The Vectorial Optimization Method Of Well Pattern Infilling In Waterflooding Oilfield

Vectorial infilling method is one of the cores of development adjustment project which matches with the permeability anisotropy,reservoir heterogeneity and development characteristics.Nowadays,the vectorial parameter representing the displacement degree in the vectorial infilling method has not been developed.The vector property has not been fully reflected.Meanwhile,the heterogeneity of interlayer and horizons is serious during high water cut stage.It is necessary to establish the well pattern infilling vectorial development project for multilayers heterogeneous reservoir.Besides,the traditional optimization method is difficult to optimize the well pattern infilling vectorial development project effectively and automatically,which limits the application of vectorial development project in actual reservoir.Therefore,based on the theory of vectorial development,in order to realize the balanced displacement,it is of practical significance to investigate the vectorial optimization method of well pattern infilling in waterflooding oilfield.Based on reservoir engineering theory and fluid flow in porous media,for the whole development stage,taking the permeability anisotropy,reservoir heterogeneity and development characteristics into account,the displacement vector parameter was derived for the optimization of the vectorial infilling project.Based on Theil index,the evaluation method of balanced displacement degree was built to provide a more comprehensive evaluation for the whole reservoir,the plane and the vertical.And the validation of the proposed method was performed.Because the number of variables was unknown in well pattern infilling optimization which could not be solved,the gradual optimization was used to solve this problem.Firstly the vectorial primary optimization of well pattern infilling was performed.Then the vectorial precise optimization of well placement was conducted to optimize the parameters of infilling wells.For the vectorial primary optimization of well pattern infilling,the development indexes predication method for multi injection-production groups and multilayers was developed.Based on the theory of vectorial development and predicted development indexes,the mathematical model of vectorial primary optimization of well pattern infilling was built,in which the objective functions were the maximum balanced displacement degree and net present value.This model can determine the number of infilling wells,primary well placement,infilling well type and the well type change method of existed wells.The mathematic model of well pattern infilling vectorial primary optimization is multi-objective,nonlinear and 0/1 planning.The Multi-objective Binary Cellular Cat Swarm Optimization algorithm(MOBcCSO)was developed.The population diversity is improved by cellular.The results of problem testing show that MOBcCSO has the advantage in convergence property,the diversity and uniformity of solution distribution.The vectorial primary optimization method of well pattern infilling was established and validated in well pattern infilling examples with five-spot well pattern and inverted nine-spot well pattern.In order to precisely optimize the well placement,with the objective function of maximum balanced displacement degree,the infilling well placement vectorial precise optimization mathematic model was established,involving available infilling area constraint,minimum interwell distance constraint,well length constraint,azimuth angle constraint and deviation angle constraint.This model can solve the optimization of the vertical well,the inclined well and the horizontal well in multilayered reservoir.This model is high-dimensional,multimodal,and nonlinear.Meta-optimized Hybrid Cat Swarm Mesh Adaptive Direct Search algorithm(O-CSMADS)was established.O-CSMADS has both global and local search ability.It can also automatically determine the optimal algorithmic controlling parameter.The testing results on benchmark functions show that O-CSMADS has the advantage in optimization accuracy,convergence speed and stability.Rejection method,repair method,static penalization method,dynamic penalization method and adaptive penalization method were respectively applied to handle well placement constraints and then adaptive penalization method was selected to handle nonlinear constraint.The vectorial precise optimization method of well placement was established and was compared with traditional well placement optimization method.Results show that the proposed method can greatly improve optimization efficiency,while obtaining high precision.The established vectorial optimization method of well pattern infilling in waterflooding reservoir was applied to optimize the infilling well placement in an actual reservoir.This method was validated by comparing the development performance of optimized and actual placements.Then this method was applied to optimize the well pattern infilling projects for another two actual oilfields.The development performance under the current and optimized projects was compared and analyzed,which could provide guidance for the two oilfields’vectorial well pattern infilling projects.