基于遗传算法的低渗透气藏水平井压裂参数优化

采用传统的正交设计方法对低渗透气藏水平井压裂进行优化设计，难以涵盖整个最优解搜索空间，优化对象数目的增加也会导致方案数量大幅增加。为了突破这一限制，基于遗传算法建立了新的水平井压裂智能优化设计方法，并编程实现了优化设计的全程自动化。计算结果表明：遗传算法在优化过程中不需要人为干预，大大降低了工作量，采用二进制编码形式替代水平井参数和裂缝参数极大地拓展了最优解搜索空间，算法复杂程度受优化对象数目的影响较小。与传统方法相比，该方法具有很大的优越性和实用价值，值得推广。

GENETIC ALGORITHM-BASED OPTIMIZING HYDRAULIC FRACTURING PARAMETERS FOR HORIZONTAL WELLS IN LOW-PERMEABILITY RESERVOIRS

If the orthogonal design method is conventionally adopted to optimize the hydraulic fracturing parameters for a horizontal well in low permeability gas reservoir, the optimization design will be difficult to cover the whole search scope for the optimum solution, and the increasing numbers of optimization object will surely add burden to simulation projects on a large increase. In order to break through this limit, a new intelligent optimum design method is established based on the genetic algorithm. What's more, with the external call program of gas reservoir numerical simulation software, program is done to automate the whole process of optimum design. The result shows that: the process of optimum design using genetic algorithm doesn't need any man made interference and reduces the workload to a large degree; the method of binary code representing the parameters of horizontal well and hydraulic fracture enlarges the search scope greatly and eliminate the effect of the number of optimization object on the complexity extent of algorithm. Comparing with the traditional method, this method is much better and more practical to be widely used on fields.