A methodology for data-driven modeling and prediction of the drag losses of wet clutches

Forschung auf dem Gebiet des Ingenieurwesens A - 2023
Lukas Pointner-Gabriel1, Katharina Voelkel1, Hermann Pflaum1, Karsten Stahl1
1School of Engineering and Design, Department of Mechanical Engineering, Gear Research Center (FZG), Technical University of Munich, Boltzmannstraße 15, 85748, Garching near Munich, Germany

Tóm tắt

AbstractIn wet clutches, load-independent drag losses occur in the disengaged state and under differential speed due to fluid shearing. The drag torque of a wet clutch can be determined accurately and reliably by means of costly and time-consuming measurements. As an alternative, the drag losses can already be precisely calculated in the early development phase using computing-intensive CFD models. In contrast, simple analytical calculation models allow a rough but non-time-consuming estimation. Therefore, the aim of this study was to develop a methodology that can be used to build a data-driven model for the prediction of the drag losses of wet clutches with low computational effort and, at the same time, sufficient accuracy under consideration of a high number of influencing parameters. For building the model, we use supervised machine learning algorithms. The methodology covers all relevant steps, from data generation to the validated prediction model as well as its usage. The methodology comprises six main steps. In Step 1, the data is generated on a suitable test rig. In Step 2, characteristic values of each measurement are evaluated to quantify the drag loss behavior. The characteristic values serve as target values to train the model. In Step 3, the structure and quality of the dataset are analyzed and, subsequently, the model input parameters are defined. In Step 4, the relationships between the investigated influencing parameters (model input) and the characteristic values (model output) are determined. Symbolic regression and Gaussian process regression have both been proven to be suitable for this task. Lastly, the model is used in Step 5 to predict the characteristic values. Based on the predictions, the drag torque can be predicted as a function of differential speed in Step 6, using an approximation function. The model allows a user-oriented prediction of the drag torque even for a high number of parameters with low computational effort and sufficient accuracy at the same time.

Từ khóa


Tài liệu tham khảo

Plothe A, Graswald C, Grüning A et al. (2017) Effizienzsteigerung bei modernen Antriebssystemen durch Kombination von Simulation und Versuch. In: Liebl J (ed) Reibungsminimierung im Antriebsstrang. Springer Vieweg, Wiesbaden, pp 143–158. https://doi.org/10.1007/978-3-658-19521-2_8

Neupert T, Benke E, Bartel D (2018) Parameter study on the influence of a radial groove design on the drag torque of wet clutch discs in comparison with analytical models. Tribol Int 119:809–821. https://doi.org/10.1016/j.triboint.2017.12.005

Iqbal S, Al-Bender F, Pluymers B et al. (2013) Experimental characterization of drag torque in open multi-disks wet clutches. SAE Int J Fuels Lubr 6(3):894–906. https://doi.org/10.4271/2013-01-9073

Pointner-Gabriel L, Forleo C, Voelkel K et al. (2022) Investigation of the drag losses of wet clutches at dip lubrication. SAE Technical Paper 2022-01-0650. https://doi.org/10.4271/2022-01-0650

Kitabayashi H, Li CY, Hiraki H (2003) Analysis of the various factors affecting drag torque in multiple-plate wet clutches. SAE Technical Paper 2003-01-1973. https://doi.org/10.4271/2003-01-1973

Draexl T, Pflaum H, Stahl K (2013) Schleppverluste Lamellenkupplungen: Wirkungsgradverbesserung durch Reduzierung der Schleppverluste an Lamellenkupplungen, FVV 1012. Final Report. Forschungsvereinigung Verbrennungskraftmaschinen (FVV) e. V., Frankfurt a. M. (in German)

Schade CW (1971) Effects of transmission fluid on clutch performance. SAE Technical Paper 710734. https://doi.org/10.4271/710734

Kaebernick H (1973) Untersuchungen zum thermischen Verhalten von Elektromagnet-Lamellenkupplungen in Werkzeugmaschinengetrieben. Dissertation, TU Berlin (in German)

Szalai G, Ray R, Bansal H et al. (2022) Wet clutch drag loss simulation for different clutch patterns. SAE Technical Paper 2022-01-1118. https://doi.org/10.4271/2022-01-1118

Oerleke C, Funk W (2000) Leerlaufverhalten von ölgekühlten Lamellenkupplungen, FVA 290. Final Report. Forschungsvereinigung Antriebstechnik (FVA) e. V., Frankfurt a. M. (in German)

Groetsch D, Niedenthal R, Voelkel K et al. (2020) Volume of fluid vs. cavitation CFD-models to calculate drag torque in multi-plate clutches. SAE Technical Paper 2020-01-0495. https://doi.org/10.4271/2020-01-0495

