摘要
为了解决公共政策优化仿真时考察点不全面及耗费时间大的问题,提出了基于遗传Data Farming的公共政策优化仿真模型。所提出的模型中,Data Farming被用于全面地考察输入参数空间中各输入参数水平对应的输出响应,通过把仿真过程的输出过程看作适应度的计算,遗传算法被用于优化仿真过程,提高优化效率。分析了公共政策作用效果的影响因素,并介绍了传统的Data Farming技术,然后从仿真目标、模型框架、算法等三方面介绍了所提出的优化仿真模型。最后进行了实验,证明了所提出模型的有效性。
In order to solve the problems of incomprehensive investigation and large time-consuming, a genetic data farming optimization simulation model for public policies was proposed. In the proposed model, data farming was used for a comprehensive study of output response corresponding to the levels in input parameters space. And by taking simulation process as fitness calculation, the genetic optimization algorithm was used to optimizing simulation process and improving simulation efficiency. Factors affecting effects of public policies were analyzed, and traditional data farming was introduced. And then, the proposed optimization simulation model was introduced from three aspects of the simulation goal, framework and the algorithm. At last, experiments were conducted to verify the proposed model.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2008年第21期5963-5966,5988,共5页
Journal of System Simulation
基金
国家自然科学基金(70471074)
广东省科技计划项目资助(2004B36001051)
