PHASE-FREQUENCY DETECTION OF SEISMIC SIGNALS

Abstract

Relevance. Currently, in the search for oil and gas fields, much attention is paid to more active involvement of poorly explored territories in exploration, as well as to the study of natural reservoirs in areas of industrial development of deposits. For these purposes, seismic methods are widely used, as a result of which huge data arrays are formed, which are subject to subsequent processing and interpretation. One of the tasks to be solved at the processing stage is reliable detection of seismic signals, this task is especially important when tracking waves, since the quality of construction of reflecting boundaries depends on the reliability of detection. Therefore, the construction of effective methods and algorithms for detecting signals based on the analysis of the phase response, with a higher degree of resistance to interference, is a very urgent task. The aim of the research is to construct methods for detecting signals based on the optimal and suboptimal processing of the phase-frequency characteristic of seismic waves obtained when searching for oil and gas fields and to study these methods effectiveness on models of seismic pulses by calculating the probability characteristics of detection errors. Methods and means of research: methods of statistical analysis and synthesis of algorithms, theory of random processes, discrete Fourier transform, methods of processing and interpretation of seismic data, mathematical modeling and computational experiment. Results. Based on the maximum likelihood method, an optimal procedure for detecting seismic signals by their phase-frequency characteristics is constructed. Analytical expressions are obtained for the probabilities of detection errors. It is shown that optimal phase-frequency detection provides results close to the results of the absolutely optimal signal detection method. For practical implementation of the developed method, suboptimal processing of the phase response (equilibrium processing) is proposed, which does not require information on distribution of signal energy in the frequency domain. The noise immunity of the detection method with equilibrium processing was studied and estimates of losses were obtained depending on the number of used frequency components in the signal spectrum. The place of application of the developed detection methods for solving problems of tracking seismic waves is shown.

Keywords

• seismic waves
• optimal phase-frequency detection
• likelihood ratio
• equilibrium processing
• seismic wave tracking