论文部分内容阅读
润滑油的润滑性能与抗氧剂的含量密切相关,快速准确地检测在用润滑油的抗氧剂损耗程度是非常重要的。本文研究傅里叶红外光谱结合BP神经网络快速测定合成润滑油中胺类抗氧剂的方法。对神经网络参数和波长范围进行选择,当隐含层神经元个数为13,学习速率、动量因子均为0.2,选择光谱波长范围(3420~3330)cm~(-1)和(1650~1510)cm~(-1)时,获得最佳的抗氧剂定量分析模型。模型的校正偏差SEC为0.052,相关系数R~2达到0.986,预测集的标准偏差SEP为0.059,与校正集结果接近。研究结果表明,BP神经网络结合红外光谱可实现润滑油抗氧剂的快速、准确分析。
The lubricity of lubricants is closely related to the content of antioxidants. It is very important to detect the degree of antioxidant depletion in lubricants quickly and accurately. In this paper, the rapid determination of amine antioxidants in synthetic lubricating oil by Fourier transform infrared spectroscopy combined with BP neural network is studied. When the number of neurons in hidden layer is 13, the learning rate and momentum factor are both 0.2, the spectral range (3420 ~ 3330) cm ~ (-1) and (1650 ~ 1510) ) cm ~ (-1), the best antioxidant quantitative analysis model was obtained. The model has a corrected deviation SEC of 0.052, a correlation coefficient of R ~ 2 of 0.986, and a standard deviation SEP of 0.059, which is close to the result of the calibration set. The results show that the combination of BP neural network and infrared spectroscopy can quickly and accurately analyze the antioxidant of lubricating oil.