Virtual Flow Meter Sensor Fault Isolation Using PCA Residuals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtual flow meters in resource production contexts, such as oil and gas, face challenges in maintaining accuracy due to model errors and sensor failures, particularly in subsea locations where sensor biases, drifts, and faults can compromise measurement quality.
Innovation Solution
A data-driven multivariate statistical method employing Principal Components Analysis (PCA), Weighted Squared Prediction Error, and Partial Decomposition Contribution Plots is used to detect and isolate faulty sensors in a virtual flow metering system, enhancing the robustness and reliability of sensor measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If virtual flow meters use less complex sensor systems (temperature and pressure sensors), then cost and reliability are improved, but measurement precision and accuracy deteriorate due to model errors and sensor faults
Solution Approach 1:
The system implements continuous feedback through residual analysis, where the difference between actual sensor measurements and model-predicted measurements is constantly monitored. When residuals exceed thresholds, the system triggers fault detection and isolation protocols, creating a closed-loop feedback mechanism that maintains measurement accuracy despite using simpler sensors
Solution Approach 2:
The patent introduces mathematical models and data fusion algorithms as intermediaries between the simple temperature and pressure sensors and the final flow rate calculation. These intermediaries process and reconcile sensor data, compensating for individual sensor inaccuracies and maintaining overall measurement precision
2Productivity
If virtual flow meters operate over long production site lifetimes, then productivity is improved, but measurement precision deteriorates due to sensor bias, drift, and failure
Solution Approach 1:
The system performs preliminary fault detection and isolation actions before complete sensor failure occurs. By continuously monitoring residuals and detecting anomalies early, the system can identify deteriorating sensors and isolate them from the measurement process, preventing total measurement system failure and maintaining productivity
Solution Approach 2:
The fault detection and isolation system operates autonomously without requiring manual intervention. The automated monitoring, detection, and isolation processes enable the system to self-diagnose and self-correct measurement issues, ensuring continuous productive operation while maintaining accuracy
3Measurement precision
If complex data-fusion algorithms are used to estimate flow rates, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The complex data fusion process is segmented into distinct functional modules: data acquisition from multiple sensors, model-based prediction, residual calculation, threshold comparison, and fault isolation. This modular segmentation makes the complex algorithm more manageable, easier to implement, and simpler to maintain while preserving measurement precision
Data Source
AI summary
Technical effects of the invention include use of a data-driven multivariate statistical method for the detection and isolation of sensor faults applied in a virtual flow metering context. In one implementation, the data-driven multivariate statistical method employs principal components analysis, weighted squared prediction error, and partial decomposition contribution plots.