Rudloff M (2013) Experimentelle Untersuchung und Strömungssimulation zur Beschreibung von Schleppmomenten in ölgekühlten Lamellenkupplungen. Dissertation, Otto-von-Guericke-Universität Magdeburg (in German)

Draexl T, Pflaum H, Stahl K (2016) Schleppverluste Lamellenkupplungen II: Wirkungsgradverbesserung durch Reduzierung der Schleppverluste an Lamellenkupplungen, FVA 671 I. Final Report. Forschungsvereinigung Antriebstechnik (FVA) e. V., Frankfurt a. M. (in German)

Fish RL (1991) Using the SAE #2 machine to evaluate wet clutch drag losses. SAE Technical Paper 910803. https://doi.org/10.4271/910803

Hu J, Peng Z, Wei C (2012) Experimental research on drag torque for single-plate wet clutch. J Tribol 134(1):14502. https://doi.org/10.1115/1.4005528

Pointner-Gabriel L, Pflaum H, Voelkel K et al. (2022) Schleppmomentberechnung: Berechnung der Schleppmomente nasslaufender Lamellenkupplungen, FVA 671 II. Final Report. Forschungsvereinigung Antriebstechnik (FVA) e. V., Frankfurt a. M. (in German)

Wu P, Zhou X, Yang C et al. (2018) Parametric analysis of the drag torque model of wet multi-plate friction clutch with groove consideration. Ind Lubr Tribol 70(7):1268–1281. https://doi.org/10.1108/ILT-03-2017-0063

Albers A, Denda C, Basiewicz M (2017) Validierung und Untersuchung von Nutgeometrien zur Reduzierung von Schleppverlusten nasser Lamellenkupplungen. In: GfT (ed) 58. Tribologie-Fachtagung (in German)

Razzaque MM, Kato T (1999) Effects of groove orientation on hydrodynamic behavior of wet clutch coolant films. J Tribol 121(1):56–61. https://doi.org/10.1115/1.2833811

Asai K, Ito T (2018) Effect of facing groove design on drag torque of automatic transmission wet clutches. SAE Technical Paper 2018-01-0400. https://doi.org/10.4271/2018-01-0400

Wang P, Katopodes N, Fujii Y (2018) Statistical modeling of plate clearance distribution for wet clutch drag analysis. SAE Int J Passeng Cars Mech Syst 11(1):76–88. https://doi.org/10.4271/06-11-01-0007

Yuan Y, Liu EA, Hill J et al. (2007) An improved hydrodynamic model for open wet transmission clutches. J Fluids Eng 129(3):333–337. https://doi.org/10.1115/1.2427088

Mahmud SF, Pahlovy SA, Kubota M et al. (2017) A simulation model for predicting high speed torque jump up phenomena of disengaged transmission wet clutch. SAE Technical Paper 2017-01-1139. https://doi.org/10.4271/2017-01-1139

Mahmud S, Pahlovy SA (2015) Investigation on torque jump up and vibration at high rotation speed of a wet clutch. SAE Technical Paper 2015-01-2184. https://doi.org/10.4271/2015-01-2184

Hu J, Hou S, Wei C (2018) Drag torque modeling at high circumferential speed in open wet clutches considering plate wobble and mechanical contact. Tribol Int 124:102–116. https://doi.org/10.1016/j.triboint.2018.03.029

Klausner M, Funk W (1991) Lamellentaumeln: Untersuchung des Betriebsverhaltens nasslaufender Lamellenkupplungen bei höheren Relativdrehzahlen, FVA 117. Final Report. Forschungsvereinigung Antriebstechnik e. V., Frankfurt a. M. (in German)

Neupert T, Bartel D (2019) High-resolution 3D CFD multiphase simulation of the flow and the drag torque of wet clutch discs considering free surfaces. Tribol Int 129:283–296. https://doi.org/10.1016/j.triboint.2018.08.031

Pardeshi I, Shih TI‑P (2019) A computational fluid dynamics methodology for predicting aeration in wet friction clutches. J Fluids Eng 141(12):121304. https://doi.org/10.1115/1.4044071

Pan H, Zhou X (2019) Simulation research on the drag torque of disengaged wet clutches. In: 2019 IEEE 5th International Conference on Mechatronics System and Robots (ICMSR). IEEE, pp 44–48 https://doi.org/10.1109/ICMSR.2019.8835458

Peng Z, Yuan S (2019) Mathematical model of drag torque with surface tension in single-plate wet clutch. Chin J Mech Eng 32:25. https://doi.org/10.1186/s10033-019-0343-9

