山东科学 ›› 2016, Vol. 29 ›› Issue (1): 1-6.doi: 10.3976/j.issn.1002-4026.2016.01.001

• 海洋科技与装备 •    下一篇

固体氧化物燃料电池系统仿真分析与控制

刘东彦,张颖颖,吴丙伟,张颖,胡云川   

  1. 山东省科学院海洋仪器仪表研究所,山东 青岛 266001
  • 收稿日期:2015-11-27 出版日期:2016-02-20 发布日期:2016-02-20
  • 作者简介:刘东彦(1983-),女,助理研究员,研究方向为复杂系统建模与控制。Email:ldynuaa2008@163.com
  • 基金资助:

    山东省重大科技专项(2015ZDXX0602A02)

Simulation analysis and control of solid oxide fuel cell system

LIU Dongyan, ZHANG Yingying, WU Bingwei, ZHANG Ying, HU yunchuan   

  1. Institute of Oceanographic Instrumentation, Shandong Academy of Sciences,Qingdao 266001, China
  • Received:2015-11-27 Online:2016-02-20 Published:2016-02-20

摘要:

固体氧化物燃料电池(SOFC)系统是一个非线性、多变量和强耦合的系统,很难用传统的建模方法来建立。本文基于BP神经网络的方法,利用MATLAB/Simulink平台构建SOFC系统模型,并在该模型的基础上增加PID控制,实现了闭环控制系统的分析。实验结果表明,该模型预测精度高,由预测模型得出的温度数据与实际数据的绝对误差为0.011%,增加的PID控制算法具有很强的抗干扰能力。

关键词: 热管理, 固体氧化物燃料电池, BP神经网络, PID控制

Abstract:

Solid oxide fuel cell (SOFC) system is a nonlinear, multivariable and strong coupling system, so its model is very difficult to be constructed with traditional modeling method. We establish a SOFC system model with BP neural network and MATLAB/Simulink platform.We further increase PID control based on the model and closed loop control system analysis. Experimental results show that the model has high prediction accuracy, and the relative error between temperature data from the model and actual data is 0.011%.The increased PID control algorithm has stronger antiinterference capability.

Key words: thermal management, PID control, solid oxide fuel cell, BP neural network

中图分类号: 

  • TM 911.4