AI Anomaly Detection for Intelligent Well Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high volume of data from well instrumentation with intelligent completion systems requires significant time and effort for analysis, diverting professionals from more critical tasks, and existing AI methods often rely on labeled data, require extensive human monitoring, or are not specific to reservoir anomalies.
Innovation Solution
An unsupervised AI method using stochastic outlier detection, which processes pressure, temperature, and flow rate data to identify anomalies by extracting signatures and applying dimensionality reduction techniques, allowing for continuous monitoring and reducing the need for constant human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data preparation and analysis is performed manually by specialists, then analysis accuracy can be maintained, but the time and effort required becomes prohibitive
Solution Approach 1:
The patent replaces manual mechanical analysis processes with an automated AI system that uses machine learning models to detect anomalies in well data. The system automatically processes pressure, temperature, and flow rate data from multiple sensors, substituting the manual work of specialists with an automated computational system that maintains high accuracy while dramatically reducing time consumption.
Solution Approach 2:
The patent introduces an AI-based intermediary system that acts as a mediator between raw well data and specialist analysis. This intermediary automatically performs data preparation, processing, and initial anomaly detection, presenting only relevant findings to specialists for final review, thus reducing their time burden while preserving analysis quality.
2Reliability
If more professionals are dedicated to analyzing data from highly instrumented wells, then detection capability improves, but operational costs and complexity increase
Solution Approach 1:
The patent implements a self-service anomaly detection system that automatically monitors well data without requiring constant human intervention. The AI system independently processes data from multiple sensors, detects anomalies, and generates alerts, enabling the system to serve itself rather than requiring a large team of professionals for continuous monitoring.
Solution Approach 2:
The patent creates a universal AI system that can handle data from multiple wells and various sensor types simultaneously. This single multi-functional system replaces the need for multiple specialized analysts, as the AI model is trained to recognize diverse anomaly patterns across different well configurations and operational conditions.
3Reliability
If constant human monitoring is maintained, then all anomalies can be detected, but the focus of professionals is diluted across too many wells
Solution Approach 1:
The patent extracts and automates the routine monitoring tasks from human specialists, separating this function into an independent AI system. This allows professionals to focus their attention on complex anomaly analysis and decision-making, while the AI handles continuous monitoring and initial detection across all wells, ensuring comprehensive coverage without diluting expert focus.
Solution Approach 2:
The patent implements a feedback mechanism where the AI system continuously monitors well data, detects anomalies, and provides targeted alerts to specialists. This feedback loop ensures that professional attention is directed only to wells exhibiting anomalous behavior, maximizing the effectiveness of human analysis while maintaining comprehensive monitoring coverage through automated surveillance.
Data Source
AI summary
The main objective of the present invention is to enable continuous monitoring and detect anomalous behavior in wells equipped with intelligent completion automatically by means of a method implemented with artificial intelligence. The present invention applies AI techniques to monitor wells in an oil field and has the ability to understand what the usual behavior of each well would be, based on temperature, pressure and flow rate sensors, and then identify by means of a stochastic technique of selection of outliers which a deviation from usual behavior would be. From the outlier detection, it is possible to quantify an anomaly probability and associate the same with a possible event, such as: sensor failure and loss of data, closure of one of the producing intervals, scale deposition, among others.