支持向量机在储层厚度预测中的应用
|
摘要
将支持向量机运用到储层厚度估计中,利用地震属性及少量测井作为学习样本进行储层预测。通过引入 窗口核函数,准确地反映了不同深度的地质信息。通过实例将支持向量机预测结果与BP网络预测结果进行比 较,结果表明该方法有效可靠,预测精度高,可很好地解决BP网络方法中无法避免的局部极值问题。
This paper presents a new approach to predict reservoir thickness from seismic attributes and well information through support vector machine (SVM). Geological information at different depth was described by introducing a mixed window kernel function to SVM. Results from the proposed method and BP neural network show that the method can avoid the problem of local optimal solution of BP network, and obtain the estimation of thickness with higher precision.
This paper presents a new approach to predict reservoir thickness from seismic attributes and well information through support vector machine (SVM). Geological information at different depth was described by introducing a mixed window kernel function to SVM. Results from the proposed method and BP neural network show that the method can avoid the problem of local optimal solution of BP network, and obtain the estimation of thickness with higher precision.
引文
1 乐友喜,王永刚,张军华．由地震属性向储层参数转化的 综合效果分析[J]．石油物探,2002,41(2):202～206
2 Hirsche K,Boerner S,Kalkomey C,et al. Avoiding pitfalls in geostatic reservoir characterization: A survival guide [J]. The Leading Edge,1998,7(4):493-504
3 Vapnik V. The nature of statistical learning theory[M]. New York:Spring-Verlag, 1995. 314
4 Vapnik V. An overview of statistical learning theory[J]. IEEE Transactions on Neural Networks, 1999, 10(5): 988-999
5罗公亮．从神经网络到支撑矢量机(上)[J]．冶金自动 化,2001,25(5):1～5
6彭传圣,常国贞．试论地震约束反演的不适定性[J]．油 气地质与采收率,2002,9(2):88～91
7 Vapnik V,Golowich S,Smola A. Support vector method for function approximation, regression estimation, and signal processing[J]. Advances in Neural Information Processing Systems, 1996,281 - 287
8 Cherkassky V,Mulier F. Vapnik-Chervonenkis(VC) learning theory and its applications[J]. IEEE Transactions on Neural Networks, 1999,10(5):985-987
9许建华,张学工,李衍达．应用核Fisher判别技术预测油 气储集层[J]．石油地球物理勘探,2002,37(2):170～175
10 Brailovsky V L, Barizily O, Shahave R . On global , mixed and neighborhood kernels for support vector machines [J]. Pattern Recognition Letters, 1999, 20(11-13):1 183-1 190
11 Platt J C, Sequential minimal optimization: A fast algorithm for training support vector machines[C/OL]. Microsoft Research Tech. http://research.microsoft. com/users/jplatt/smoTR. pdf
12姚凯丰,陆文凯,丁文龙,等．一种基于SVM特征选择 的油气预测方法[J]．天然气工业,2004,24(7):36-38
13高美娟,田景文．基于径向高斯网络的薄互储层参数 预测[J]．石油地球物理勘探,2001,36(5):540-546
14魏永佩,陈惠鑫．人工神经网络及其在油气勘探与开 发中的应用[J]．大自然探索,1998,17(63):43～46
2 Hirsche K,Boerner S,Kalkomey C,et al. Avoiding pitfalls in geostatic reservoir characterization: A survival guide [J]. The Leading Edge,1998,7(4):493-504
3 Vapnik V. The nature of statistical learning theory[M]. New York:Spring-Verlag, 1995. 314
4 Vapnik V. An overview of statistical learning theory[J]. IEEE Transactions on Neural Networks, 1999, 10(5): 988-999
5罗公亮．从神经网络到支撑矢量机(上)[J]．冶金自动 化,2001,25(5):1～5
6彭传圣,常国贞．试论地震约束反演的不适定性[J]．油 气地质与采收率,2002,9(2):88～91
7 Vapnik V,Golowich S,Smola A. Support vector method for function approximation, regression estimation, and signal processing[J]. Advances in Neural Information Processing Systems, 1996,281 - 287
8 Cherkassky V,Mulier F. Vapnik-Chervonenkis(VC) learning theory and its applications[J]. IEEE Transactions on Neural Networks, 1999,10(5):985-987
9许建华,张学工,李衍达．应用核Fisher判别技术预测油 气储集层[J]．石油地球物理勘探,2002,37(2):170～175
10 Brailovsky V L, Barizily O, Shahave R . On global , mixed and neighborhood kernels for support vector machines [J]. Pattern Recognition Letters, 1999, 20(11-13):1 183-1 190
11 Platt J C, Sequential minimal optimization: A fast algorithm for training support vector machines[C/OL]. Microsoft Research Tech. http://research.microsoft. com/users/jplatt/smoTR. pdf
12姚凯丰,陆文凯,丁文龙,等．一种基于SVM特征选择 的油气预测方法[J]．天然气工业,2004,24(7):36-38
13高美娟,田景文．基于径向高斯网络的薄互储层参数 预测[J]．石油地球物理勘探,2001,36(5):540-546
14魏永佩,陈惠鑫．人工神经网络及其在油气勘探与开 发中的应用[J]．大自然探索,1998,17(63):43～46
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