Research On Production Prediction Of Dynamic System Waterflooding Model Based On Fuzzy-PSO

Oilfield production process can be regarded as a complex dynamic system problem,and oilfield production is affected by a variety of factors(such as geology,fluid,development mode,etc.),so there is both certainty and randomness for the production well.Scientific production prediction is particularly important for oilfield production management and next work plan.At present,empirical statistical method and numerical simulation method are commonly used in oil fields.However,these two methods have limitations.There are too many subjective factors in empirical statistical method,which requires a deep understanding of reservoir engineering knowledge and historical development of oil fields.The numerical simulation method requires very comprehensive data to accurately reflect the formation and production situation,which takes a long time.For low permeability reservoirs,there is a large start-up pressure gradient,and the percolation law of the fluid in the formation no longer conforms to Darcy’s law,which cannot be described by the current common well connectivity model.Based on the above analysis,this paper established a novel model suitable for predicting oil field fluid production based on the analysis of inter-well connectivity.This model did not make quantitative assumptions on flow rate(injection rate and production rate)and pressure,but indirectly expressed the relationship between pressure and flow by introducing the implicit variable of volume change rate of water flooding.Through the combination of the maximum expectation algorithm and PSO fuzzy self-adaptive solution,the model parameters and system state variables optimized to make the prediction results more accurate.The results show that the improved algorithm is more accurate.At the same time,the accuracy of the novel water flooding model is very high in the four different physical properties of the conceptual model which include high permeability,medium permeability,low permeability homogeneity and low permeability heterogeneity.In the conceptual model the relative error is within 5%,while the error in the actual oil field is relatively large,but the trend of liquid production is consistent.The error accuracy depends on the interference of human factors in the actual production process,which can provide some guidance for the subsequent development.