摘要
本文以大豆样品为例 ,研究不同的样品温度对于脂肪含量近红外定量分析数学模型的影响 ,并且通过在回归建模过程中包含温度影响因子的混合建模方法来减小温度对于模型的影响 ,使得模型对于温度的适应性增强。实验结果表明 。
In this article we have researched the influence of different sample temperatures against the mathematical model of NIR quantitative analysis to fat content in soybean.In order to diminish the influence, a temperature-mixed model was built,which took variable temperature factor into account in the process of modeling.The temperature-mixed model is more flexible and adaptable to different sample temperatures.The result of experiments shows that the ability to predictio of the temperature-mixed model is much better than that of the single temperature model.
出处
《现代仪器》
2004年第5期29-31,共3页
Modern Instruments
基金
国家高技术研究发展计划 ( 863计划 )课题--稻麦品质遥感监测与预报技术研究 ( 2 0 0 2AA2 43 0 11)
国家高技术研究发展计划 ( 863计划 )课题---农畜产品品质快速无损检测技术 ( 2 0 0 2AA2 480 5 1-2 )的资助
关键词
数学模型
温度
近红外定量分析法
建模方法
大豆
脂肪
Near infrared spectroscopy Sample temperature Mathematical model Quantitative analysis
