The Research Of Medium Density Fibreboard's Glue System Modeling And Simulation

With the economic development and population increasing, the demand for wood is constantly increasing. The man-made board can use wood resource efficiency, it is main industry in the wood domain because it has high increment and high-tech content. Medium density fibreboard's development is the fastest in the man-made board industry, with its development, our country has been the biggest country of production medium density fibreboard in the world.In the process of medium density fibreboard's production, applying glue is the important working procedure, it has big influence on the quality and cost production of medium density fibreboard. The task of applying glue is to apply gule for fibre according to its change, in order to glue quantity and fibre quantity keep a reason proportion.Glue system is a nonlinear multivariable and coupling complicated system . It's difficult to build up an exact model using classical system identification method, but the system identification based on Neural Networks can directly learn the input and output data of the system and make the object function to get minimal value, then generalize the relation which hide in the input and output data of the system, the description is the model of system. In this thesis, the glue system model are built based on Neural Networks.According to the characteristic of medium density fibreboard's glue system, built up the model of medium density fibreboard's glue system based on the static multi-layer feedforward Neural Networks. This model provide the theoretics gist to carry out online control in the procedure of medium density fibreboard's glue system.The structures of the Neural Networks and the learning algorithms were given, in addition, we verify the accuracy of the model, the simulation results show that the model is effective and reliable; at the same time, we compared with results between with the classical method and Neural Networks method, the results show that the later is better. Neural Networks is the physics implementation for actual system, and can be used in on-line control. The methods of identification can be popularized to other similar systems.