NEI learns more about LongOps, the new international program for robotics in nuclear decommissioning and fusion.
Above: A British-Japanese research alliance aims to develop long-range robotics
LONGOPS IS A NEW UK-Japanese R&D program valued at £ 12 million to research and expand the capabilities of digital robotic technologies for use in nuclear decommissioning. The project will focus in particular on the digital tools used to control long-range robotics in long-term operation – hence also known as “LongOps”.
The key theme is the development of digital twins – a range of technologies including virtual reality (VR) and simulations of robotics that connect the virtual and physical worlds. These can enable strategic planning, training, remote operation, storage and analysis of data, including the prediction of maintenance events and potential operational problems, during the decommissioning of nuclear power plants.
The aim is to create skills and knowledge that will enable end-users in the nuclear space to become intelligent customers of products and services in this rapidly evolving area. The project is also expected to provide direct benefits such as employment opportunities, advances in ‘fusion-related’ technologies and improving robotics skills in the UK and Japan.
RACE – the Center for Remote Applications in Challenging Environments of the UK Atomic Energy Agency based in Culham, Oxfordshire – is leading the LongOps project on behalf of a collaboration funded equally by the UK Nuclear Decommissioning Authority (NDA), Japan’s Tepco and UK Research & Innovation (UKRI).
Safer, faster and cheaper decommissioning
The LongOps program deployment comes at a crucial time in the world of nuclear decommissioning. In Japan, the Fukushima Daiichi power plant is stable, including the four units badly affected by the earthquake in March 2011, and Tepco has presented an action plan for its decommissioning.
In the UK, Sellafield Ltd in Cumbria is working with industry to consider how best to use robotics and artificial intelligence to make the decommissioning more effective and efficient. This includes the Windscale Pile 1 reactor, which was damaged by fire in 1957. Challenges in decommissioning.
LongOps technologies will open more avenues for safer, faster, and cheaper decommissioning. It will also promote a cost effective reactor design for fusion and fission.
Top-class challenges
Old nuclear and fusion facilities are complex, large-scale projects whose safe decommissioning is time-consuming.
Windscale Pile 1, Fukushima Daiichi, and JET are some of the most popular use cases for emerging decommissioning technologies. However, it’s hard to imagine that a standard approach to decommissioning all three would be suitable – a graphite-moderated reactor from the 1940s, a BWR from the 1970s, and an experimental fusion reactor.
Although they pose different dangers, the decommissioning strategies overlap. In all three cases, access is through small ports with sleek, long-range robotics, and while the interior of the JET device is “as built,” all three environments present significant challenges that will evolve over time.
Tepco (via Mitsubishi Heavy Industries) recently acquired one from Veolia Nuclear Solutions UK Ltd. developed articulated boom. The boom, which will reach over 20m at full reach, will be used to inspect fuel residues in the primary containment containers at Fukushima Daiichi.
In Sellafield, Windscale Pile 1 requires robotics to safely and efficiently reach and extract the fuel assemblies housed in the 7.5 m deep reactor core.
LongOps comes into play in these long-term operations in high-radiation environments that are likely to require long-range devices. All three locations will benefit from advanced digital tools throughout the process in planning, training and possibly automating remote operations to reduce operator doses and lower costs.
RACE has experience from decades of use of the unique JET Remote Handling System. This system with two 12 m booms and haptic manipulators was used for maintenance and upgrades of JET, including a complete replacement of the inner wall. This means that it was used to disassemble and rebuild the device from the inside.
Next generation digital tools
At the heart of LongOps is a portfolio of research and development contract opportunities aimed at industry and academia. The contracts are scheduled to start in September 2021 and explore the potential for a new generation of digital tools to support nuclear operations.
These tools, collectively known as digital twin or digital mock-up, make operations easier for people by providing additional information and relieving some of the cognitive load. Even outside the control room, digital twins can provide useful information for task planning, operator training, documentation, and decision-making.
At RACE, workshops are held with employees from JET, Tepco and Sellafield to ensure that research and development are aligned with user requirements.
Physics simulations
One research interest is the inclusion of physical simulations, including the control of flexible long-range booms and haptic manipulators.
Real joint positions are difficult to measure or predict accurately, and disruption can lead to booms
distracting, making precise, repeatable movement difficult to achieve. Control algorithms that can take this into account are a key area of LongOps research, with particular relevance to the tight geometric constraints that occur in nuclear environments. RACE’s TARM long reach boom will be available for testing advanced algorithms.
automation
Automation is another area that can improve operations. Demanding tasks such as cutting, sorting, separating and collecting waste within tight operational constraints could be carried out faster and with fewer errors with the help of autonomous planning support.
Routine inspection tasks could be automated through machine learning, including the ability to detect anomalies and changes in the environment, with this information stored in the digital twin. Relevant information
can then be made available to operators or used in automated planning and control systems. LongOps will also explore how recent advances in AI and machine learning can be applied to robot control and perception. However, machine learning relies on the availability of relevant and sufficient data sets that do not currently exist. LongOps intends to start building these records.
Feel
At JET, remote dispatch operators rely on haptic feedback to control manipulators and tools, allowing them to experience the same forces and torques on the local device that are exerted on the remote device. This enables the operator to “feel” objects that are being gripped. However, virtual simulations of haptics that could be used for exercise cannot currently accurately recreate the feel of manipulators interacting with the environment. Currently, operator training for haptic devices requires the use of real devices.
LongOps will investigate the development of high frequency haptic simulators with high fidelity.
Modular software
All of this research is bundled into a modular software package called the prototype of the next generation digital mock-up.
Digital tools are often designed for stand-alone use or with a limited selection of interface modules. LongOps will use a modular, interoperable software architecture to facilitate reconfiguration.
This is what makes LongOps such an ambitious and exciting program – in addition to developing new tools to improve operations, it requires these tools to work together in a single, modular system.
Different end users may only need some software features or prioritize only some features. By managing a software suite and based on the requirements of real tasks, LongOps aims to create standardized, modular software that meets the various requirements of nuclear decommissioning projects.
Manipulators
Digital twins connect the virtual world with the physical world. In 2021 LongOps awarded contracts for the procurement of two teleoperated manipulators.
For the next two years, LongOps will conduct performance reviews, benchmarking tests and comparison activities within a test environment. The research program will seek to improve their use through the use of digital twin technologies and build on preparatory work on glove box automation led by RACE as part of the UKRI-funded academic network Robotics and AI in Nuclear Research (RAIN).
Meeting end-user training requirements
The nuclear sector needs suitably qualified and experienced personnel for successful remote operation. Currently, the training requires operators to use real hardware, which is expensive and time consuming. Digital environments that map more of the operator experience offer the potential to accelerate training, increase the number of fully qualified operators, and maintain and test competency over the years. The diverse situations in real use can also be better trained.
Change of nuclear decommissioning
The next generation of digital tools provide opportunities to develop decommissioning strategies, train and perform operations in high impact environments that are constantly changing.
RACE, UKRI, TEPCO and the NDA see digital tools and remote-controlled machines as an important path to safer, faster and more cost-effective decommissioning and a cost-effective future reactor design, be it in fusion or in nuclear fission.
For more details on ways to contribute to LongOps, please visit: https://ukaeaevents.com/longops-supplier-day
Comments are closed.