Flowrate Measurement of Coal Powder Using Fuzzy Neural Network

摘要

The paper proposes a method to measure coal powder flowrate with fuzzy neural network. In general, industry process have a character of nonlinear, the input space of a nonlinear system is initially divided into a number of fuzzy operating regions in which local models are able to represent the system. The complete system model output is obtained through the conjunction of the outputs of the local models. The fuzzy neural network approach to nonlinear process modeling provides a way of openin up the purely 'black box' approach normally seen in the neural network applications. Process knowledge is used to identify appropriate local operating regions and as an aid to initializing the network structure. Fuzzy neural network models are also easier to interpret than conventional neural network models. The weights in a trained fuzzy network models can be interpreted in terms of process information such as the partition of operating regions and the process gain and time constatn in each region. The technique has been applied to coal powder flowrate. The fuzzy network technique is proven to be valid by experiments. Indexing term :Neural network, Fuzzy model, Membership function, Coal powder flowrate.