Schlumberger. has filed a patent for a system and method that uses sensor data to select a drilling mode during the drilling of a borehole in a geologic environment. The system then simulates the drilling process and generates a reward based on the state of the borehole and a planned trajectory. This reward is used in deep reinforcement learning to maximize future rewards for drilling actions, enabling the system to follow the planned trajectory and drill to a target. GlobalData’s report on Schlumberger gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Schlumberger, Mobile payments was a key innovation area identified from patents. Schlumberger's grant share as of September 2023 was 59%. Grant share is based on the ratio of number of grants to total number of patents.
A drilling system and method using deep reinforcement learning
A recently filed patent (Publication Number: US20230313664A1) describes a method and system for optimizing drilling operations in a geologic environment using deep reinforcement learning. The method involves receiving sensor data during the drilling of a borehole and selecting a drilling mode based on the sensor data. The drilling mode can be chosen from a variety of options, including rotary drilling and sliding drilling modes.
Once the drilling mode is selected, the method simulates the drilling process and generates a state of the borehole in the geologic environment based on the simulated drilling. A reward is then generated using the state of the borehole and a planned borehole trajectory. This reward is used through deep reinforcement learning to maximize future rewards for drilling actions. The deep reinforcement learning algorithm determines a sequence of decisions for rotating and sliding actions to follow the planned borehole trajectory and drill towards a target.
The system described in the patent includes a processor and memory accessible by the processor. The memory stores processor-executable instructions that instruct the system to receive sensor data, select a drilling mode, simulate drilling, generate a reward, and use deep reinforcement learning to optimize drilling actions. The system can also issue control instructions for drilling additional portions of the borehole using the selected drilling mode.
The deep reinforcement learning algorithm used in the method and system incorporates penalties for selecting a sliding drilling mode compared to a rotary drilling mode. It also includes penalties for transitioning from a rotary drilling mode to a sliding drilling mode, while rewarding forward drilling.
In summary, the patent describes a method and system for optimizing drilling operations in a geologic environment using deep reinforcement learning. By analyzing sensor data and simulating drilling processes, the system can determine the most effective drilling mode and sequence of actions to follow a planned borehole trajectory and reach a target. The use of deep reinforcement learning allows for the maximization of future rewards and the optimization of drilling actions.
To know more about GlobalData’s detailed insights on Schlumberger, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.