This investigation reports the rheological behavior of ZnO-MWCNT/10W40 nanofluid at the temperature range of 5 °C–45 °C, in the volume fraction (VF) of 0.05%–1%, and in the shear rate (SR) of 666.5 (1/s) to 11,997 (1/s). ZnO and multiwall carbon nanotube were mixed by a volume ratio of 60:40 and dispersed into 10W40 lubricant by a two-phase mixing method. In order to use the sets data of the current experimental research for future goals, a mathematical relation was proposed to calculate the NF viscosity. The results show a direct relationship between the dynamic viscosity and the VF of nanoparticles and the relation inversely to temperature and SR. The performed sensitivity analysis also showed that the viscosity is dependent on temperature in comparison with other independent variables. In fact, based on sensitivity analysis nanoparticles, VF and SR introduced in the next ranks of importance, respectively. In the ANN design, the R² value is equal to 0.991. For more precise estimation of the hybrid NF dynamic viscosity, an optimum artificial neural network (ANN) modeling selected from 400 different structures was employed. The comparison of the experimental results with the ANN ones reported an appropriate accuracy of ANN results.

本研究报告了 ZnO-MWCNT/10W40 纳米流体在 5 °C–45 °C 的温度范围、0.05%–1% 的体积分数 (VF) 和 666.5 的剪切速率 (SR) 下的流变行为(1/s) 到 11,997 (1/s)。ZnO和多壁碳纳米管按60:40的体积比混合，通过两相混合法分散到10W40润滑剂中。为了将当前实验研究的集合数据用于未来目标，提出了一种数学关系来计算NF粘度。结果表明，动态粘度与纳米颗粒的 VF 之间存在直接关系，而与温度和 SR 呈负相关。进行的敏感性分析还表明，与其他自变量相比，粘度取决于温度。事实上，基于灵敏度分析纳米粒子，VF 和 SR 分别在重要性的下一个等级中引入。在 ANN 设计中，R ²值等于0.991。为了更精确地估计混合 NF 动态粘度，采用了从 400 种不同结构中选择的最佳人工神经网络 (ANN) 模型。实验结果与人工神经网络的比较报告了人工神经网络结果的适当准确性。