多Agent的复杂经济仿真系统构建策略  

A strategy to construct multi-Agent-based complex economic simulation systems

作　　者：曲国华[1] 张振华[2] 徐岭[3] 刘增良[4] 曲卫华[4] 张汉鹏[5] 张强[4] 

机构地区：[1]山西财经大学管理科学与工程学院,山西太原030006 [2]广东外语外贸大学经济贸易学院,广东广州510006 [3]北京石油化工学院经济管理学院,北京102600 [4]北京理工大学管理与经济学院,北京100081 [5]西南财经大学工商管理学院,四川成都610074

出　　处：《智能系统学报》2016年第2期163-171,共9页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金项目(61175122;71071018;71201089);北京市哲学社会科学规划项目(SZ201410017006);广东省哲学社科和软科学基金项目(GD12XGL14;2015A070704051;2014A030313575);教育部人文社科项目(14XJC630010)

摘　　要：为了给复杂经济仿真系统找到通用的理论模型,通过对国内Agent经济系统的研究,提出复杂经济系统微观模型(CESM)。采用智能Agent CESM和亚小CESM模拟的实现方式,界定了个人、企业、政府、国外贸易的经济行为,采用模型转换和代码生成支持计算实验。根据所提出的CESM策略,设计和实现四部门复杂经济仿真系统(FEAS),最后用经济学的观点对结果进行解释,结果表明,CESM最吸引人的地方是它通过异质性微观Agent来涌现宏观特征。系统的仿真过程表明提出的CESM策略是一种自治的、贴近复杂经济仿真系统问题研究的建模方法,可以在集成项目的工作中发挥重要的作用。To find a suitable theory model for complex economic simulation system, complex economic system microcosmic model （CESM） was proposed through the domestic economic system for research and Agent classification, analyzing CESM mechanism. An implementation approach used CESM-simulation and sub-CESM simulation based on intelligent agent and person, enterprise, govemment and foreign trade actions was defined, and the simu- lation models are generated by model transformation in order to support computational experiments. Four depart- ments complex economic simulation system FEAS designing and implementation is developed according to CESM strategy proposed. Finally, simulation results are explained from Macroeconomics view, and the results show that the most attractive characteristics of CESM is that it can analyze its macro character by research heterogeneous agents behavior. FEAS simulation testifies that CESM strategy proposed is self contained, and it plays an important role in the integration of simulation system project.

关 键 词：AGENT 复杂经济系统微观模型 结点 四部门复杂经济仿真系统 

分 类 号：F224[经济管理—国民经济] N945.13[自然科学总论—系统科学]