Continuous wave mud pulse data transmission method based on continuous gradation frequency keying modulation and...

Area of Science:

• Petroleum Engineering
• Signal Processing
• Machine Learning

Background:

• Measurement While Drilling (MWD) is crucial for complex wellbore navigation and subsurface evaluation.
• High-speed data transmission in MWD faces inter-symbol interference (ISI) due to motor inertia, limiting traditional methods like Frequency Shift Keying (FSK) and Phase Shift Keying (PSK).

Purpose of the Study:

• To introduce and evaluate a novel Continuous Gradation Frequency Keying (CGFK) modulation combined with Convolution Neural Network (CNN) demodulation for continuous mud pulse data transmission.
• To address and mitigate ISI caused by motor inertia in high-speed MWD data transmission.

Main Methods:

• Developed a CGFK modulation technique controlling waveform frequency ramps within symbol periods.
• Employed Convolution Neural Network (CNN) for demodulating CGFK signals, leveraging the rate of frequency change as a key feature.
• Compared CGFK-CNN performance against traditional methods (Matched Filter, SVM, LSTM) via simulations and physical tests.

Main Results:

• CNN demonstrated superior performance in recognizing CGFK signals compared to other machine learning models and traditional methods.
• CGFK with CNN demodulation effectively avoided or reduced ISI caused by motor inertia.
• Achieved favorable information transmission rates and bit error rates (BER) exceeding those of traditional FSK and PSK.

Conclusions:

• CGFK modulation coupled with CNN demodulation offers a robust solution for high-speed, reliable data transmission in MWD systems.
• This innovative approach enhances the capabilities of MWD in challenging drilling environments by overcoming limitations of conventional modulation techniques.