Research On Optimization Method Of Separate Layer Water Injection Scheme In Multi-layer Section

Nowadays,separate layer water injection is the most common production method in oilfield,and it can achieve accurate separate injection technically.However,with the continuous exploitation of oilfields,the major oilfields have gradually entered the middle and late stage of ultra-high water cut,and the accuracy of injection allocation scheme is required to be higher and higher.The calculation of stratified water injection volume and injection pressure has become an urgent problem to be solved in the research of stratified water injection in China.Therefore,an optimization method of injection allocation scheme is proposed to optimize the design of stratified water injection volume and injection pressure.In this paper,Daqing Oilfield is taken as the research object.Through careful research and combining with the theory of engineering fluid mechanics,the pressure system model is established,and the relationship between pressure loss and water injection volume in separate layer water injection and separate layer polymer injection is studied.Combined with MOTU and Nicholas curves,the calculation models of pressure loss along the way of laminar and turbulent flow in separate layer water injection and separate layer polymer injection in Daqing Oilfield are established respectively.By analyzing the existing formation parameter data of Daqing Oilfield,selecting the main control formation parameters that affect the water injection,Filling missing values of formation data with grey correlation method,combining with machine learning,the layered water injection prediction model based on BP neural network is established.10000 groups of existing data are used for model training,and 20 groups of field data are used for testing.The results show that the maximum error of prediction is 10%,which is better than the existing15%,which is the accurate classification of the oilfield Through the analysis of the simulation results,it is proved that BP neural network has good applicability in the prediction of layered water injection.Finally,an experimental test platform for optimization method of separate layer water injection and separate layer polymer injection scheme is built.The relationship between the actual pressure loss and injection volume is measured by experiment,and compared with the above model calculation value;the oil production effect of the water injection volume prediction method in this paper is measured by experiment,and compared with the oil production effect of other water injection volume prediction methods.The experimental results show that the prediction method of stratified water injection volume and the calculation method of pressure loss along the way based on BP neural network proposed in this paper are meaningful to control the stratified water injection volume and injection pressure,and can improve the accuracy of injection allocation scheme and enhance oil recovery.