New consortium will explore how a mixture of human and robot-controlled techniques can help monitor cables and perform other tasks in the offshore industry.
Advances in automation and sensing are changing the debate over offshore safety. As technical capabilities increase, industry thinking is moving away from discussions about mitigating risk and towards ways to minimise the need for personnel to be offshore altogether. Doing this efficiently can also carry with it some innate cost benefits. Yet not all construction, monitoring and maintenance can be done remotely – the marine and offshore industries still need a degree of flexibility.
Such a scenario is being explored by researchers at Heriot-Watt University, who are developing a human-robotics hybrid solution for the maintenance and operation of offshore wind farms. Their consortium – the Holistic Operation and Maintenance for Energy from Offshore Wind Farms (HOME-Offshore) – has recently been given a share of a GBP4 million (US$5 million) grant to create advanced health monitoring of these complex assets, which will integrate the remote inspection and repair capabilities of robotics and autonomous systems to inspect the condition of critical sub-systems, such as subsea power cables, in order to identify problems early and, ultimately, extend their lifespan.
Industry is keen to see the implications of such technology. The grant includes a GBP1 million (US$1.3 million) industry contribution from DONG, Siemens Wind, GE Energy Solutions, Scottish Power Energy Networks, Offshore Renewable Energy (ORE) Catapult, Hydrason, Nova Innovation, British Approvals Service for Cables (BASEC), JDR Cables and the European Marine Energy Centre (EMEC). GBP3 million (US$3.8 million) has been awarded by the Engineering and Physical Sciences Research Council (EPSRC).
The Heriot-Watt team includes Dr David Flynn, Dr Keith Brown and Professor David Lane, and combines expertise from the university’s Ocean Systems Laboratory and the Smart Systems Group (SSG). With the UK government targeting 40 GW of offshore wind capacity by 2050, the need for such techniques is only set to become more pressing. Flynn, the director of the SSG at Heriot-Watt, commented: “The cost of achieving these targets has, until now, focused on the capital outlay for wind turbines, but budgets have largely ignored the operation and maintenance of wind farm assets, including subsea cabling.”
According to the HOME blueprint, the project will therefore involve taking a “fusion prognostics” approach, integrating online and offline data into digital representations of the asset, to predict critical subsystem Remaining Useful Life (RUL). This will be coupled with innovation in sensor technology, robotics and autonomous systems.
Fail better The crux of the matter is that even with current technologies, 70% of cable failure modes cannot be monitored in-situ. Truly accurate cable and asset health monitoring is still hard to come by, and investigations into how this can be done more accurately and effectively will form a key plank of the group’s efforts.
“After reviewing historical failure records spanning 50 years of UK subsea power cable assets, it’s clear that [these] subsea cable failures can be attributed to abrasion, corrosion and third-party impacts,” Flynn explained to InnovOil. “There is also a lack of industrial standards to account for abrasion and corrosion failure modes on the sea bed. The team will therefore look to understand the dynamics of subsea cables and the rates of degradation incurred by these failure modes.”
One possible solution lies in a technology previously explored by a Heriot-Watt team. Low Frequency Sonar was originally pioneered for use in pipeline inspection in the oil and gas industry. Subsea cables are more complex multi-layer structures, so the team will focus on advanced data analysis – based on co-central scattering theory. “The return echoes from the subsea cable structure will be used to inform a library of failure modes associated to the cable,” Flynn added. “The university is always keen to transfer technology from the laboratory to the field, so will be working closely with its spin-out company Hydrason.”
Hydrason has commercialised a wideband sonar (WBS) system modelled on the signals used by dolphins for echolocation, and uses signal data that would otherwise be discarded in conventional surveys. The BioSonar system has an effective bandwidth of 30 to 130 kHz, and can recognise, classify and detect objects and structures beneath the seafloor in real time. From the perspective of cabling, the technology allows for more accurate surveying and classification of buried cables and their contents, improving monitoring and asset management, and better informing the causes and consequences of failures.
The next stage would be to design a system with a form factor which would allow the technology to be deployed on an AUV, and to integrate it with current sensors and processing to improve its overall detection capability. In addition to funding, firms are also providing resources. In particular, Flynn said that JDR Cables would be providing access to subsea power cables and that this would help inform how the group’s findings can be used to improve future cable design, planning and installation, as well as ongoing asset management.
The big picture A further problem, as identified by the team, is that remote Condition and Process Monitoring (CPM) applications are singularly focused tools in a complex system. They target only smaller sub-systems – for example gearboxes or generators – with sensors suited to a limited task. That leaves other critical sub-systems in isolation and limits the potential to co-ordinate multiple feeds and actions in an asset.
Ensuring that automated systems can work in tandem with human-controlled ones – and vice versa – requires a conceptual approach. Two main sources of information are available in this context: knowledge extracted from the human expert and knowledge which can be inferred from processed sensor data. For decision-making to be effective, both types of data must be stored, accessed and shared by the humans or systems involved, and in as close to real time as possible.
“It is widely documented that within the oil and gas sector, 70% of offshore data collected is never analysed,” Flynn noted. “Within this work we are focused on creating actionable information where the emphasis will be on front-end data analysis as opposed to centralised back-end data processing.”
The first part of the project, as is the case with most engineering endeavours, will involve extensive modelling of a wind farm. Many current systems are isolated, and part of the scope of HOME will be investigating how these might be better monitored, or incorporated into larger programmes better able to monitor asset health across the entire system.
The team will also explore numerous physics-based modelling options – be they electrical, mechanical, thermal or lifetime/ageing – to examine benefits, and how they might be incorporated into a smarter monitoring regime. This will provide a much improved understanding of what is happening across the farm and ideally go some way towards integrating disciplines which have traditionally operated as independent ‘silos’.
In the project’s second stage, this model will be validated against real-world and experimental data, and a data analytics-driven model interconnecting performance and diagnostic information from (previously isolated) sub-systems.
Using the two together, the team will then identify advanced sensing methods, including automated and robotic inspection, which best fill information gaps, improve the quality of information and/or reduce the need for manual inspections.
A busy and interesting path lies ahead for the SSG and its collaborators as the project moves into a number of development phases. Research will inform tests at tanks at Heriot-Watt, before the team moves to demonstrations at the ORE Catapult and a final offshore trial with a working AUV. Thanks to the project partners on board, Flynn is highly optimistic about the prospects and, thanks to the addition of BASEC, that they will even help to inform new best practices and standards in the burgeoning sector.