|
| 基于改进的GRNN的固体氧化物燃料电池辨识模型研究 |
| Identification model for SOFC based on improved GRNN |
| 投稿时间：2013-07-03 |
| DOI： |
| 中文关键词: 固体氧化物燃料电池 广义回归神经网络 粒子群算法 辨识建模 |
| 英文关键词:solid oxide fuel cell (SOFC) generalized regression neural networks (GRNN) particle swarm optimization (PSO) modeling identification |
| 基金项目:国家自然科学基金资助项目（41175033） |
|
| 摘要点击次数: 2875 |
| 全文下载次数: 2357 |
| 中文摘要: |
| 固体氧化物燃料电池（SOFC）作为一种新的能源形式，日益受到重视.针对SOFC 系统过于复杂，现有的理论电压模型存在明显不足的特点，绕开了SOFC 的内部复杂性，利用经过粒子群算法（PSO）优化的广义回归神经网络（GRNN ） 对SOFC 系统进行辨识建模.以氢气流速为神经网络辨识模型的输入量，电流/电压为输出量，建立SOFC 在不同氢气流速下的电池电流/电压动态响应模型.仿真结果表明所建模型能基本表示出SOFC系统的电流/电压的动态响应，说明利用GRNN建模的有效性，所建模型精度也较高. |
| 英文摘要: |
| As a new form of energy,the solid oxide fuel cell (SOFC) is getting more and more attention.For the SOFC system is too complex,the existing theoretic voltage model is obviously insufficient to describe the system.This paper tries to avoid the internal complexity of SOFC and set up a SOFC model using a generalized regression neural networks (GRNN) based on a particle swarm optimization (PSO).Taking the flow rate of hydrogen for SOFC as the input and the voltage and current as the output of the neural network model,a cell voltage and current identification model of SOFC is established.The simulation results show that the model can basically show the current/voltage dynamic response of the SOFC system and therefore verifies the effectiveness of GRNN modeling and high accuracy of the model. |
| HTML 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
