Hybrid-DNNs: Hybrid Deep Neural Networks for Mixed Inputs
Rapid development of big data and high-performance computing have encouraged explosive studies of deep learning in geoscience. However, most studies only take single-type data as input, frittering away invaluable multisource, multi-scale information. We develop a general architecture of hybrid deep neural networks (HDNNs) to support mixed inputs. Regarding as a combination of feature learning and target learning, the new proposed networks provide great capacity in high-hierarchy feature extraction and in-depth data mining. Furthermore, the hybrid architecture is an aggregation of multiple networks, demonstrating good flexibility and wide applicability. The configuration of multiple networks depends on application tasks and varies with inputs and targets. Concentrating on reservoir production prediction, a specific HDNN model is configured and applied to an oil development block. Considering their contributions to hydrocarbon production, core photos, logging images and curves, geologic and engineering parameters can all be taken as inputs. After preprocessing, the mixed inputs are prepared as regular-sampled structural and numerical data. For feature learning, convolutional neural networks (CNN) and multilayer perceptron (MLP) network are configured to separately process structural and numerical inputs. Learned features are then concatenated and fed to subsequent networks for target learning. Comparison with typical MLP model and CNN model highlights the superiority of proposed HDNN model with high accuracy and good generalization.