基于相关系数法与遗传算法的啤酒酒精度近红外光谱分析

以啤酒的酒精度的快速检测为研究对象，针对采用偏最小二乘(Partial Least Squares，PLS)法建立近红外光谱预测模型时波长筛选问题，提出将相关系数法与遗传算法(Genetic Algorithms，GA)相结合提取光谱有效信息，提高预测模型的精度的方法。结果表明：该方法应用于啤酒酒精度近红外光谱检测中，吸收光谱和一阶导数光谱的预测建模的波长个数分别减少了83%、82%，预测平均相对误差分别降低了0.42%、0.64%，不仅简化、优化了模型，而且增强了预测建模型的预测能力，是一种采用PLS法建立预测模型前行之有效的降低和优选波长的方法。

Analysis of near infrared spectroscopy of beer alcohol content by correlation coefficients and genetic algorithms

In order to select wavelengths of near infrared spectroscopy in the prediction model of partial least squares regression of beer alochol content detection, correlation coefficients and genetic algorithms were used to select wavelength from near infrared spectroscopy in partial least squares regression model. The method was applied to the study of beer alcohol content with spectroscopy. The result shows that the numbers of wavelengths of original absorption spectra and first derivative spectra for developing models can be reduced by 83% and 82% by this method, while the root mean square error of prediction reduces by 0.42% and 0.64% respectively. This wavelengths selection method for PLS modeling not only simplifies and optimizes calibration model but also increases the prediction ability of calibration model. Therefore, correlation coefficients and genetic algorithms are effective and feasible methods applied in developing mutlivariate calibration model based on partial least squares regression.