Pump Jack Acoustic Anomaly Detection via Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Pump jacks used in low-volume oil wells face challenges in early anomaly detection and predictive maintenance due to the complexity of their mechanical operations, leading to potential performance degradation and failure, which can significantly impact oil production.
Innovation Solution
A system utilizing sensors and data processing techniques, including Bayesian inference, machine learning, and sensor fusion, to create a unique performance signature for pump jacks, allowing for real-time monitoring and visualization of downhole loads, crank motion, and bridle abnormalities, enabling early detection of anomalies and predictive maintenance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pump jacks operate without continuous monitoring, then operational costs are reduced, but anomaly detection capability deteriorates leading to delayed failure prediction
Solution Approach 1:
The patent replaces complex mechanical monitoring systems with acoustic sensing and data processing. Acoustic sensors capture pump jack operation sounds, which are then processed through signal processing and machine learning algorithms to detect anomalies. This substitution reduces mechanical complexity while maintaining or improving anomaly detection capability.
Solution Approach 2:
The patent introduces acoustic signals as an intermediary between the pump jack operation and anomaly detection. Instead of directly monitoring mechanical parameters, the system uses acoustic emissions as a mediator that carries information about pump jack health, enabling indirect but effective monitoring with reduced complexity.
2Measurement precision
If traditional monitoring methods are used, then system complexity is minimized, but measurement precision of operational parameters deteriorates
Solution Approach 1:
The patent replaces direct mechanical load measurement systems with acoustic-based inference systems. By analyzing acoustic signals generated during pump jack operation, the system derives downhole load information without requiring complex mechanical sensors, achieving high measurement precision through signal processing and machine learning.
Solution Approach 2:
The patent creates an acoustic copy or representation of the mechanical operation state. Instead of directly measuring mechanical parameters, the system captures acoustic emissions that replicate operational information, then processes these acoustic copies to extract precise measurements of downhole loads and other parameters.
3Productivity
If pump jacks are maintained only after failure, then maintenance costs are minimized, but productivity is reduced due to unplanned downtime
Solution Approach 1:
The patent implements a feedback system where acoustic monitoring continuously provides information about pump jack health status. This feedback loop enables the system to detect degradation trends and predict failures before they occur, allowing maintenance to be scheduled proactively to maintain productivity without requiring full automation of the maintenance process itself.
Solution Approach 2:
The patent enables preliminary detection of anomalies and prediction of failures before they actually occur. By analyzing acoustic signals and identifying early signs of degradation, the system allows maintenance actions to be taken in advance, preventing unplanned downtime and maintaining productivity without requiring complete automation of maintenance operations.
Data Source
AI summary
A computer-implemented method may comprise attaching a plurality of wireless sensors to a pump jack; receiving time-stamped data from at least some of the plurality of wireless sensors attached to the pump jack, at least one of the plurality of wireless sensors comprising an accelerometer and a gyroscope and being attached to a crank arm of the pump jack; synchronizing the received time-stamped data; from the synchronized time-stamped data, calculating and generating information related to: a downhole load versus polished rod position of the pump jack; a relative balance of a counterweight of the pump jack relative to a horse head of the pump jack; deviations from a nominal acceleration profile of a bridle of the pump jack; and an angle of inclination of the bridle of the pump jack; and selectively generating, on a computing device, visualizations of the generated information.