Wu W, Xiong Z, Hu J et al. (2015) Application of CFD to model oil–air flow in a grooved two-disc system. Int J Heat Mass Transf 91:293–301. https://doi.org/10.1016/j.ijheatmasstransfer.2015.07.092

Takagi Y, Okano Y, Miyagawa M et al. (2011) Combined numerical and experimental study on drag torque in a wet clutch. In: Proceedings of the ASME-JSME-KSME 2011 Joint Fluids Engineering Conference, vol 1. ASME, pp 2425–2430. https://doi.org/10.1115/ajk2011-10006

Yuan Y, Attibele P, Dong Y (2013) CFD simulation of the flows within disengaged wet clutches of an automatic transmission. SAE Technical Paper 2003-01-0320. https://doi.org/10.4271/2003-01-0320

Groetsch D, Niedenthal R, Voelkel K et al. (2021) Efficient CFD simulation method for calculation of drag torque in wet multi-plate clutches in comparison to test rig results. In: CTI SYMPOSIUM 2019, vol 2245. Springer, Berlin, Heidelberg, pp 164–176. https://doi.org/10.1007/978-3-662-61515-7_15

Hirt C, Nichols B (1981) Volume of fluid (VOF) method for the dynamics of free boundaries. J Comput Phys 39(1):201–225. https://doi.org/10.1016/0021-9991(81)90145-5

Takagi Y, Nakata H, Okano Y et al. (2011) Effect of two-phase flow on drag torque in a wet clutch. J Adv Res Phys 2(2). article number 021108

Groetsch D, Niedenthal R, Voelkel K et al. (2019) Effiziente CFD-Simulationen zur Berechnung des Schleppmoments nasslaufender Lamellenkupplungen im Abgleich mit Prüfstandmessungen. Forsch Ingenieurwes 83(2):227–237. https://doi.org/10.1007/s10010-019-00302-3

Singhal AK, Athavale MM, Li H et al. (2002) Mathematical basis and validation of the full cavitation model. J Fluids Eng 124(3):617–624. https://doi.org/10.1115/1.1486223

Iqbal S, Al-Bender F, Pluymers B et al. (2013) Mathematical model and experimental evaluation of drag torque in disengaged wet clutches. Int Sch Res Notices. https://doi.org/10.5402/2013/206539

Rao G (2011) Modellierung und Simulation des Systemverhaltens nasslaufender Lamellenkupplungen. Dissertation, TU Dresden (in German)

Cui H, Yao S, Yan Q et al. (2014) Mathematical model and experiment validation of fluid torque by shear stress under influence of fluid temperature in hydro-viscous clutch. Chin J Mech Eng 27(1):32–40. https://doi.org/10.3901/CJME.2014.01.032

Pahlovy SA, Mahmud SF, Kubota M et al. (2014) Multiphase drag modeling for prediction of the drag torque characteristics in disengaged wet clutches. SAE Int J Commer Veh 7(2):441–447. https://doi.org/10.4271/2014-01-2333

Pahlovy SA, Mahmud SF, Kubota M et al. (2017) Development of an analytical model for prediction of drag torque characteristics of disengaged wet clutches in high speed region. SAE Technical Paper 2017-01-1132. https://doi.org/10.4271/2017-01-1132

Pahlovy SA, Mahmud SF, Kubota M et al. (2016) New development of a gas cavitation model for evaluation of drag torque characteristics in disengaged wet clutches. SAE Int J Engines 9(3):1910–1915. https://doi.org/10.4271/2016-01-1137

Iqbal S, Al-Bender F, Pluymers B et al. (2014) Model for predicting drag torque in open multi-disks wet clutches. J Fluids Eng 136(2):21103. https://doi.org/10.1115/1.4025650

Yuan S, Peng Z, Jing C (2011) Experimental research and mathematical model of drag torque in single-plate wet clutch. Chin J Mech Eng 24(1):91. https://doi.org/10.3901/CJME.2011.01.091

Nasiri H, Delprete C, Brusa E et al. (2022) Analytical simulation of influential parameters affecting grooved wet clutches performance under disengagement condition. Proc Inst Mech Eng Part J: J Eng Tribol 236(6):1113–1122. https://doi.org/10.1177/13506501211047783

Montáns FJ, Chinesta F, Gómez-Bombarelli R et al. (2019) Data-driven modeling and learning in science and engineering. C R Méc 347(11):845–855. https://doi.org/10.1016/j.crme.2019.11.009

Automatic Transmission and Transaxle Committee (2012) SAE no. 2 clutch friction test machine guidelines. SAE International https://doi.org/10.4271/J286_201203

Oh Y, Jang S (2021) Study on the drag torque characteristics wet clutch system under the operating environment conditions. Trans KSAE 29(1):35–41. https://doi.org/10.7467/KSAE.2021.29.1.035

