There is a story related to Giotto’s Perfect Circle, where legend has it that Pope Benedict IX, seeking to test Giotto di Bondone’s skill before commissioning him to paint a fresco, asked Giotto to prove his talent by drawing a perfect circle freehand. Giotto (ref) was an Italian painter and architect from Florence during the late Middle Ages and is often considered one of the first great artists of the Renaissance (see his artwork in the main article image, Capella degli Scrovegni Chapel). Giotto accomplished this feat effortlessly, impressing the Pope and securing the commission. While the story may be a bit of a legend, it reflects the admiration for Giotto’s artistic abilities and his reputation as a master artist.
The test itself is a tough one, have you ever tried to draw a perfect circle? Why not give it a go; on a piece of paper and without using any tools to help, attempt to draw the best circle you can…if anyone is brave enough to post their attempt in the comments, there’s a one&zero moleskin up for grabs! The point being, there are limitations with humans to be able to accurately, and repeatably do a seemingly simple task.
But do these limitations apply to the digital world? A computer system using Boolean Algebra (ones & zeros…) can produce a perfect circle using mathematical equations (the mathematical equation for a perfect circle in a Cartesian coordinate system is (x-h)2 + (y-k)2 = r2)). It can not only produce a perfect circle, but it can do this consistently, repetitively, and at a much faster rate than a human could produce.
But how does this story relate to drilling and logging a well bore you might ask? The answer is automation. After attending a new technology showcase by Halliburton Sperry Drilling, several newly developed and released services caught our attention and deserve further exploration. A key theme of the event was automation, and how it enables us to leverage technology to optimise well construction and data interpretation, ultimately delivering a better wellbore with better data quality. Automation was presented as a means to augment, not replace human activities, supplementing and enhancing the services being provided.
A prime example of this was LOGIX® Autonomous Drilling Platform (halliburton.com), a closed loop drilling control system that can automatically respond to changes in well path to correct to the planned well trajectory, adjust drilling parameters to reduce vibration patterns detected by the downhole tools, as well as automatically and continuously scan for offset wells for anticollision avoidance. By using automation, sophisticated algorithms combined with high-frequency downhole data, this system can deliver a smoother well bore, with less tortuosity.
If we remember how difficult it was for a human to draw a circle, extend this principle to plotting a smooth curve (substitute ‘curve’ with a build/drop/turn section as appropriate). To meet the required curve rate, multiple adjustments are required to produce the doglegs in the wellbore to stay on well plan. For a human, this would typically mean multiple calculated steering commands sent to the downhole tools (e.g. downlinks) to instruct the Rotary Steerable System (RSS) in hole when to hold angle, or when to build or drop angle. But there’s more than that to consider; what force is required to achieve the curve, or what target angle is required to be met, both of which can be affected by various drilling and downhole factors impacting the dogleg output of the BHA. Throw in changes in geology, changes in formation tendency, changes in drilling parameters, changes in downhole vibration, and potentially surface changes or limitations (say a rig pump goes down, or MWD detection is poor so missing At Bit Inc updates), as well as a dogleg limit to get the completion down, and there’s an awful lot of information to take into account, incorporate into the plan, then implement to try and make sure that the planned curve isn’t compromised with a horrible looking dogleg or risk falling behind the curve. All this, whilst trying to maximise the ROP while limiting vibration to keep the BHA tools alive. There can be a lot to take in!
What if automation could take some of this human effort out the equation by performing key tasks or actions independently (and consistently). By using real time sensor measurements made by the drilling systems down hole, the digital system is able to autonomously and continuously monitor, adjust and performing micro corrections in the well path, to create a smoother well bore, with less tortuosity and doglegs. This alone has multiple benefits, with more consistent drilling parameters, increased on bottom time, faster ROPs, less downlinking resulting in smoother wellbores and perhaps the most critical benefit, fewer issues when running casing or completions into the wellbore. Other, less tangible benefits, include better hole cleaning, less friction in the wellbore, potentially less wear or erosion of the BHA components (RSS, MWD, LWD) and stabilisers, lower dogleg bending forces being applied to rotary connections and tool/drill collars, possibly leading to fatigue failures, and better transmission of drilling forces without stabiliser hang-ups. Logically, if the planning and drilling of the well is smoother, then there’s less likelihood of harsh drilling and as a result, should reduce NPT or service quality events from tool failures. The digital system can also be set up to automatically adjust drilling parameters for better BHA performance, or to mitigate when vibration is detected, helping prolong BHA and bit life, keeping you in hole for longer and cutting section times as well as costs.
Revisiting the perfect circle principle, imagine asking 2 different people to draw the same size circle, it would be very challenging to produce the same outputs, either as a one off or consistently. Imagine asking an individual to draw a circle once in the morning, and again at the end of a long tiring shift, would the end result be the same? Or if the individual slept poorly, has a cold, or is fatigued at all. What about crew change day, when short shifting or at the end of a long hitch, would the circle output be affected by the physical or mental changes that humans encounter? A key benefit of automation is consistency and repeatability; by eliminating some of the human tasks, input identification, forward planning and calculations, and output actions, this also helps to reduce possible human errors.
By employing automation for some of these tasks, many of these variables are removed. While not infallible, and suitable back ups, redundancies, checks and balances must be factored into the any automated task, generally once a system has been developed, tested and implemented, digital systems produce uniform results. A key benefit of automation is consistent well delivery, the digital system will autonomously optimise and deliver repeatable and accurate well positioning to ensure the wellbore stays within the desired targets.
