Surface Sensor Drilling Data Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems for detecting downhole events during drilling are ineffective due to the high cost of deploying and maintaining downhole sensors, and the challenges of transmitting real-time data from downhole to the surface.
Innovation Solution
A system and method using surface sensors, a processor, and a pre-trained machine learning model to detect and mitigate downhole events by processing drilling data from the surface sensors into segments and applying these segments to the model to determine labels corresponding to downhole events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If downhole sensors are deployed to detect downhole events, then detection accuracy is improved, but deployment and maintenance cost increases significantly
Solution Approach 1:
The patent creates a virtual copy of downhole sensor data by training a machine learning model on paired surface and downhole sensor data. The model learns to predict downhole conditions from surface measurements, effectively copying downhole information without physical downhole sensors during operation.
Solution Approach 2:
The patent introduces surface sensors as an intermediary that indirectly measure downhole conditions. Instead of placing sensors directly in the harsh downhole environment, surface sensors detect vibrations and other signals that correlate with downhole events, serving as a mediator between the drilling operation and detection system.
2Loss of time
If downhole sensors are used to transmit real-time data, then detection timeliness is improved, but data transmission complexity increases due to bandwidth constraints
Solution Approach 1:
The patent extracts the data transmission problem by moving all sensing operations to the surface where communication infrastructure already exists. Downhole event detection capabilities are extracted from physical downhole sensors and embedded in a machine learning model that runs on surface computing equipment, eliminating the need for complex downhole-to-surface data transmission systems.
Solution Approach 2:
The patent replaces the mechanical/physical data transmission system (downhole sensors transmitting through mud telemetry or wired connections) with an information-processing system. Surface sensors capture data that is processed by machine learning algorithms to infer downhole conditions, substituting direct physical measurement and transmission with indirect measurement and computational analysis.
3Ease of manufacture
If conventional surface sensor systems are used, then cost is reduced, but detection capability is limited without pre-defined features
Solution Approach 1:
The patent makes the detection system dynamic by using machine learning models that can adapt to different drilling conditions, well types, and formation characteristics. The model is trained on diverse data and can be retrained or fine-tuned for specific applications, allowing the same surface sensor system to detect multiple types of downhole events across various drilling scenarios without requiring pre-programmed detection rules.
Solution Approach 2:
The patent changes the operational parameters of the detection system by transitioning from rule-based detection with fixed thresholds to machine learning-based detection with adaptive parameters. The model learns optimal detection parameters from training data, allowing it to automatically adjust to different operating conditions and improve detection capability while maintaining cost-effectiveness.
Data Source
AI summary
A method is provided to determine and mitigate one or more downhole events of a drill string in a wellbore. The method includes receiving, via a processor, drilling data from one or more surface sensors. The drilling data is processed into a plurality of segments. Each of the plurality of segments are processed by a pre-trained model such that one or more labels are determined. The one or more labels are relating to one or more downhole events corresponding to each of the plurality of segments.