Oilfield Development Planning Based On Genetic Algorithm Multi-objective Optimization Study

The regular dynamic prediction methods have respectively shown deficiencies because their applied development stages are different. Genetic algorithm is a global search algorithm based upon the mechanics of natural evolution. It shows its robustness in handling some complicated optimization problems. In this paper the author researched the oilfield and set up several multi-objective optimal models of oilfield development programming, on the basis of the establishment of the correlative relationship of the development index of oilfield by use of functional simulation theory, such as differential simulation method and NN method. The following work is performed:1) The in-out conjunctional relation of dynamic index of oilfield development has been presented, by using of the theory of function simulation (differential simulation, NN), on the basis of researching traditional method of reservoir mechanism to carry out the dynamic prediction for development index and the idea of system theory.2) Several multi-objective optimum models which composes the oilfield development programming has been established, based on the analysis of "decision-making variable", "objection" and "conditional restriction", by means of the in-out conjunctional relation of 1), it is followed:a) Optimization model of the production distribution (optimally distribute each index to each oil extraction plant).b) Optimization model of the production structure (solving the optimum structure of each production, including natural production, measure production, new-wells production in new zone, new-wells production in old zone, with the whole oilfield or oil extraction plant).c) Optimization model of measure structure (to decide the optimum structure of production of each measure, work-hour and measure production, including fracture, acidulation, capital repair, reperforation, drain etc.).3) Pareto genetic algorithm is formed to handle multi-objective optimization problems by combining genetic algorithm with pareto strategy. Compared with conventional approach, Pareto genetic optimization are capable of dealing with multi-objective optimization problems more conveniently and more efficiently. Based on the practical data of one oil field, actual optimized results with the cases analysis are feasible. The development programming and the architecture of oil field of structural optimization software are discussed in the last part of this paper.