Reservoir Production Prediction Based on Variational Mode Decomposition and Gated Recurrent Unit Networks

Fractured-vuggy carbonate reservoirs have complex geological structures including pores, caves and fractures, which causes frequent working system adjustments makes the production prediction extremely challenging. Currently, most widely used methods in such are water drive characteristic curve machine learning based models. However, (such as well shut-in) could make unstable, provides unsatisfactory accuracy of model. In this paper, by integrating variational mode decomposition (VMD) gated recurrent unit (GRU), a novel model termed VMD-GRU is proposed to address limitations methods. The time-series data firstly decomposed using VMD into several sub-series that represent different characteristics data. GRU establish autoregression model, can extract inner each prediction. final outputs obtained aggregating result verified with real-world from Tahe oilfield China. experimental results demonstrate outperforms existing models for fractured-vuggy reservoirs.