铜颗粒污染物对变压器油运动黏度的影响

变压器油中铜颗粒污染物的存在极易影响油液的理化性能，甚至导致重大生产事故，因而根据ISO4406：1999标准，配制了24组不同污染度的含铜颗粒油样并通过实验测试了变压器油样的运动黏度性能（40℃），获得了实验范围内铜颗粒污染度对油液运动黏度的影响规律；然后根据测试数据采用偏最小二乘（PLS）法和支持向量机（SVM）法建立了油样中不同粒径、含量的铜颗粒与油液运动黏度的数学模型，探讨了不同粒径、含量的铜颗粒污染物对油液运动黏度的影响规律。结果表明：随着铜颗粒污染度的增加，油液的运动黏度减小；铜颗粒污染物粒径在15～25 μm范围内对油液运动黏度的影响较大，且粒径越大，油液运动黏度也越大；所建立SVM模型对验证集进行预测的相关系数和均方根误差（RMSE）分别是0.962 6、4.597×10^-5。为消除实验误差，减少人为因素的影响比较准确地掌握铜颗粒污染物对油液运动黏度的影响提供了新的途径。

Influence of copper particle pollution on transformer oil kinematic viscosity

Copper particle pollution in the transformer oil is easy to affect the physical and chemical properties of the oil, and even leads to major accidents. On the basis of ISO4406：1999, 24 groups of samples of different pollution particles containing copper were made up, and the kinematic viscosity (40℃) of the samples was tested; the influence rules were achieved between copper particle pollution and oil kinematic viscosity. A simulation model was established between the different particle sizes, content copper particles in the sample and the kinematic viscosity of oil samples using partial least squares (PLS) method and support vector machine (SVM) method. The influence laws of the different particle sizes and content copper particle pollution on the oil kinematic viscosity were investigated. The results show that the oil kinematic viscosity decreases with the increase of copper particles' pollution degree. Copper particle pollution within the particle size range of 15-25 μm has obvious influence on the oil kinematic viscosity, and the oil kinematic viscosity increases with the increase of particle size. The correlation coefficient and root mean square error (RMSE) of the established SVM model are obtained to forecast the validation set, which are 0.962 6 and 4.597×10^-5, respectively. This provides a new way to eliminate the experimental error, reduce the influence of artificial factors, and know the copper particle pollution impact on oil kinematic viscosity more accurately.