发动机润滑油含水量近红外特征波段选取

应用近红外光谱技术结合间隔偏最小二乘法,探讨发动机润滑油中含水率检测的特征波段及检测方法。采用间隔偏最小二乘法对发动机润滑油含水量的近红外光谱特征波段进行选择,结合偏最小二乘回归建模方法对所选波段进行验证。结果表明:选用7 432～6 321 cm-1和5 583～4 846 cm-1 2个波段建模可以得到与全谱一致的预测结果,从而实现了模型的简化,为润滑油中含水率近红外光谱分析提供理论依据。

Selection of Near-infrared Characteristic Wavelength Bands for Water Content of Engine Oil

The characteristic wavelength bands and detection methods of the water content in engine oil were probed by applying near-infrared(NIR)spectroscopy combined with interval partial least square method.The interval partial least square(iPLS)method was used to select the NIR characteristic wavelength bands of water content in engine oil,the selected wavelength bands were verified with partial least square regression(PLSR)modeling method.The results show that modeling with 7 432~6 321 cm-1 and 5 583~4 846 cm-1 wavelength bands can get the same results as the whole wavelength bands,so the model can be simplified.This research provides the theoretical basis for the analysis of water content in engine oil with NIR spectroscopy.