Methods for automated eddy current non-destructive testing analysis
A system and method for precisely detecting tubing flaws using a computer architecture that combines scalable processing power with an extensible array of detection and classification possibilities involving eddy current data analysis, as well as detection algorithms for pinpointing exact tubing regions and wherein these regions can be further divided into manageable segments for flaw analysis. Multiple classification tools utilized to discriminate detection and to precisely assign proper report codes to detect flaws repeatedly and accurately.
1. A system for the comprehensive monitoring, analysis and managing of tubing flaws, said system comprising:
a computer platform whose architecture involves a distributed processing system for supporting parallel execution of automated analysis tasks, said platform comprising a locale analyzer and a detection classification unit for pinpointing tubing flaws;
means for channel aliasing for analyzing data in terms of a particular class of datastream;
an auto final acceptance means that automatically applies analyst-configured rules to reduce forms of overcall that would result from using high probability of detection automated analysis methods; and
locale mapping means for generating settings to certain areas of the tubing and associated components, said locale mapping means implementing landmark edge correction, mapping a damage mechanism to a plurality of related analysis locales, and algorithms that involve expansion transition detection, adaptive threshold detection and locating schemes.
2. The system of claim 1 further comprising an algorithm training interface.
3. The system of claim 1 wherein said detection classification unit comprises a plurality of ratio or different measurements to be made between frequencies for voltage, angle, depth and length of said tubing.
4. A method for the comprehensive monitoring, analysis and managing of tubing flaws, said method comprising:
loading and calibrating historical data and acquired data on a computer platform;
sorting said historical and acquired data by locale;
mapping said data for generating settings to certain areas of the tubing and associated components;
detecting flaws based on said mapped data;
classifying said flaws using at least one ratio or direct difference measurement between frequencies for at least one of the group consisting of voltage, angle, depth and length; and
applying analyst-configured rules for reducing overcall and for selecting a single final result to report from among a plurality of redundant results.
5. The method of claim 4 further comprising the step of utilizing channel aliases to permit analyzing said data of a particular class of a data stream.
6. The method of claim 4 wherein said step of mapping comprises implementing landmark edge correction.
7. The method of claim 4 wherein said step of mapping comprises mapping a damage mechanism to a plurality of related analysis locales.
8. The method of claim 4 wherein said step of mapping comprises implementing algorithms that involve expansion transition detection.
9. The method of claim 4 wherein said step of mapping comprises implementing algorithms that involve adaptive threshold detection.
10. The method of claim 4 wherein said step of mapping comprises implementing algorithms that involve locating schemes.