Neural Network Modeling of Duplex Inlet and Outlet Ball Pulverizer System Based on Grey Relational Analysis

The direct-fired system with duplex inlet and outlet ball pulverizer is more difficult to establish the precise mathematical model because of its multivariable strongly coupling and strongly nonlinear, and the impact factors of mill output are numerous. The gray relational analysis theory used to analyze the duplex inlet and outlet ball mill system, and obtained the gray correlation degree of every instrumental variable which affect the mill output. Ignoring minor factors, the neural network model of mill system was established, and implemented the model dimensionality reduction. The two models of before and after grey relational analyzing was compared with each other. Simulation results show that the learning speed and convergence rate of dimension reduction model is better. This will lay a foundation in theory for the neural network application online in pulverizing systems.