基于偏最小二乘与神经网络耦合的储层参数预测

Prediction of the reservoir parameters based on the coupling of neural network model with partial least square method

将偏最小二乘回归(PLS)与神经网络(NN)耦合,建立了储层参数预报模型。利用偏最小二乘对影响储层参数的诸多因素进行分析,提取对因变量影响强的成分,从而克服了变量间的多重相关性问题,降低了神经网络的输入维数;同时,利用神经网络建模可以较好地解决非线性的储层参数预测问题。计算实例表明,本耦合模型的拟合和预报精度优于独立使用神经网络模型的精度。

This paper proposes a model for predicting the reservoir parameters based on the combination of neural network with the partial least square method. The factors affecting the reservoir parameters are analyzed by means of partial least square method to extract the most important components, so that not only the problem of multi-correlation among variables can be solved but also the amount of input dimensions of the neural network can be reduced. Besides, the application of neural network helps to solve the problem of nonlinearity of the model. The applied example shows that the proposed model has higher precision than those models based on neural network method only.