Development Sequence Optimization of Fault-Block Oilfields Based on the Back Propagation Neural Network Model Improved by Imperial Competition Algorithm

A well-defined development sequence of each fault block in a fault-block oilfield is the prerequisite for field development plan design. Where there are very few fault-block oilfields, the method of numerical simulation in combination with economic evaluation is acceptable for comparison. However, with the increase in the number of fault-block oilfields, there will be enormous arrangements and combinations that require massive data, heavy workload, high cost and long time. In response, an improved BP neural network model based on imperialist competitive algorithm (ICA) was proposed to predict the net present value (NPV) of each fault-block oilfield in the fault-block oilfields, allowing the determination of production sequence and prioritized producing range of fault block exploitation according to NPV. Based on the actual geological reservoir database, this method avoids human subjectivity and overcomes the limitations of the existing methods, so it is of great significance to enhance the overall benefit of the development of fault-block oilfields.