化学计量学建模软件的开发及应用

本文利用数据库技术,针对Y=F(X)数学模型,编写集模型管理、数据样本管理和建模、预报为一体的通用性处理软件。软件自动适应不同分析体系中,自变量数和目标函数的变化,并支持在模型中对每个自变量进行定义,及其物理意义的显示。软件中提供了MLR、PCR、PLS、神经网络和遗传算法-PLS联用5种建模方法,并将交叉验证方法嵌入PCR、PLS方法中,用交叉验证法检验样本中的奇异样本点。利用模拟数据对软件进行了检验。

The development of model building software in chemometrics

In this paper, a more generalized modeling building software, which integrates model management, sample management and model building together is developed. It is mainly designed for solving Y=F(X) modeling problem in a more general thought through database technology. This software can automatically adjust to the changes in number of variable, target function and samples. It also supports to add, delete each variable and functions in the model dynamically, and to display the physical meaning of each variable respectively. In this software, five model-building techniques such as MLR, PCR, PLS, Artificial Neural Network and Genetic Algo-rithms-PLS are employed. In addition, Cross Validation is added into PCR and PLS in order to select the effective principal components and also used to test the abnormal sample point. At last simulated data is employed to testify the application of the software.