How Ex certified robots are finding a niche in remote and hazardous oil and gas sites
02 February 2018
It is perhaps surprising that although robots have been making cars for decades and are roaming the surface of Mars, we do not see them at onshore oil and gas facilities. Of course tethered, remotely operated vehicles are widely used by the subsea sector, but they have never made the evolutionary step onto land. That is about to change, says Ian Peerless of ExRobotics.
One of the Kashagan field's islands in the frozen Caspian Sea - Image: ENI
The case for using robots to improve financial performance and safety has never been stronger, and the technology has progressed to the point that credible solutions are ready to be deployed.
There are a number of good reasons to deploy robots in remote and hazardous facilities such as Kashagan in the Kazakhstan zone of the Caspian Sea, which, with its harsh climate and high levels of hydrogen sulphide, is not a welcoming place for humans. Another important factor is that when humans are deployed in these hostile environments, their productivity is hampered by protective clothing and breathing apparatus.
At Kashagan, there are plans for robot operators to be permanently stationed on production islands, to be activated and operated by humans sitting in the comfort of a distant control room. Should the unexpected occur, the robot would immediately respond and might be able to resolve the problem and restart production long before humans arrive on the scene. And even if it cannot restart production, it can diagnose the situation so that the intervention team is better prepared and aware of any hazards it might face. The team might then use the robot to perform the most hazardous tasks.
At the other end of the scale, some companies are operating large numbers of normally unmanned facilities such as NAM’s onshore gas fields in The Netherlands. For operational and/or regulatory reasons, these need to be visited to respond to unplanned events or check for potential issues.
In the past, human operators would drive around the motorway system and visit three or four sites in a day. Now, they can send out a robot from a central control station, significantly improving productivity as well as reducing response times and the risk of road accidents.
Unconventional oil and gas fields may also have hundreds of widely dispersed well-pads and facilities.
And in the offshore environment, normally unmanned offshore platforms can be expensive and slow to reach, especially when the weather is bad, making robots an effective solution here also.
A common response is to ask “why robots?” Remote instrumentation and controls seem much simpler. However the reality is that it is often inefficient to install these items where they will seldom be used. It is also difficult to predict everywhere they might be required. How many normally unmanned installations do you know that are truly unmanned? Robots offer mobile instruments and tools that can move from one need to another. In this way they complement the conventional fixed-instrument approach.
Truly practical robot operators for land and platform applications are a recent phenomenon. The technical challenges are considerable. The most obvious is the need for IECEx, ATEX or similar certification. Others are:
• Communication – how does the human at a computer screen control and monitor the robot?
• Charging – how is the robot kept operable for months or even years without human intervention?
• Robustness – the robot needs to resist the harsh conditions in difficult locations.
• Reliability – it is no use removing humans from a location if the robot’s maintenance engineer has to visit every week.
• Working Environment – for the moment at least, these robots need to operate in facilities designed for humans. They need to be scaled accordingly and must be able to traverse steps and sharp corners.
In 2010, these were the challenges faced by a project team funded by the North Caspian Operating Company (NCOC) and led by Shell’s Frontier Automation Team. Over the last seven years, and three generations of robot, the challenges have been progressively overcome to the point that a commercially available robot operator is about to hit the market.
Three generations of Ex robot operators
The first generation robot (left) was called Sensabot. It was developed by the National Robotics Engineering Centre (NREC) at Carnegie Mellon University and was amazingly successful for a first attempt. It was still being driven years later and has the honour of being signed by US President Obama, UK Prime Minister Cameron, and Kazakh President Nazarbayov. However it also demonstrated the challenges of certifying such a complex electrical machine for use in potentially explosive environments. The over-pressure approach (Ex “p”) worked, but was unsuitable for use without regular human intervention. The pressure regulation system was complex, the over-pressured compartment leaked, and topping up the pressure tank was impractical.
Sensabot Mark 2 (below) tackled this challenge head-on. Instead of over-pressure it used a wide range of protection methods especially Ex “d”, Ex “e”, and Ex “i”. This resulted in a hugely complex certification programme for Shell, who had taken over the project from NREC. UL LLC, the certification body, ultimately issued 13 new IECEx certificates for the robot which in 2016 became the first remotely operated resident robot certified for use in potentially explosive environments. Sensabot Mark 2 is now with the customer, NCOC, in Kazakhstan. It has continued the tradition of attracting attention. In 2017 the development team was awarded Shell’s Excellence Award for Technology and Innovation.
