煤层弹性波探地雷达接收信号处理方法研究

发布时间：2018-06-06 00:36

本文选题：弹性波 + 自适应对消　；参考：《山东科技大学》2017年硕士论文

【摘要】：课题源于国家自然科学基金项目“基于连续震源的煤层反射式槽波精细探测理论研究”。课题主要目的是对超低频弹性波接收信号进行处理,包括对直达波干扰和多界面杂波干扰的对消降噪,以及对淹没在不同反射特征信号中远端地质小构造特征信号的提取。分析了超低频弹性波在煤层中的传播特性。复杂的煤系地层结构,导致弹性波在传播过程中会受多径效应影响,产生不同程度的信号时延和能量衰减,导致接收信号往往是低频微弱信号。微弱的低频接收信号又受直达波干扰、界面干扰影响较大,难以识别。而且超低频信号又具有波长大,界面干扰拖尾较严重等一系列特点。在实际工程探测中,接收到的回波信号,除了干扰强度较大的直达波及多界面杂波干扰之外,同源干扰更是影响小构造特征信号的提取因素。建立了超低频多界面信号模型,选取非因果滤波器结构的归一化最小均方对消算法,进行干扰抑制。针对接收信号特点以及干扰特点,为减小稳态误差,提高收敛速度,对步长的选取进行了改进,采用新Sigmoid函数进行步长改变,应用到非因果变步长的自适应对消中。为提取远端小构造微弱特征信号,仅利用对消处理达不到实际应用效果,需要对接收信号进行二次处理。根据盲源分离可以分离多个源信号的特点,采用盲源分离与自适应对消相结合的算法,先利用盲源分离,分离出强信号分量,并将此信号作为负反馈信号,再进行自适应多反射强信号分量对消处理。进行仿真分析,得到第三章所提的改进算法可以在更小的稳态误差和较快收敛速度条件下,取得了较好对消效果,但发现同源干扰仍存在。继续进行信号处理,仿真发现利用第四章所提混合算法可以消除绝大多数的同源干扰,尤其是强分量成分,且回波剖面图显示小构造特征也较明显。
[Abstract]:The research is based on the National Natural Science Foundation of China. The main purpose of this paper is to process the received signals of ultra-low frequency elastic waves, including the de-noising of direct wave interference and multi-interface clutter interference, and the extraction of remote geological small structure characteristic signals submerged in different reflected characteristic signals. The propagation characteristics of ultra low frequency elastic wave in coal seam are analyzed. Because of the complex formation structure of coal measure, elastic wave will be affected by multipath effect in the process of propagation, resulting in different degrees of signal delay and energy attenuation, resulting in the received signal is usually low frequency weak signal. The weak low frequency received signal is also affected by direct wave, and the interface interference is very difficult to identify. And ultralow frequency signal has a series of characteristics, such as large wavelength and serious interfacial interference. In practical engineering detection, the received echo signal, in addition to the interference intensity of direct and multi-interface clutter interference, the homologous interference is also a factor affecting the extraction of small structural characteristic signals. An ultra-low frequency multi-interface signal model is established, and the normalized least mean square cancellation algorithm with non-causal filter structure is selected for interference suppression. In order to reduce the steady-state error and improve the convergence speed, the selection of step size is improved, and the new Sigmoid function is used to change the step size, which is applied to adaptive cancellation of non-causal variable step size. In order to extract the weak feature signal of remote small structure, only using cancellation processing can not achieve the practical application effect, so it is necessary to carry on the secondary processing to the received signal. According to the characteristics of blind source separation, a combination of blind source separation and adaptive cancellation is used to separate the strong signal components, and the signal is used as a negative feedback signal. Then the adaptive multi-reflection strong signal component cancellation processing is carried out. The simulation results show that the improved algorithm proposed in Chapter 3 can achieve better cancellation effect under the condition of smaller steady state error and faster convergence rate, but it is found that homologous interference still exists. The simulation results show that most of the homologous interference, especially the strong component, can be eliminated by using the hybrid algorithm proposed in Chapter 4, and the echo profile shows that the small structure features are obvious.
【学位授予单位】：山东科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN957.51

【参考文献】