化工学报 ›› 2016, Vol. 67 ›› Issue (11): 4599-4608.DOI: 10.11949/j.issn.0438-1157.20160721

• 流体力学与传递现象 • 上一篇    下一篇

余热锅炉单相受热面动态建模与模型参数优化

李金波, 程林   

  1. 山东大学热科学与工程研究中心, 山东 济南 250061
  • 收稿日期:2016-05-27 修回日期:2016-07-12 出版日期:2016-11-05 发布日期:2016-11-05
  • 通讯作者: 程林,cheng@sdu.edu.cn
  • 基金资助:

    国家重点基础研究发展计划项目(2013CB228305)。

Dynamic modeling and parameter optimization of single phase heating surface of heat recovery steam generator

LI Jinbo, CHENG Lin   

  1. Center of Thermal Science and Technology, Shandong University, Jinan 250061, Shandong, China
  • Received:2016-05-27 Revised:2016-07-12 Online:2016-11-05 Published:2016-11-05
  • Supported by:

    supported by the National Basic Research Program of China (2013CB228305).

摘要:

作为余热利用环节中最重要的部分,余热锅炉的启动、变工况运行和停机特性直接决定锅炉的寿命及效率。基于工质热力学性质和质量、动量及能量守恒方程,以Matlab/Simulink为平台,构建了余热锅炉单相受热面的动态仿真模型。结合某水泥厂自主设计的直流余热锅炉实验数据,基于遗传算法和粒子群算法,对动态模型进行了参数优化。结果表明,经过优化后,余热锅炉动态模型与实验数据匹配程度高,模拟与实验结果的误差为0.93%~4.39%。因此,本文所建立的单相受热面变工况动态模型可以准确反映余热锅炉受热面动态特性。两种算法的对比表明,粒子群算法适应度函数收敛更优;在收敛迭代次数上,粒子群算法在54~64代达到收敛,遗传算法在93代后达到收敛。粒子群算法在参数优化方面优于遗传算法。

关键词: 动态仿真, 参数优化, 实验验证, 粒子群算法, 遗传算法

Abstract:

Heat recovery steam generator (HRSG) is the most important part of the waste heat utilization, and the start-up, shutdown and off-design operation of the HRSG directly determines its life and efficiency. Based on the thermodynamic properties and mass, momentum and energy conservation equations, and taking Matlab/Simulink as the research platform, a dynamic simulation model of the single phase heat transfer surface of HRSG is built in this paper. Combined with the experimental data of the HRSG designed by research group in a cement plant, the parameter optimization of the dynamic model is carried out based on genetic algorithm and particle swarm optimization algorithm. The results show that after optimization, the dynamic model of the waste heat boiler is matched with the experimental data, and the error of the model is 0.93%-4.39%. The dynamic model can be used to simulate the temperature change of the heat transfer surface under different working conditions. And through the comparison of the two algorithms, it shows that particle swarm optimization algorithm has obvious advantages in parameter optimization. The fitness function convergence value is better, and in the convergence iteration, it finishes between the 54 and 64 generation. The genetic algorithm achieves convergence after the 93 generation.

Key words: dynamic simulation, parameter optimization, experimental validation, particle swarm optimization, genetic algorithm

中图分类号: