用模糊神经网络构造煤粉流量模型

COAL POWDER FLOWRATE MODELING USING FUZZY NEURAL NETWORK

  • 摘要: 提出了用模糊神经网络的方法来建立煤粉流量模型.针对煤粉喷吹系统,模糊神经网络可以把它的工作空间分成若干个模糊区间,在每个区间中,用一局部模型来代表此系统;此方法克服了神经网络无法解释其联接权物理意义的不足.利用现场采集的数据进行实验,发现此方法不仅提高了准确度,而且可充分运用工作经验、专家知识,具有很大的工程应用价值.

     

    Abstract: This paper proposes a method of coal powder flow rate modeling using fuzzy neural network.For puffing system of furnace,the input space of the system is divided into several fuzzy operating regions,in which alocal model can represent the system; the method avoids the short coming that neural network cannot interpret the physical meaning of its associating weights. Making use of data which were sampled in practice,the accuracy is improved using this method,and work experience and expert knowledg eare used sufficiently,which has great engineering application value.

     

/

返回文章
返回