应用红外光声光谱技术结合区间、组合区间偏最小二乘,建立了油菜籽含氮量和含油量的校正模型。结果表明,红外光声光谱技术可以应用于油菜籽品质的快速测定。相对于全谱偏最小二乘建模,区间、组合区间偏最小二乘的采用筛选出了含氮量和含油量的相关波段,使模型简化,并提高了模型预测精度。 A calibration model of nitrogen content and oil content in rapeseed was established by using infrared photoacoustic spectroscopy in combination with interval and partial least squares. The results show that infrared photoacoustic spectroscopy can be applied to the rapid determination of rapeseed quality. Relative to full-spectrum partial least squares modeling, the interval and combination interval partial least squares are used to filter out the relevant bands of nitrogen and oil content, which simplifies the model and improves the prediction accuracy of the model.