Production Optimization Study Based On Cooperative Multi-objective Artificial Bee Colony Algorithm

The production and development of an oil field is a dynamic process.The form of the process is complex and changeable,and the actual needs also changes with it.How to satisfy the actual needs of different stages of oilfield development is an urgent problem to be solved.The goals of different stages of development are diverse and not singular and immutable.Therefore,simple single-objective optimization is not enough to solve the current problems.From the perspective of multi-objective optimization,this paper combines the main contradictions in the early and middle stages of oilfield development,and specifically analyzes the multi-objective optimization problems in different oilfield development stages.Firstly,in order to get the good performance solution of multi-objective optimization problems,this paper proposes a novel multi-objective optimization algorithm,which is based on artificial bee colony algorithm.The algorithm makes full use of the colony's own experience and the collective experience which can be gotten from external archives.Through the information sharing strategy and the elite search strategy in external archives,it can quickly approach to the Pareto front and jump out the local Pareto front.The cooperative multiobjective artificial algorithm has the characteristics of fast convergence,strong diversity,and uniform Pareto front distribution.In the early stage of oilfield development,because of the lack of geological data and other reasons,the geological model has great uncertainty.At this time,the current main contradiction is how to reduce the development uncertainty.This paper takes the average economic net present value,standard deviation,minimum economic net present value as the goals,and combines multi-objective optimization theory to obtain the Pareto optimal solution set.Comparison of various optimization schemes,we can find the two points.On the one hand,the solution of the the multi-objective optimization based on the average economic net present value and the standard deviation,reduces the uncertainty at the expense of the average economic net present value,which violates the goal of oilfield actual demands.On the other hand,the solution of the the multi-objective optimization based on the average economy net present value and the minimum economic net present value reduces the development risk through the minimum economic net present value,which is more in line with the actual conditions of the oil field.In the middle and late stages of oilfield development,geological uncertainty is not the main contradiction with the improvement of geological data,production data and other data.How to maxize oil recovery has become the most important thing.In this paper,flow field intensity is constructed by parameters such as the increment of water flow,flow water saturation,etc.When we compare the optimization solution of targeting at flow field intensity and the optimization solution of targeting at the economic net present value,we find that the optimization solution of targeting at the economic net present value pays more attention to the current economic benefits,and their subsequent oil production capacity is weak.The optimization solution of targeting at flow field intensity is based on the nature of polymer,exerting the maximum effect of the polymer,and the subsequent oil production ability is strongger.Multi-objective production optimization targeting at flow field intensity and the economic net present value can get a series of Pareto frontier solutions,and the oil field can select suitable development schemes according to different needs.