The Down-hole Data Processing Module For LWD-NMR Tool

Compared with the conventional NMR wire logging tool, nuclear magnetic resonance logging-while-drilling(LWD-NMR) tool measures reservoir fluid properties real-time in the drilling process, avoiding the measurement error caused by the change of original fluid characteristic of oil well in the process of well cementation. There are two key problems in engineering implementation of LWD-NMR tool: one is that identify eligible NMR logging data according to the vibration state of drilling; the other is that compress echo data and upload it to the ground system for getting the formation information in real time.In this paper, a data processing module is designed in LWD-NMR tool, which is composed of two parts of motion measurement circuit and main control circuit. Motion measurement circuit output vibration acceleration and displacement information. Main control circuit identify NMR logging data according to the vibration state of drilling and the valid data is compressed to upload to ground data processing system of while drilling(ELIS) for getting formation fluid information.The hardware of main control circuit consists of communication circuit, data processing and its peripheral circuit. The software realizes the control function of data transmission, tool self-management and self-starting. A kind of echo data compression algorithm realized in main control circuit is analyzed emphatically(the compression ratio is 1000:6). Test results show that the T2 spectrum of decompressed echo train is not only well matched with the T2 spectrum of original model, but also has a significant improvement in signal-to-noise ratio(SNR).Motion measurement circuit consists of two two-axis accelerometer and a triaxial magnetometer and their own signal conditioning circuit. This circuit will measure tool vibration acceleration in real-time, and obtain downhole real-time vibration displacement by the time-domain integral algorithm of acceleration, and get the correction displacement by using polynomial fitting method to eliminate the integral trend error. Single axis random vibration test shows that the measurement error of acceleration power spectrum density(PSD) is less than 5%; Sweep frequency vibration test shows that the fitting precision of displacement curve through acceleration single integral is good.