Directional Drilling is a complex process that involves the remote control of tool alignment and force application to a very long drill string subject to various external forces. The first use of directional drilling system in oil fields was motivated by economics, and the motivation has not changed today.
Since the time when onshore drillers employed the directional drilling technology to reach offshore reservoirs, the technology has greatly developed into a distinct complex form known as the Extended-reach drilling (ERD) – a form of directional drilling considered as expensive and complex, used increasingly to tap into hard-to-produce reservoirs, making viable projects that might otherwise be considered as non-commercial. ERD systems are considered complex due to challenges with regard to geological aspects of data requirement and transmittal, reactive geosteering response times, and accuracy of well placement.
Irrespective of these challenges, the introduction of directional drilling systems is to allow production from multiple reservoirs through a single well, while maintaining the proper orientation and control of BHA trajectory at the same time, which in return reduces drilling costs and minimises the environmental impact of the drilling process.
With the immense benefits provided by directional drilling, the amount of uncertainty related to directional drilling makes accurate controlling of the bit tool face orientation while ensure the adequate rate of penetration challenging, leaving much to the traditional technique of drilling operators.
To address these challenges, an effective approach in drilling Management that specialises in the development, testing, and implementation of directional drilling systems utilizing artificial intelligence, geoscience and petrophysics should be adopted.
To adopt the synergistic utilization of cutting-edge AI, geoscience and Petrophysics in directional drilling, it is note-worthy that each approach is uniquely important. However, the successful integration of AI techniques into existing reservoir, drilling and production aspects has provided an increase in the productivity for service operations in the directional drilling system.
Directional drilling success, depending on prevalent conditions, is a function of several general factors. These include the selection of best technologies and tools, procedural optimization, concrete problem-solving, accurate prediction and rapid decision-making.
Although there is growing need for precise execution of complicated well trajectories increased the demand for directional drilling expertise. It is good to know that there are two main categories of systems used in directional drilling today: Bent-sub downhole motors and Rotary steerable systems.
Bent-sub downhole motors
Directional drilling with bent-sub downhole motors according to OTC-28633-MS is generally more cost effective and prevalent in shale and it involves two distinct operations: rotation and sliding drilling.
In rotating drilling of the bent-sub downhole motors, the entire drill string is rotated using a motor located at the surface called the top drive motor. This motor applies torque to the entire drill string and turns the slightly angled drill bit symmetrically in the hole. This action leads to a straight run for the drill bit and an increased ROP. Meanwhile, in slide drilling, the top drive motor is not used and instead the drive that rotates the bit is supplied by a device called the mud motor located near the tip of the drilling tool.
Tool face angle orientation
Maintaining the right tool face angle orientation is seen here as the Key performance Indicator when using bent-sub downhole motors. Between rotary and slide modes during the course of drilling a well to reach the target depth and location. Alignment of the direction of the drill bit is known as tool face angle orientation. There are many factors that complicate the directional drilling process including borehole friction, mud motor stall, weighting of the bit during drilling and torque effects from the drill string itself, and low bit-rate telemetry. Awareness and control of tool face orientation is critical to preserve the proper well trajectory and eliminate deviations that require corrective measures and add to well costs.
Rotary steerable systems (RSS)
Using a Steerable bottomhole assembly (SBHA) comprising a mud motor with a bent housing, an experienced directional driller can orient the bend of the motor in the direction prescribed on the well plan to steer the well on the intended trajectory in a process known as sliding. When the well trajectory is intended to remain relatively straight, the entire drill string is rotated from surface in a process referred to as rotating. A SBHA is supported from a drill string that extends from surface and is driven by a subsurface directional tool combined with the drill bit.
The conventional approach to directional drilling is to ‘slide’ the SBHA, using mud motors and bent housings, to adjust the geometry/direction of the well path. During this process, only the drill bit rotates to excavate rock while orientation of the motor’s bend angle holds constant direction as surface weight is applied and drilling fluid clears debris from the face of the bit.
Read also: ND Western calls for innovation in achieving Nigeria’s decarbonization, sustainable energy goals
Mud pulse “downlinking” sequence
When using RSS some performance indicators are easily verified and measured, while others are more complex and require in-depth analysis to extract results. Dogleg severity (DLS), build rate, turn rate, and steering times achieved by RSS settings have been used as criteria to quantify steerability. Settings programmed in the RSS tool electronics control the inclination and azimuth in which a well is drilled using a steering mechanism that continually rotates.
The drilling team utilises a technique known as mud pulse “downlinking” to modify the RSS settings in real-time. Steering commands are transmitted in a series of mud pulses called a “downlink sequence” by adjusting the rig pumps to change the flow-in rate from the surface. Location, frequency and durations of RSS downlinking sequences while on and off bottom are key reference points to extract performance indicators that impact the drilling process. Service providers have direct accessibility to RSS data retrieved from the tool’s memory (post-run) or during real-time drilling; the latter being dependent on availability of parameters sent up hole via MWD mud-pulse transmission.
Having known the two main categories of systems used in directional drilling today and the key performance indicators that quantify steerability of each directional drilling system. The assessment of the impact on drilling performance and drive directional efficiency will be based on the selected directional drilling system.
Synergistic utilisation of cutting-edge AI, geoscience and petrophysics in development, testing and implementation of directional drilling operation.
Well planning
It is no longer news that directional drilling in the energy industry is a challenging task where gains in efficiency may be realised using Artificial Intelligence (AI) that provide tangible value to the drilling operator. Complications arise in managing and controlling the directional drilling system in such a way as to maximise ROP and minimize corrective steering actions.
To ensure that the adequate ROP is achieved, the synergistic use of AI with the help of petrophysical properties is employed at each step of the directional well planning. Processing past drilling data and drilling simulations using deep learning AI systems has proven to improve directional drilling guidance information for directional drillers to design and develop a directional drilling system with least possible complications.
The deep learning system which is a type of machine learning that trains multi-layered neural networks, to generate predictions based on complex patterns in previously collected data, will ingest historical and simulation data corresponding to the information used and actions taken by expert directional drillers and use that data to generate decisions that correspond with those experts would have made when presented with new situations.
Reinforcement learning is another type of machine learning that uses simulation to inform the actions taken in a given environment to maximise a particular goal or reward. Typically, an agent takes actions in an environment which is interpreted into a reward and updated state which are fed back into the agent. Applying experience in artificial neural networks directional drilling.
At the end of the project, the stated objectives were met and a feed-forward multilayer perceptron network was constructed as the first step toward deep machine learning for directional drilling.
In conclusion, the use of AI with geoscience and petrophysical properties have been of great benefit to the directional drilling systems.
Nwakwesi is an energy expert