Part of Sensabot Mark 2’s complexity stemmed from its extensive functionality:
* 4 stereo cameras and 1 pan tilt zoom camera
* Multiple warning and safety devices
* An arm for accessing hidden spots.
* 6 cameras mounted on the arm and robot body.
* 14 lights
* 1 thermal imaging camera
* 1 hydrocarbon and hydrogen sulphide gas detector.
* 1 microphone
* 1 vibration detector with 2 positioning lasers.
* Cogs for using vertical racks to access different module levels.
In 2017, some of the companies with experience of Sensabot Mark 2 formed a new company, ExRobotics B.V. This company accepted a different challenge from NAM - “We need a much cheaper, simpler robot operator to deploy in 2018 on our unmanned locations in The Netherlands”. The result is ExR-1 (below) which is a commercially-available robot and whose field trials will be completed by the end of 2017. This robot comes with three cameras and a microphone as standard, with the option to add lights and a gas detector. It has added two more protection methods (Ex “q” and Ex “m”) and more ‘off-the-shelf’ Ex components to reduce cost. The results have been impressive.
Whereas a Sensabot Mark 2 would cost almost €1 million, ExR-1 costs between €50,000 and €100,000 depending on functionality and the quantity ordered. When compared to the cost of mobilising a conventional operator, this makes the use of a robot operator financially attractive even in relevantly benign deployments.
So what’s in store for the future? ExR-1 perhaps indicates that future robot operators will focus on specific tasks rather than being multi-functional brutes. Each type of robot may build on and add to the assemblies and learnings that are encapsulated in ExR-1. ExRobotics has an ambitious development roadmap for these future robots. It adds increasing levels of autonomy and manipulation.
Both of these development paths are within reach:
* Robots already exist that can play football or chess autonomously. Compared to this, navigating a pre-determined route and triggering an alarm if gas is detected seems rather simple. More sophisticated autonomy such as pre-programmed missions, automatic sensor reading with data transfer, and collaborating robots will inevitably follow once robot operators are being used.
* Strong, highly dextrous robotic arms have been in existence for decades. The remaining challenges are to add Ex certification, reliability, and robustness in harsh environments to the mix.
However ExRobotics will not be pursuing these developments in splendid isolation. The successes to date have been achieved through intimate collaboration between robot developers, strategic suppliers, certifying authorities like UL LLC, and end-users like NCOC, NAM and Shell. Sharing financial and technical risks has ensured that the next generation of robot is targeted at a real business case and has optimised the robot’s capabilities. Future robots will be developed in partnership with similar thought-leaders and the roadmap will evolve based on tangible market needs.
Will the use of the robot operators be constrained to upstream and downstream oil and gas? This seems unlikely. Although the hazards are different, industries such as nuclear, chemicals and mining would surely benefit from deploying more robots. The robust independence of robots such as the ExR series is likely to evolve to meet the challenges of these different environments.
“With the ExR-1 robots, the industry will have a new alternative way to do plant supervision & inspection; the remote control & visualisation eliminates the need for travelling to locations, greatly reducing safety exposure and costs. The low cost & ATEX Zone 1 certification of the ExR-1 platform will allow for easy enhancement with additional sensors & tools, increasing the value by executing more complex tasks. The ExR-1 helps the industry to make a big step in digitalisation, in which autonomous robots will permanently monitor assets.” Berry Mulder, Head of the Robotics Centre of Expertise in Shell.
“In the past 12 months, every single business leader—including our executive committee—has digitalisation on his or her agenda. The technology of using robotics, remote monitoring, and advanced analytics to drive efficiency in our manufacturing and production is key to our core business.” Alisa Choong, EVP digitalisation in Shell.
About the author
Ian Peerless has degrees in Civil and Petroleum Engineering and started his career as an engineer on Shell’s North Sea oil platforms, later specialising in subsea facilities. After leaving Shell, he spent 15 years in the steel industry in a range of senior positions.
In 2006 he returned to Shell as Executive Consultant, leading to a technology role within a Kashagan project team where his involvement with robot operators began. Since 2010 he has been the customer representative for the Sensabot Mark 1 project and has been the project manager during much of the development of Sensabot Mark 2 and ExR-1. During this time he has built up a wealth of experience in Ex certification and the challenges facing robot operators.