实现从构造勘探向岩性勘探阶段的转变,是煤田地震勘探亟待解决的重要问题。其中,地震反演技术是岩性勘探的一种重要手段。为了规避常规反演方法的固有限制,利用概率神经网络技术预测井数据和地震数据之间的非线性关系,得到密度数据体和速度数据体,并获得相应的波阻抗数据体。对某矿区的实际地震资料采用该技术进行岩性反演,得到了较为准确的波阻抗数据体,为岩性解释提供了不可或缺的资料。
It is transformation from the structure to lithologic exploration that needs to be urgently solved in coalfield seismic exploration.As we all know,seismic inversion technique is a significant tool in lithologic exploration.We chose to apply probabilistic neural network method for predicting the nonlinear relationship between well data and seismic data so as to avoid the inherent limitations of the common inversion method and then predicte density and velocity volume to derive the impendence volume.Based on the thought,we use the technique to lithological inversion and obtained a quite accurate impendence volume which provide indispensible data to lithologic interpretation in one coalfield.