Font Size: a A A

Weak Signal Detection Algorithm For The Array Induction Logging Tools

Posted on:2010-02-24Degree:MasterType:Thesis
Country:ChinaCandidate:J W FanFull Text:PDF
GTID:2120330338487108Subject:Detection Technology and Automation
Abstract/Summary:PDF Full Text Request
Logging signals have the characteristics of low power and low SNR. One of the key technologies that logging tool developed by is how to detect logging signals exactly and effectively.The weak signal processing algorithms on the array induction logging signals is discussed. According to the characteristics of the logging signal, a noise analysis model that contains the sinusoidal signal and Gaussian white noise is established, and the noise analysis model is processed by some weak signal processing algorithms, such as wavelet analysis, digital phase-sensitive detector, multiple autocorrelation and cumulative average. Matlab simulation results show,wavelet analysis is suitable to process non-stationary transient signal, the SNR of which is not too low. Digital phase-sensitive is suitable to process sinusoidal signal, the SNR of which is lower, the longer the length of the signal, the better the effect. Multiple autocorrelation can detect the frequency of signal accurately, but the amplitude and phase of signal have some errors. Cumulative average is for processing the periodic signal, the more average number, the better and the effect. According to the advantages and disadvantages of these algorithms, on this basis, a new detection method that firstly uses multiple autocorrelation and digital phase-sensitive detector is given. After simulation, the effect of the new method is better than single algorithm. In order to verify the capability that the method processes actual signals, a weak signal detection platform which simulates the logging tool to detect 100nv level signal is set up. The platform contains attenuator, low-noise pre-amplifier circuit and band-pass filter circuit .When the amplitude of input signal is 100nv, Amplitude error and phase error of output data is 6.4% and 6.3%.Sum of all, the algorithm that firstly uses multiple autocorrelation and digital phase-sensitive detector can detect 100nv level signal accurately. Migrating this algorithm to the array induction logging tool helps to improve the ability that array induction logging tool detects weak signal.
Keywords/Search Tags:array induction logging, weak signal detection, wavelet Analysis, digital phase-sensitive detector, multiple autocorrelation, cumulative average
PDF Full Text Request
Related items