Practically, to allow autonomous drilling operations, various levels of rig integration is required, from low level surface software for planning optimisations, to full rig integration to allow control of rig site parameters. During planning stages, computer software (DrillingXpert™) is used for risk assessment, BHA design, optimising engineering and hydraulic modelling. To allow rig site autonomy, surface systems (the LOGIX ® system) installed and running on surface PCs, along with a compatible downhole BHA system, such as the iCruise® Intelligent Rotary Steerable System (halliburton.com) allow steering and well bore placement optimisation. To allow bi-directional control of the downhole tools, a downlink skid must be installed, typically connected to the standpipe and flow line (other, manually controlled downlinks are available however require more human actions, varying of pump speed for example, to complete). For autonomous control of the drilling parameters, such as adjusting of the flow rate (mud pumps), drilling RPMs and Torque (topdrive), interface with the rig site systems is required, along with a dedicated data feed to allow remote monitoring and intervention, should it be required.
A similar system is available from Baker Hughes via their i-Trak service Drilling automation | Baker Hughes, with complimentary systems, able to monitor surface fluid properties (rheology, density, temperature, gel strengths and pressure readings), and even autonomous geosteering services, able to streamline steering decisions and increase reservoir exposure.
While the above examples are vendor specific, it is worth noting that similar systems exist directly from rig equipment providers, such as NOVs Kaizen system, IDO | Intelligent Drilling Optimizer | Kaizen | NOV which promises similar though discreetly different optimisations. Focused more on drilling optimisation, using AI and machine learning, the NOV automation system is able to either propose suggested surface drilling parameters to optimise drilling performance for a human (e.g. Driller or AD) to action, or if in control mode, directly set autodriller parameters (WOB/RPM). These parameters can be continuously and automatically adjusted to maximise ROP by analysing the currently drilled well bore drilling parameters at a much higher frequency than is possible by humans, then by comparing them to offset well data sets the system can recognise where changes should or could be made, all without the need for human input or intervention.
As well as drilling efficiencies, faster ROPs and less section times, these systems also have less tangible benefits by allowing for drillers to spend less time on repetitive and process orientated tasks, and more time, energy and a laser focus on other critical well control tasks or events, helping ensure both a safer rig floor and ultimately well.
According to research published by NOV (ref), with their autonomous platform implemented at various rig sites, several key benefits were consistently identified:
- Repetitive tasks efficiency: With its ability to consistently reduce flat time, up to 40% reductions in connection times have been recorded.
- Drilling performance improvement: the ability to optimize drilling parameters to meet the goals of the operators and drilling contractors.
- Advanced optimization techniques: the ability to allow third-party developers to build apps that can be used in the platform. Operators, contractors, R&D institutes and service providers have created successful apps being used.
A similar system is deployed by SLB via their PRECISE Automated Drilling System | SLB, a digital drilling control system that is integrated into the rig systems to allow full control of drawworks, topdrive, mud pumps and various rig floor equipment.
Technical Advisory
The adoption of automation in the drilling and logging industry has arguably been late to the party, while other industries have embraced this technology and implement it to a larger degree (see automotive manufacturing assembly lines for an example of this), the technology itself has shown to be reliable and provide clear benefits for those making use of it. While the benefits can be demonstrated with KPIs against offset wells and historic comparisons, it is important to recognise that human input is still a large part of the overall system. From installation, set up and feeding data into the system, to overview of the planning, recommendations and engineering outputs, and the actual delivery of the wells, it is vital that there is a human element to, at a minimum, sense check the work being done, or intervene at any point in the process should the outputs not match expectations. Automation should be used a means to augment and empower operations, not replace or create a ‘hands off’ well delivery. Some key points to consider when investigating if automation systems are right for your operations:
- Training & competency: As with all new systems, training and competency is vital, the alternative of simply rolling out the software and hardware and expecting the benefits may lead to the wrong results, or failure to take advance of the benefits. Specific training and competency directly related to autonomous drilling systems should be reviewed for those involved in both the planning and rig site operations. Individuals must also be experienced enough to question the outputs of the system, and empowered to react accordingly should the systems recommendations or actions do not meet expectations.
- Implementation costs & logistical limitations: The software and hardware required to deliver autonomous well delivery can be prohibitive and may not logistically be possible depending on the well/rig location, age of rig site equipment, or financial constraints. While the software for planning and drilling optimisation is less prohibitive logistically, most services will come with a financial outlay to the client. It is wise to evaluate the potential benefits (and risks) Vs the initial or ongoing outlay by carrying out sensitivity analysis to help justify any system being evaluated.
- Offset data & benchmarking: The most noticeable gains for drilling performance are typically experienced when there are sufficient data to base decisions on, i.e. offset well or field data. If you are planning a one off E&P well in a previously undeveloped field with little to no offset wells, then automation is potentially not the best option to deliver your well. Alternatively, if batch drilling or factory type wells, drilled in a developed and well understood geological area is planned, then automation should be able to leverage the significant offset data sets to improve planning and operational performance. The added benefit here is that any optimisations will be more visible against offset well performance, thus benchmarking and KPI monitoring can be carried out. This may also help justify any cost increase for the software and hardware set ups Vs improved well delivery.
Conclusion
As the energy industry sees wider adoption of autonomous systems, it is a positive to see that drilling and evaluation is also part of this digital transformation. The technology that allows us to digitise or mechanise actions and tasks that were traditionally performed by humans is now mature enough to be implemented in the often harsh environments we experience in the energy industry. With increased maturity, we also see increased reliability, consistency, and enhanced features, allowing us to do more with less. With the current industry pressures and targets we are facing, technology that allows wells to be drilled faster, smoother and importantly safer, should be embraced and applauded. By being more efficient, not just with the drilling of the hole, but with the overall well delivery process, helps reduced emissions and CO2 output, helping the industry move towards NetZero targets.
Who doesn’t like the idea of faster ROPs, getting the casing down on the first run, being ahead of AFE, all while reducing well costs and impact to the environment?
