上海交通大学学报

上海交通大学学报（自然版）

• 自动化技术、计算机技术 • 上一篇    下一篇

基于演化参数辨识的流体网络建模

倪何a,b，程刚a，孙丰瑞b
  

  1. (海军工程大学 a. 装备仿真研究所；b. 动力工程系， 武汉 430033)
  • 收稿日期:2010-05-14 修回日期:1900-01-01 出版日期:2011-02-28 发布日期:2011-02-28

Modeling of Fluid Network Using EABased Parameter Identification

NI Hea,b,CHENG Ganga,SUN Fengruib
  

  1. (a. Research Institute of Equipment Simulation Technique; b. Department of
    Power Engineering, Naval University of Engineering, Wuhan 430033, China)
  • Received:2010-05-14 Revised:1900-01-01 Online:2011-02-28 Published:2011-02-28

摘要： 应用演化计算理论研究了热动力系统中热工流体网络的仿真建模问题.从基本物理规律出发，建立了热工流体网络的通用仿真数学模型；采用演化算法对流体网络的参数进行辨识，避免了以往采用经验值或者网络在理想状态下的试验数据作为建模参数所造成的偏差，在提高计算精度的同时使模型的动态特性更加符合实际.将该方法应用于某蒸汽动力装置新蒸汽系统的仿真建模中，通过仿真结果和实测值的对比，证明采用演化算法辨识后的流体网络模型具有较高的计算精度，能够很好地反映网络对象的特性，在工程上是实用的.

关键词: 流体网络, 演化计算, 参数辨识, 数学模型

Abstract: Using evolutionary arithmetic this paper studied the modeling of thermofluid network which existed in thermopower system rifely. First, a general mathematic modeling of fluid network was established based on the basic physical laws, and then evolutionary arithmetic was adopted to identify the parameters of network. This method avoids the calculational errors of conventional methods that using experiential values or testing data, enhances the computational precision of model and makes the behaviors of model accordant to the real object. This method was used to simulate the steam network in a certain type marine power plant, and the outputs of model against the real values were measured. The results show that the parameters identified model has preferable accuracy, and the EAbased modeling method is useful.

中图分类号: