Multi-objective Optimization Algorithm And Its Application In Reservoir Management

The majority of the engineering and scientific problem is a multi-objective optimization problem, the presence of multiple conflicting goals, how to obtain the optimal solution of the problem, always is academic and engineering circles the center of attention problems. With the single objective optimization problem, multi-objective optimization is, in most cases, a target improvement may cause other target performance was reduced, while the multiple targets are reached optimal is impossible, only in the target coordinate between balance and compromise, so that all the objective function as much as possible to achieve optimal, and the optimal solution of the problem by a large number, or even infinite Pareto optimal solution.The traditional multi-objective optimization method is to the target into a single target with positive coefficient, coefficient by decision makers or by the optimization method of adaptive adjustment. These methods include the linear weighted method, constraint method. As a result of these methods require the multi-objective optimization problem of the objective function and the constraint problem is linear, continuous or differentiable. In reality, these conditions are hard to meet, so the algorithm more steering intelligent optimization algorithm, which is in a class by simulating a natural phenomenon or process and set up optimization method, this algorithm including evolutionary algorithms, particle swarm algorithm, tabu search, scatter search, simulated annealing, artificial immune systems and ant colony algorithm. And the traditional mathematical programming method, the intelligent optimization algorithm for solving multi-objective optimization problems. First of all, most intelligent optimization algorithm can simultaneously process a set of solution algorithm, each run once, can obtain more efficient solution. Secondly, the intelligent optimization algorithm for Pareto optimal front end shape and continuity is not sensitive, can be very good approximation of convex or discontinuous optimal front. The artificial immune system algorithm performance is especially outstanding, which improve the convergence speed at the same time, can better maintain the population diversity. As a heuristic search algorithm, has been successfully applied in the field of multi objective optimization, there are some hot research direction.The modern reservoir management is first to obtain the best interests of the target, using a variety of effective means and methods, access to as much as possible the economic benefits and relatively high recovery. Development to the modern reservoir management has not only focused on integrated geological and engineering, but also because of the information, tools, technology and talent in all aspects of the comprehensive. Is inevitable to take into account many aspects of the problem, need to establish a variety of different objective functions, these functions often can not be taken into account, even contradictory. Multi-objective optimization can be applied to investment decisions, timing target cloth well and measures to optimize. So multi-objective optimization methods can just be combined with reservoir management together so that enterprises can gain more benefits.At the beginning, the paper describes a multi-objective problem, and on this basis, the establishment of a feasible solution is Pareto optimal solution to determine the matrix, thus for the selection of Pareto optimal solution to provide new ideas and methods, then discusses the problem of multi-objective optimization algorithm, given the merits of the standard evaluation algorithm, a more comprehensive understanding of the multi-objective optimization. Then details of the artificial immune algorithm, and made great improvements. After the discourse the concept of reservoir management, and the characteristics of modern reservoir management, reservoir management and given the appropriate steps and evaluation indicators, defined the modern reservoir management using multi-objective optimization to improve efficiency. Finally, oil and gas fields in the development process instance to point to an area demonstrated multi-objective optimization in reservoir management.