在傳統使用有限元素作工程問題分析時，當問題越趨複雜時，需耦合許多物理量來使模擬更趨近於真實情況。但是當耦合許多物理量時，常常會耗費大量的計算時間與計算資源。因此若能利用機器學習來輔助，將能使耗費時間縮短，節省計算時間與計算資源的消耗。
本研究使用修正型的雷諾方程式並加入彈性變形的影響，再給定負載後依序求出各軸承性能參數，並將不同細長比及負載的軸承依序對壓力、膜厚和變形量建立資料集，然後使用不同機器學習模型做訓練，在經過最佳化的調校後，會評估各模型的準確度與花費時間，藉此探討使用機器學習的可行性。
結果表明極限梯度法XGBoost和隨機森林樹Random forest的準確度平均都有9成以上，並且訓練集的大小會有一定程度的影響，基本上資料集越多越完整，準確度會越準確。其中又以極限梯度法XGBoost和隨機森林樹Random forest不管在計算時間還是準確上，表現皆更為出色。因此在軸承特性的評估上，極限梯度法和隨機森林樹是一個可行的方法。

When using traditional finite element methods for engineering problem analysis, increasing problem complexity requires coupling multiple physical quantities to make simulations more realistic. However, this coupling often leads to significant computation time and resource consumption. Utilizing machine learning as a supplement can help reduce time consumption, saving both computation time and resources.
This study employs a modified Reynolds equation, incorporating the effects of elastic deformation, and calculates bearing performance parameters under applied loads. Datasets were generated for various slenderness ratios and loads, focusing on pressure, film thickness, and deformation. These datasets were then used to train different machine learning models. After optimizing the models, their accuracy and computation times were evaluated to assess the feasibility of using machine learning.
The results show that both Extreme Gradient Boosting (XGBoost) and Random Forest achieved average accuracies above 90%. Additionally, the size of the training dataset has a certain impact—the larger and more complete the dataset, the more accurate the predictions. Among the models, XGBoost and Random Forest outperformed others in terms of both computational time and accuracy. Therefore, XGBoost and Random Forest are viable methods for evaluating bearing characteristics.

[1] G.-A. Hirn, "Sur les principaux phénomènes que présentent les frottements médiats, et sur les diverses manières de déterminer la valeur mécanique des matières employées au graissage des machines: mémoire présenté par MG-Ad. Hirn, dans la séance du 28 Juin 1854," Bulletin de la société industrielle de Mulhouse, vol. 26, pp. 188-237, 1858.
[2] Petrov, N. P. , “Friction in Machines and the Effect of the Lubricant”, Inzh. Zh., St. Petersburg, 1883.
[3] Tower B., “First Report on Friction Experiments”, Proceedings of the Institution of Mechanical Engineers. 1884;35(1):29-35, 1883.
[4] Reynolds O., “IV. On the theory of lubrication and its application to Mr. Beauchamp tower’s experiments, including an experimental determination of the viscosity of olive oil”, Philosophical transactions of the Royal Society of London, 177, 157-234, 1886.
[5] Dowson, D., & Taylor, C. M., “Cavitation in bearings”. Annual Review of Fluid Mechanics, 11(1), 35-65, 1979.
[6] Wang, N., & Chang, C., “An application of Newton's method to the lubrication analysis of air-lubricated bearings”, Tribology transactions, 42(2), 419-424, 1999.
[7] Nicoletti, R., & Santos, I. F., “Linear and non-linear control techniques applied to actively lubricated journal bearings”, Journal of Sound and Vibration, 260(5), 927-947, 2003.
[8] Gertzos, K. P., Nikolakopoulos, P. G., & Papadopoulos, C. A., “CFD analysis of journal bearing hydrodynamic lubrication by Bingham lubricant”, Tribology international, 41(12), 1190-1204, 2008.
[9] Higginson, G. R., “The theoretical effects of elastic deformation of the bearing liner on journal bearing performance”, In Proceedings of the Institution of Mechanical Engineers, Conference Proceedings Vol. 180, No. 2, pp. 31-38, 1965.
[10] Mokhiamer, U. M., Crosby, W. A., & El-Gamal, H. A., “A study of a journal bearing lubricated by fluids with couple stress considering the elasticity of the liner. Wear”, 224(2), 194-201, 1999.
