from others. When it comes to why EDF opted for a digital twin for this component, one has to consider nuclear reactor design and how they generate electricity.
The steam generator is used to extract heat. In AGRs, they were designed and built completely encased in concrete, which made their replacement impractical as they got older. This is one of the reasons where this digital twin application in the nuclear sector diverges from other industries, where specific components might be replaceable.
The other areas where it differs from other industries relate directly to that issue. It is important to consider how the materials and the component itself age throughout the lifetime. The digital twin helps make investigations into what is happening at a fundamental level inside the structure to determine the degradation mechanisms from a scientific perspective.
Moreover, understanding these mechanisms through the use of the digital twin helps justify decisions related to the operation of this component with enough certainty. Therefore, the digital twin is also used to conduct in-depth scientific studies and scenario studies. This process generates modelling based data, which is interpreted into information that supports judgement calls.
In contrast to other digital twins used in other industries, which might have more real-time impact directly into the real world, the steam generator digital twin lacks that functionality. That is because of the nature by which the nuclear sector operates – it is by committee, there is no one person making a decision. Still, decisions have impacts on the real world but not through direct interventions. Options are consulted with experts, and ultimately choices are justified and approved by regulators.
The route back for intervention in the real world is convoluted, which represents another area where the nuclear sector differs from others. If we look at an example from another industry, vibration sensors in an aero engine might be able to make a direct change to power output. Implementing a similar mechanism in a power plant would not be ideal as you don’t always want to generate an immediate response. The complexity by which a decision occurs is not really at a state where it can be fully automated, which is why nuclear digital twins are at a different stage compared to others.
For EDF in the UK, the lead time on the development of the physical models used in steam generator digital twins was approximately five years of research and development. This is because the methods to understand how the components might behave under all conditions were not yet available and some still remain unavailable. During this time, a research stream with the academic institutions and the entire supply chain had to be established. This teams helps understand phenomena, development of models, and most importantly is to get validation data, which are used as a trust metric for digital twins.
This process was followed by a period of another five years, in which the steam generator model itself was developed. The challenge there is robust quality control, verification and validation of the development. It took five years to get to a stage where it was mature enough to be usable.
How does it work?
Steam Generator Digital Twin (EDF)
In principle, The Steam Generator Digital Twin typically works with materials data collected over time, processed, and securely stored on record. This includes data related to the system parameters, pressures, and temperatures. The data is then transferred into the digital space. In many cases, that represents a manual process brought about by the need to comply with the rules and regulations of the sector.
Once the data is inputted, it is typically used together with the 3D representation of a component to answer ‘what if’ type scenarios to generate simulations. The original simulation will generate vast quantities of numerical data, due to the types of approaches used. The data is processed and information is extracted, typically in the form of temperature and fields within the component.
These are cross-compared to online measurements, either historical or live, depending on what is undergoing assessment. To reach a decision, the data and information are processed and undergo the regulatory procedure before making an intervention in the real system and component. This process takes place either online or during an outage.
What are the benefits?
The operational cost benefits represent a significant outcome from a business perspective. In the current landscape of the UK energy market, the revenue from running one of EDF’s Advanced Gas-cooled Reactors (AGR) per day is around two million pounds. Through the use of the Steam Generator Digital Twin, the outage period can be minimised, thereby leading to savings of lost revenue to that daily value.
The Steam Generator Digital Twin is also used for in-service performance assessments and loading studies, which supports the operation of the plant. Through in-depth analysis of maintenance, chemical and operational data, the lifespan of steam generators can be optimised by reducing the number of descaling operations that need to be carried out. That is how we were able to anticipate the development of clogging in the generators according to operating conditions, and take actions to limit it. This aspect is contributing directly to the lifetime extension of the plant and therefore revenue gain.
During the operational stage, the use of the Steam Generator Digital Twin boosts competitiveness and performance, enabling the use of techniques, such as uncertainty quantification, which improves risk or safety analysis. Utilising digital twins enables them to extract value from national infrastructure, such as high performance computing facilities, and helps with building connections with partners in the supply chain.
In terms of wider benefits, the development of the Steam Generator Digital Twin prompted the use of innovative tools, methods, and techniques. By developing the research and development supply chain, EDF made direct contributions to the education of society by creating new techniques which are published in the wider literature. Indeed, upskilling the community often pays dividends as high skilled workers often join the company having benefited from that investment.
Moreover, through these advancements, EDF has directly contributed to the training of individuals in the use of these techniques. The upskilling of individuals is part of EDF’s ethos to be technically excellent.
What’s next?
The learnings obtained in the development process have since been applied to other Steam Generator Digital Twin projects of different types and technology, facilitating the process. All the learnings have been directly applied to another plant, and it could potentially be applied to new plants being built in the UK as well.
Currently, EDF operates its own designs of power plants, but with an increase of recent government funding for development of new technologies, it is possible that EDF may operate reactors developed by external vendors. Indeed, EDF is committed to host an Advanced Modular Reactor (AMR) on one of its decommissioned AGR sites in Hartlepool. In that case, they are acutely aware that many potential vendors are applying digital twins technologies in the design of their systems, and as a company we would have to be capable of accepting and working with that digital twin.
Being able to access, share, and use tools which are interoperable with other digital twin systems outside of the EDF is crucial. Having standardised ways in which digital twins are developed across the world is important. Essentially,: having a worldwide web version of a digital twin is something that would be extremely beneficial for industry as a whole.
Aside from developing digital twins themselves, EDF are involved in support infrastructure for digital twins, particularly environments for data sharing cross-organisations. Initiatives such as the Nuclear Virtual Engineering Capability (NVEC) are collaborative attempts by the UK nuclear sector to develop a capability of that nature. Another example is the EPSRC funded SINDRI project, which is working on on data centric tools and methods used to interpret data generated from models. These we technologies we consider to be essential support infrastructure for digital twins.
Thank you to Jefri Draup, Lead Research Engineer (Structural Integrity) at EDF R&D UK, for making the time to share his insights and knowledge into the subject matter.