Application of Inter-Well Connectivity Analysis with a Data-Driven Method in the SAGD Development of Heavy Oil Reservoirs

The development of heavy oil reservoirs in China is great significance to safeguard national energy security, but challenges are faced due the complex and heterogeneous reservoir properties. Inter-well connectivity analysis critical enhancing performance, as it a good way interpret fluid flow provides theoretical basis for injection-production optimization. Data-driven deep learning methods have been widely used can be employed develop surrogate models injection production infer inter-well connectivity. In this study, model performance recurrent neural network (RNN) its four variants were evaluated compared temporal prediction. comparison results showed that bidirectional gated unit (Bi-GRU) optimal algorithm with highest accuracy 0.94. A was established simulate steam-assisted gravity drainage (SAGD) research area by utilizing Bi-GRU algorithm. global sensitivity method, Fourier amplitude testing (FAST), introduced combined explain influence input variables on output quantitatively calculating each variable. Quantitative SAGD derived from proposed which effectively applied typical linear patterns five-spot patterns. varied 0.1 0.58 test applications, mutual corroboration previous geological knowledge further determine distribution interlayer reservoir. workflow study new direction analyzing inferring Northeast reservoirs.