Drilling Dysfunction Detection via Recursive Amplitude Envelopes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Real-time well drilling operations face challenges in detecting dysfunctions that can lead to costly drill system failures due to difficulties in achieving computational efficiency, causality, and minimum information drift, particularly in signal anomaly detection.
Innovation Solution
A method is developed to compute dysfunctions via amplitude envelopes that deviate from a mean state behavior, using a recursive application of the maximum signal value within a given window size, with a dysfunction operator defined as the relative change of the envelope with respect to the mean signal, enabling causal and computationally affordable calculations suitable for high-sample-rate data streams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time signal processing is performed using traditional methods, then dysfunction detection capability is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent segments the continuous signal processing task into discrete temporal windows. By dividing the signal into segments and processing each window independently to extract envelope and spectral features, the computational complexity is reduced while maintaining dysfunction detection capability. This segmentation allows for efficient real-time processing without requiring complex global analysis of the entire signal stream.
Solution Approach 2:
The patent extracts only the most relevant features (envelope characteristics and spectral components) from the raw drilling signals, discarding redundant information. By taking out and focusing solely on the critical dysfunction-indicating features rather than processing the complete raw signal, the system achieves high detection precision with reduced computational burden.
2Measurement precision
If comprehensive signal analysis is performed to detect all dysfunctions, then detection accuracy is improved, but processing speed decreases
Solution Approach 1:
The patent applies local quality analysis by examining signal characteristics within specific temporal windows and frequency bands relevant to particular dysfunctions. Rather than performing uniform comprehensive analysis across the entire signal, the system focuses computational resources on local signal segments and frequency ranges where dysfunction indicators are most prominent, thereby maintaining high detection accuracy while improving processing speed.
Solution Approach 2:
The patent implements partial action by performing analysis on selected signal features and windows rather than exhaustive processing of all signal data. The system processes only the necessary subset of signal information required for dysfunction detection, avoiding excessive computation on redundant data, thus achieving fast processing without sacrificing detection accuracy.
3Stability of the object's composition
If large data windows are used for signal processing, then measurement stability is improved, but responsiveness to sudden dysfunctions deteriorates
Solution Approach 1:
The patent employs dynamic windowing where the analysis window size and positioning adapt based on the drilling conditions and detected signal characteristics. The system can adjust the window duration and overlap dynamically, using larger windows for stable operations to improve measurement stability, and smaller, more frequent windows when dysfunctions are detected to reduce detection delay and respond quickly to changing conditions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods compute dysfunctions via amplitude envelopes that deviate from a mean (normal) state behavior. The envelope function is constructed from a recursive application of the maximum signal value within a given window size. The aforementioned operations are causal and computationally affordable as relative short moving windows are required to trail the current point. Therefore, the proposed envelope and dysfunction calculations are amenable for any source of data streams measured at high sample rates. The effectiveness of the computing is validated as representing multiple physics in real time field drilling operations.