Goszczak J, Leyko J, Mitukiewicz G et al. (2022) Experimental study of drag torque between wet clutch discs. Appl Sci 12(8):3900. https://doi.org/10.3390/app12083900

Lloyd FA (1974) Parameters contributing to power loss in disengaged wet clutches. SAE Technical Paper 740676. https://doi.org/10.4271/740676

Mahmud SF, Pahlovy SA, Kubota M et al. (2016) Multi-phase simulation for studying the effect of different groove profiles on the drag torque characteristics of transmission wet clutch. SAE Technical Paper 2016-01-1144. https://doi.org/10.4271/2016-01-1144

Leighton M, Morris N, Trimmer G et al. (2019) Efficiency of disengaged wet brake packs. Proc Inst Mech Eng Part D: J Automob Eng 233(6):1562–1569. https://doi.org/10.1177/0954407018758567

Aphale CR, Schultz WW, Ceccio SL (2011) Aeration in lubrication with application to drag torque reduction. J Tribol 133(3):31701. https://doi.org/10.1115/1.4004303

Neupert T, Bartel D (2021) Measurement of pressure distribution and hydrodynamic axial forces of wet clutch discs. Tribol Int 163:107172. https://doi.org/10.1016/j.triboint.2021.107172

Albers A, Ott S, Basiewicz M et al. (2017) Variation von Nutbildern mittels generativer Verfahren zur Untersuchung von Schleppverlusten in Lamellenkupplungen. In: VDI (ed) Kupplungen und Kupplungssysteme in Antrieben 2017. VDI, Düsseldorf, pp 293–300. https://doi.org/10.51202/9783181023099-293 (in German)

Pan H, Zhou X (2019) Experimental and theoretical analysis of the drag torque in wet clutches. Fluid Dyn Mater Process 15(4):403–417. https://doi.org/10.32604/fdmp.2019.07808

Aphale CR, Schultz WW, Ceccio SL (2010) The influence of grooves on the fully wetted and aerated flow between open clutch plates. J Tribol 132(1):11104. https://doi.org/10.1115/1.3195037

Beisel W, Federn K (1982) Lamellenwellung: Untersuchung des Einflusses einer Sinuswellung der Stahllamellen auf das Leerlaufverhalten von Lamellenkupplungen mit der Reibpaarung Stahl/Sinterbronze bei unterschiedlicher Reibflächengestaltung, FVA 53 II. Final Report. Forschungsvereinigung Antriebstechnik (FVA) e. V., Frankfurt a. M. (in German)

Neupert T, Bartel D (2015) Schleppmomentuntersuchungen an nasslaufenden Kupplungslamellen mithilfe von Prüfstandsmessung und CFD-Simulation. In: VDI (ed) Kupplungen und Kupplungssysteme in Antrieben 2015. VDI, Düsseldorf (in German)

DIN 51563 (2011) Testing of mineral oils and related materials: determination of viscosity temperature relation—Slope m

DIN 51757 (2011) Testing of mineral oils and related materials: Determination of density

Greenhill S, Rana S, Gupta S et al. (2020) Bayesian optimization for adaptive experimental design: a review. IEEE Access 8:13937–13948. https://doi.org/10.1109/ACCESS.2020.2966228

Woods DC, Lewis SM (2016) Design of experiments for screening. In: Ghanem R, Higdon D, Owhadi H (eds) Handbook of uncertainty quantification. Springer, Cham, pp 1–43 https://doi.org/10.1007/978-3-319-11259-6_33-1

Arboretti R, Ceccato R, Pegoraro L et al. (2022) Design of Experiments and machine learning for product innovation: A systematic literature review. Qual Reliab Eng 38(2):1131–1156. https://doi.org/10.1002/qre.3025

Joseph VR (2016) Space-filling designs for computer experiments: A review. Qual Eng 28(1):28–35. https://doi.org/10.1080/08982112.2015.1100447

Liu H, Ong Y‑S, Cai J (2018) A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design. Struct Multidiscipl Optim 57(1):393–416. https://doi.org/10.1007/s00158-017-1739-8

Bishop CM (2006) Pattern recognition and machine learning. Computer science. Springer New York, New York

Poli R, Langdon WB, McPhee NF (2008) A field guide to genetic programming. Lulu Press, Morrisville, NC

Koza JR (1994) Genetic programming as a means for programming computers by natural selection. Stat Comput 4:87–112. https://doi.org/10.1007/BF00175355

gplearn Genetic programming in python, with a scikit-learn inspired API. https://gplearn.readthedocs.io/en/stable/. Accessed 20 Jan 2023

Rasmussen CE, Williams CKI (2006) Gaussian processes for machine learning. