摘要
清除工业废气中低浓度挥发性有机物(VOCS) 的流向变换催化燃烧反应器的床层温度依时变化,为了实现实时预测和控制,用动态RBF(Radial Basis Function)神经网络建立了反应器床层瞬态温度分布的预测模型。着重讨论了动态RBF神经网络的基本结构,依据RBF网络线性输出的特点,给出了预测模型参数的在线修正方法。仿真结果与中试装置现场数据的对照表明,所建立的模型简单、精度高,能满足控制要求。
The temperature profile for a reverse flow reactor with catalytic combustion of air contaminated with volatile organic compounds (VOCs) varies with real time. In order to predict and control, a real-time prognosticate model of temperature profile for a reverse flow reactor was built based on dynamic RBF (Radial Basis Function) neural networks. An on-line correcting method of model parameters was proposed based on RBF networks' linear outputs, with special emphasis on constructing dynamic model. The simulation result has proved that the model presented is simple, highly accurate and can meet control demand.
出处
《高校化学工程学报》
EI
CAS
CSCD
北大核心
2003年第5期559-563,共5页
Journal of Chemical Engineering of Chinese Universities
基金
国家自然科学基金资助项目(20076002)。
关键词
挥发性有机物
流向变换
催化燃烧
动态系统
RBF神经网络
在线参数校正
Catalysis
Chemical reactors
Neural networks
Radial basis function networks
Reverse combustion
Temperature distribution