[11] Ebrat, O., Mourelatos, Z. P., Vlahopoulos, N., & Vaidyanathan, K., “Calculation of journal bearing dynamic characteristics including journal misalignment and bearing structural deformation”, Tribology Transactions, 47(1), 94-102, 2004.
[12] Osman, T. A., “Effect of lubricant non-Newtonian behavior and elastic deformation on the dynamic performance of finite journal plastic bearings”, Tribology Letters, 17, 31-40, 2004.
[13] Saber, E., & El-Gamal, H. A., “Effect of elastic deformation of the journal bearing shell on its dynamic stability”, In IUTAM Symposium on Elastohydrodynamics and Micro-elastohydrodynamics, pp. 121-131, 2006.
[14] Sun, J., Zhu, X., Zhang, L., Wang, X., Wang, C., Wang, H., & Zhao, X., “Effect of surface roughness, viscosity-pressure relationship and elastic deformation on lubrication performance of misaligned journal bearings”, Industrial Lubrication and Tribology, 66(3), 337-345, 2014.
[15] Wang, Y., Yin, Z., Jiang, D., Gao, G., & Zhang, X., “Study of the lubrication performance of water-lubricated journal bearings with CFD and FSI method”, Industrial Lubrication and Tribology, 68(3), 341-348, 2016.
[16] Chetti, B., “Combined effects of turbulence and elastic deformation on the performance of a journal bearing lubricated with a couple stress fluid” Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, 232(12), 1597-1603, 2018.
[17] Gong, J., Jin, Y., Liu, Z., Jiang, H., & Xiao, M., “Study on influencing factors of lubrication performance of water-lubricated micro-groove bearing”, Tribology International, 129, 390-397, 2019.
[18] Malki, M., Larbi, S., Boubendir, S. A., Hammoudi, D., & Bennacer, R., “Elastic deformation effects on self-lubricating journal bearings using pseudo-plastic lubricants”, Tribology Online, 16(4), 299-308, 2021.
[19] Song, X., & Wu, W., “The effect of journal misalignment on the lubrication performance of partially textured journal bearing with elastic deformation” In Journal of Physics: Conference Series, Vol. 1549, No. 3, p. 032070, 2020.
[20] Turing, A. M., “On computable numbers, with an application to the Entscheidungsproblem”, J. of Math, 58(345-363), 5, 1936.
[21] McCulloch, W. S., & Pitts, W., “A logical calculus of the ideas immanent in nervous activity”, The bulletin of mathematical biophysics, 5, 115-133, 1943.
[22] Draper, N. R., & Smith, H., “Applied regression analysis”, John Wiley & Sons, Vol. 326, 1998.
[23] Cover, T., & Hart, P., “Nearest neighbor pattern classification”, IEEE transactions on information theory, 13(1), 21-27, 1967
[24] Cortes, C., & Vapnik, V., “Support-vector networks”. Machine learning, 20, 273-297, 1995.
[25] Rumelhart, D. E., Hinton, G. E., & Williams, R. J., “Learning representations by back-propagating errors” nature, 323(6088), 533-536., 1986.
[26] Breiman, L., “Bagging predictors”, Machine learning, 24, 123-140, 1996.
[27] Breiman, L., “Random forests”, Machine learning, 45, 5-32, 2001.
[28] Hinton, G. E., & Salakhutdinov, R. R., “Reducing the dimensionality of data with neural networks”, science, 313(5786), 504-507, 2006.
[29] Tallian T E., “Tribological design decisions using computerized databases”, J Tribol 109(3): 381–386, 1987.
[30] Hirani, H., Rao, T. V. V. L. N., Athre, K., & Biswas, S., “Rapid performance evaluation of journal bearings”, Tribology international, 30(11), 825-834, 1997.
[31] Jones, S. P., Jansen, R., & Fusaro, R. L., “Preliminary investigation of neural network techniques to predict tribological properties”, Tribology Transactions, 40(2), 312-320, 1997.
[32] Karkoub, M., & Elkamel, A., “Modelling pressure distribution in a rectangular gas bearing using neural networks”, Tribology International, 30(2), 139-150, 1997
[33] Li, B., Chow, M. Y., Tipsuwan, Y., & Hung, J. C., “Neural-network-based motor rolling bearing fault diagnosis”, IEEE transactions on industrial electronics, 47(5), 1060-1069, 2000.
