【摘要】 针对地震勘探资料湮没在随机噪声中的微弱同相轴问题提出基于混沌理论的混沌振子检测算法.利用修正的Duffing_Holmes方程建立检测微弱同相轴的混沌振子系统,之后经过对同相轴的扫描处理,构成新子波等时间间隔序列W(t),与此同时对随机噪声也进行相同的截断.截断的随机噪声在混沌振子系统中可以具有与周期信号相同的表现;经过大量仿真实验确定出满足通常地震勘探子波延续时间的使混沌振子检测子波不呈现周期相态的随机噪声截断时间范围.选用与松辽盆地T1、T2反射层类似的子波函数并构成待检微弱周期信号,经过MATLAB仿真试验成功地检测出该弱信号,信噪比达到约-10.3dB.更多还原

【Abstract】 We present an algorithm of chaotic vibrator detection for weak events submerged in stochastic noise in seismic prospecting data based on the chaotic theory.The algorithm founds the chaotic vibrator system detecting weak events utilizing the modified Duffing-Holmes equation,and then scans the events to construct new wavelet sequences W(t) with equal intervals.At the same time,stochastic noises are also truncated similarly.The truncated stochastic noises take on the same behaviors as periodic signals in the chaotic vibrator system.By a mass of simulating experiments,we confirm the time scope of stochastic noise truncation,which is comparable to the general continuous time length of seismic prospecting wavelet and ensures the chaotic vibrator system to be a nonperiodic phase-state in detecting the wavelet.We select the wavelet function similar to the reflecting layers T-(1) and T-(2) in the Songliao Basin and construct the weak periodic signal to be detected. By MATLAB simulating experiments,the algorithm successfully detects the weak signal and the signal-to-noise ratio reaches nearly-10.3dB.更多还原