摘要
Applying glue system of medium density fiberboard is a typical nonlinear time-delay system.In this paper,based on the simple introduction of the principle of the inverse model identification,using the function approximation ability of support vector machine,the inverse model of applying glue system is identified with the direct inverse model identification method based on least squares support vector regression,and the simulation results indicate that the inverse model of applying glue system built through the direct inverse model identification method on support vector machine has high accuracy and the feasibility and validity of the method is also proved.
Applying glue system of medium density fiberboard is a typical nonlinear time-delay system.In this paper,based on the simple introduction of the principle of the inverse model identification,using the function approximation ability of support vector machine,the inverse model of applying glue system is identified with the direct inverse model identification method based on least squares support vector regression,and the simulation results indicate that the inverse model of applying glue system built through the direct inverse model identification method on support vector machine has high accuracy and the feasibility and validity of the method is also proved.
出处
《林业科学》
EI
CAS
CSCD
北大核心
2010年第2期171-174,共4页
Scientia Silvae Sinicae
基金
黑龙江省科技攻关计划重点项目(GB06A505)
哈尔滨市科技攻关计划项目(2006AA1BG067)
关键词
逆模型
施胶系统
支持向量机
建模
inverse model
applying glue system
support vector machine
modeling
