The Application Research Of Mulitivariate Regression Model In Wood Drying Quality Prediction

Wood is one of the most important constructional materials, and many adornment such as furniture are welcomed to people.All of that lead to the increase of demand and the reduce of forest. How to reduce the excessive use of wood, use the limited resources effectively and improve the quality of wood products has aroused widespread concern by the government of many countries. China is a country which is short of forest resources, facing the huge need of wood and the environmental pressure, so how to improve the proformance of the wood and utilization has become an important research subject to the scientific workers of our country.Wood drying is the important technical measure to improve the physical mechanical properties of wood, reduce the loss of wood dropping, improve the utilization of wood and ensure wood products qualities. Create a right model of drying has a great significance to guide wood drying and improve the drying qualities. The stress and moisture are the important index of drying quality, and the environment influencing factors are multiplex too, so creat the model between the drying quality and the environment influencing factors will have a great significancy to improve the drying quality and control level.In this paper, according to the various indicators and the multivariate environment parameter, we can creat a multivariate model between them. To the multicollinearity among the various, we use PAC and PLS to improve the model. Compare the result of the three models, we get the best one.