基于BS-PLS回归模型的木材含水率在线预测研究  

Research on Online Prediction of Moisture Capacity in Wood Based on B-spline Partial Least Square Regression Model

作　　者：刘亚兰[1] 刘德胜[2,3] 孙丽萍[2] 李志辉[2] 

机构地区：[1]东北林业大学材料科学与工程学院,黑龙江哈尔滨150040 [2]东北林业大学机电工程学院,黑龙江哈尔滨150040 [3]佳木斯大学信息电子技术学院,黑龙江佳木斯154007

出　　处：《自动化仪表》2009年第7期61-63,共3页Process Automation Instrumentation

基　　金：国家科技支撑计划基金资助项目(编号:2006BAD18B0801)

摘　　要：由于木材干燥周期较长、干燥窑内工作状态极其复杂,这使得木材含水率的检测精度受环境影响较大,尤其是在以木材含水率为基准的干燥过程中,木材含水率的在线检测精度至关重要。针对这一问题,提出了基于样条映射的偏最小二乘法的木材干燥过程非线性建模方法。该方法将采集温度、湿度等多种传感器检测值作为模型的输入,木材含水率作为模型的输出,仿真试验验证了模型具有较好的泛化能力和较高的预测精度。The detection accuracy of moisture capacity in wood is highly depends on environmental impact, because the drying cycle of wood is very long and the operation condition in drying kiln is complicate; especially in drying process based on moisture capacity in wood, so online detection accuracy of moisture is crucial. Aiming at the problem, the nonlinear modeling method of spline-based mapping of partial least square for wood drying proves is proposed. With the method, the detected values from various sensors, such as temperature and humidity, are used as the inputs of the model, and the moisture capacity in wood is the output of model. The simulation test verifies that the model has better generalization ability and offers higher prediction accuracy.

关 键 词：木材含水率 样条函数 偏最小二乘回归 数据采集 仿真 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]