Formation Fluid Contamination Assessment via Multivariate Factor Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of formation fluid measurements in hydrocarbon producing wells is compromised by contamination from drilling mud, which affects the reliability of data obtained from downhole sensors due to improper calibration and physical perturbations, especially under over-burdened or overpressure conditions.
Innovation Solution
The implementation of multivariate factor analysis using a formation tester tool that records time series measurements from multiple sensors, normalizes and combines them, and applies principal component analysis (PCA) to reduce dimensions and generate a prediction curve, allowing for the determination of contamination levels and properties of the formation fluid, while also assigning confidence values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multivariate factor analysis with multiple sensors is used, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the measurement problem by using multiple independent sensors (optical, resistivity, density, NMR) that each measure different aspects of the formation fluid. This segmentation allows the system to capture multiple parameters simultaneously, improving overall measurement precision while managing complexity through modular sensor design
Solution Approach 2:
The patent transitions from single-parameter measurements to multi-dimensional measurement space by combining data from multiple sensor types. This dimensional expansion allows the system to distinguish between formation fluid and mud filtrate more effectively, improving measurement accuracy in contaminated conditions
2Loss of time
If dimensional reduction and prediction curves are applied, then analysis time is reduced, but measurement precision may be compromised
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing prediction curves that represent the relationship between multivariate sensor responses and fluid properties. During field operations, the system simply needs to match measured values against these pre-computed curves, dramatically reducing processing time while maintaining accuracy through the robustness of the underlying statistical model
Solution Approach 2:
The patent transforms the measurement data from raw multivariate sensor readings into reduced-dimensional parameters through dimensional reduction techniques. This parameter transformation maintains the essential information needed for accurate fluid characterization while enabling faster processing and visualization
Data Source
AI summary
A system to determine a contamination level of a formation fluid, the system including a formation tester tool to be positioned in a borehole, wherein the borehole has a mixture of the formation fluid and a drilling fluid and the formation tester tool includes a sensor to detect time series measurements from a plurality of sensor channels. The system includes a processor to dimensionally reduce the time series measurements to generate a set of reduced measurement scores in a multi-dimensional measurement space and determine an end member in the multi-dimensional measurement space based on the set of reduced measurement scores, wherein the end member comprises a position in the multi-dimensional measurement space that corresponds with a predetermined fluid concentration. The processor also determines the contamination level of the formation fluid at a time point based the set of reduced measurement scores and the end member.