[34] Sinanoğlu, C., Kurban, A. O., & Yildirim, Ş. A. H. İ. N., “Analysis of pressure variations on journal bearing system using artificial neural network”, Industrial Lubrication and Tribology, 56(2), 74-87, 2004.
[35] Hirani, H., “Multiobjective optimization of journal bearing using mass conserving and genetic algorithms”. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, 219(3), 235-248, 2005.
[36] Kim, Y. H., Yang, B. S., & Tan, A. C., “Bearing parameter identification of rotor–bearing system using clustering-based hybrid evolutionary algorithm”. Structural and Multidisciplinary Optimization, 33, 493-506, 2007.
[37] Durak, E., Salman, Ö., & Kurbanoğlu, C., “Analysis of effects of oil additive into friction coefficient variations on journal bearing using artificial neural network”, Industrial Lubrication and Tribology, 60(6), 309-316, 2008.
[38] T. Kolodziejczyk, “Artificial intelligence as efficient technique for ball bearing fretting wear damage prediction”, Wear , 268 309–315, 2010.
[39] Navthar, R. R., & Halegowda, D. N., “Analysis of oil film thickness in hydrodynamic journal bearing using artificial neural networks”, Ciit Int J Artif Intell Syst Mach Learn, 3(12), 762-766, 2011.
[40] Sadık Ünlü, B., Durmuş, H., & Meriç, C., “Determination of tribological properties at CuSn10 alloy journal bearings by experimental and means of artificial neural networks method”, Industrial Lubrication and Tribology, 64(5), 258-264, 2012.
[41] Gorasso, L., & Wang, L., “Journal bearing optimization using nonsorted genetic algorithm and artificial bee colony algorithm”, Advances in Mechanical Engineering, 6, 213548, 2014.
[42] Wu D Z, Jennings C, Terpenny J, Gao R X, Kumara S., “A comparative study on machine learning algorithms for smart manufacturing: Tool wear prediction using random forests”, J Manuf Sci Eng 139(7): 071018, 2017.
[43] Kornaev, A. V., Kornaev, N. V., Kornaeva, E. P., & Savin, L. A., “Application of artificial neural networks to calculation of oil film reaction forces and dynamics of rotors on journal bearings”, International Journal of Rotating Machinery, (1), 9196701, 2017.
[44] Kornaeva, E. P., Kornaev, A. V., Kazakov, Y. N., & Polyakov, R. N., “Application of artificial neural networks to diagnostics of fluid-film bearing lubrication”, In IOP Conference Series: Materials Science and Engineering , Vol. 734, No. 1, p. 01215, 2020.
[45] B. Roy and S. Dey, “Machine learning-based performance analysis of two-axial-groove hydrodynamic journal bearings”, Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, vol. 235, no. 10, pp. 2211-2224, 2021.
[46] Yang, J., & Palazzolo, A., “Computational fluid dynamics based mixing prediction for tilt pad journal bearing TEHD modeling—Part II: Implementation with machine learning”, Journal of Tribology, 143(1), 011802, 2021.
[47] Kazakov, Y. N., Kornaev, A. V., Shutin, D. V., Li, S., & Savin, L. A., “Active fluid-film bearing with deep Q-network agent-based control system”, Journal of Tribology, 144(8), 081803, 2022.
[48] Kazakov, Y., Stebakov, I., Shutin, D., & Savin, L., “Application of machine learning in simulation models and optimal controllers for fluid film bearings”, In International Conference on Rotor Dynamics, (pp. 224-235), 2023.
[49] Katsaros, K. P., & Nikolakopoulos, P. G., “Performance prediction model for hydrodynamically lubricated tilting pad thrust bearings operating under incomplete oil film with the combination of numerical and machine-learning techniques”, Lubricants, 11(3), 113, 2023.
[50] Gheller, E., Chatterton, S., Panara, D., Turini, G., & Pennacchi, P., “Machine Learning-Based Optimizer for Tilting Pad Journal Bearing Inlet Flowrate”, Journal of Engineering for Gas Turbines and Power, 145(12), 2023.
[51] S. Cartwright, B. C. Rothwell, G. Figueredo, H. Medina, C. Eastwick, J. Layton, and S. Ambrose, “A machine learning-driven approach to predicting thermo-elasto-hydrodynamic lubrication in journal bearings,” Tribology International, vol. 196, p. 109670, 2024.
[52] Chen, T., & Guestrin, C., “Xgboost: A scalable tree boosting system” In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794), 2016.