基于BP算法的煤层厚度预测技术应用研究
|
摘要
煤层厚度变化情况对煤矿综采工作面布设有很大的影响,查明煤层厚度的变化对煤矿开采有着极为重要的作用。依据厚度变化的非线性特点,运用三维地震数据的运动学、动力学特征,研究了煤层反射波不同类型属性信息与煤层厚度的相关性,通过非线性人工神经网络BP算法,建立了各属性与煤厚之间的人工神经网络模型,利用反向传播学习建立煤厚预测的神经网络。针对山西某矿201工作面的煤层厚度变化,通过BP人工神经网络进行了预测,经实际探采对比验证可知效果良好。该项研究也为今后通过三维地震资料预测煤层厚度提供了相关的经验。
Coal thickness variation plays a major role in coalmine fully mechanized working face layout,thus finding out thickness variation takes vary important effect in coal mining.According to nonlinear peculiarity of thickness variation,applying kinematics and dynamics characteristics of 3D seismic data,studied relevancy between different types coal seam reflection attribute information and coal thickness,through nonlinear artificial neural network BP algorithm,established artificial neural network models between each attribute and coal thickness.Using back propagation learning established coal thickness prediction neural network.In allusion to coal thickness variation at No.201 working face in a Shanxi coalmine,through artificial neural network carried out prediction.After correlation of exploration and mining information,has verified the good effects.The research has provided related experiences for coal seam thickness prediction through 3D seismic data henceforth.
Coal thickness variation plays a major role in coalmine fully mechanized working face layout,thus finding out thickness variation takes vary important effect in coal mining.According to nonlinear peculiarity of thickness variation,applying kinematics and dynamics characteristics of 3D seismic data,studied relevancy between different types coal seam reflection attribute information and coal thickness,through nonlinear artificial neural network BP algorithm,established artificial neural network models between each attribute and coal thickness.Using back propagation learning established coal thickness prediction neural network.In allusion to coal thickness variation at No.201 working face in a Shanxi coalmine,through artificial neural network carried out prediction.After correlation of exploration and mining information,has verified the good effects.The research has provided related experiences for coal seam thickness prediction through 3D seismic data henceforth.
引文
[1]姚姚.地球物理反演[M].北京:中国地质大学出版社,2002.
[2]程建远.三维地震资料资料微机解释性处理技术[M].北京:石油工业出版社,2002.
[3]孙渊,张良,等.地震属性参数在煤层厚度预测中的应用[J].煤炭地质与勘探,2008,4:58-60.
[4]郭颜省,孟召平,杨瑞召,等.地震属性及其在煤层厚度预测中的应用[J].中国矿业大学学,2004,(5):557-562.
[5]倪逸.储层油气预测中地震属性优选问题探讨[J].石油地球物理勘探,1999,34(6):614-625.
[6]蔡军,刘卫.利用人工神经网络法预测煤层厚度研究[J].中国煤田地质,2005,17(5).
[7]林建东,霍全明,吴亦峰.多井约束三维地震反演技术在煤厚预测中的应用[J].中国煤田地质,2005,15(3).
[8]崔若飞,李晋平,庞留彦,等.地震属性技术在煤田地震勘探中的应用研究[J].中国矿业大学学报,2002,31(5):267-270.
[9]钱荣钧.地震波的特征及相关技术分析[J].石油工业出版社,2008,1.
[10]钱荣钧.对地震切片中一些问题的分析[J].石油地球物理勘探,1987,22(5).
[2]程建远.三维地震资料资料微机解释性处理技术[M].北京:石油工业出版社,2002.
[3]孙渊,张良,等.地震属性参数在煤层厚度预测中的应用[J].煤炭地质与勘探,2008,4:58-60.
[4]郭颜省,孟召平,杨瑞召,等.地震属性及其在煤层厚度预测中的应用[J].中国矿业大学学,2004,(5):557-562.
[5]倪逸.储层油气预测中地震属性优选问题探讨[J].石油地球物理勘探,1999,34(6):614-625.
[6]蔡军,刘卫.利用人工神经网络法预测煤层厚度研究[J].中国煤田地质,2005,17(5).
[7]林建东,霍全明,吴亦峰.多井约束三维地震反演技术在煤厚预测中的应用[J].中国煤田地质,2005,15(3).
[8]崔若飞,李晋平,庞留彦,等.地震属性技术在煤田地震勘探中的应用研究[J].中国矿业大学学报,2002,31(5):267-270.
[9]钱荣钧.地震波的特征及相关技术分析[J].石油工业出版社,2008,1.
[10]钱荣钧.对地震切片中一些问题的分析[J].石油地球物理勘探,1987,22(5).
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心