NMR logging quadrature detection filter algorithm
|School||Huazhong University of Science and Technology|
|Course||Detection Technology and Automation|
|Keywords||NMR Orthogonal transform Digital phase- sensitive detection Wavelet Transform|
NMR logging through the echo signal amplitude and the phase inversion process can be various fluid lithology independent information to evaluate the formation . Logging instrument decay curve of the echo signal peak measurement results directly affect the accuracy of the inversion , the effective peak value of the echo NMR logging measurements are one of the key issues . After the introduction of the principle of NMR echo signal detection under the premise , this paper analyzes the echo signal with variable fluid properties with the AM signal bandwidth , low signal to noise ratio , the detection characteristics of high real-time requirements . On this basis , the paper describes the quadrature-detected echo signal detection principle of amplitude and phase information . For measurement of the echo signal , combined with Matlab simulation works on paper for the most commonly used quadrature detection algorithm ( digital phase-sensitive detection algorithm ) filters in the average performance of a detailed analysis , presented its pass, stop-band characteristic difference in the case of signal energy to ensure the signal sampling period less the equivalent noise bandwidth, the echo center offset the drawbacks of poor ability to adapt . To solve the above problems, this paper proposes wavelet transform low-pass filter design quadrature detector way to detect the echo . Through the analysis of the principle of wavelet transform , wavelet transform has a variable time-frequency windows, no output signal filtering delay characteristics. By Matlab simulation , compared to an average type filter , the equivalent filter pass , stop-band characteristics, to ensure the same signal energy in the case of equivalent noise bandwidth is smaller ; settings can be adjusted through the decomposition level equivalent bandwidth ; sampling period several multi- center shift of the echo adaptable . Finally, in order to meet the real-time detection echo high demand , the paper detection wavelet analysis to estimate the amount of computation , relying on the operating frequency of TI's DSP2812 hardware computing speed on reasonable assumptions , gifted through parallel computing , under the assumption that the wavelet meet the requirements of real-time detection of the detector .