Deep Neural Network Application for 4D Seismic Inversion to Pressure and Saturation: Enhancing Training Data Sets

In this work we tackle the challenge of estimating reservoir property variations during a period of production, directly from 4D seismic data. We employ a deep neural network to invert 4D seismic amplitude maps to the simultaneous changes in pressure, water and gas saturations. The method is applied to a real field data case, where, as is common in such applications, the data measured at the wells is insufficient for training neural networks, thus, the network is trained on synthetic data. Training on synthetic data offers much freedom in designing a training dataset, therefore, it is important to understand the impact of the data distribution on the inversion results, to define the best way to construct a synthetic training dataset. We perform a study on four different approaches to populating a training dataset and make remarks on data sizes, network generality and the impact of physics-based constraints. Using the results of a reservoir simulation model to populate our training datasets we demonstrate the benefits of constraining training samples to fluid flow consistent combinations in the dynamic reservoir property domain, uncovering the potential of deep neural networks for 4D seismic inversion, in a situation where no appropriate measured data is present.