基于知识的仿真模型验证方法的研究  

Research of Knowledge-based Method to Simulation Model Validation

作　　者：李亚男[1] 刘飞[1] 杨明[1] 

机构地区：[1]哈尔滨工业大学控制与仿真中心,哈尔滨150001

出　　处：《系统仿真学报》2009年第8期2277-2280,共4页Journal of System Simulation

基　　金：国家自然科学基金重点项目(60434010);黑龙江省杰出青年科学基金(JC200606).

摘　　要：针对仿真模型验证过程中缺少参考数据的情况,提出一种基于知识的仿真模型验证方法。首先将仿真系统的行为分割成各种基本行为模式,然后提取出行为特征,最后将已提取出的仿真系统行为特征与知识库中的专家知识进行比较,实现仿真模型的验证。实验表明,该方法能够较好的解决仿真模型验证中缺少参考数据的问题。Aiming at the situation of lacking of reference data in simulation model validation process, a knowledge-based method was proposed. First the system behavior was divided into segments according to behavior patterns. Then behavior featureswould be extracted. After that, compared features extracted with expert knowledge in the knowledge base, simulation model was validated. Experiments show that the method can solve the problem of lacking of reference data in simulation model validation effectively.

关 键 词：模型验证 专家知识 基本行为模式 行为特征 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]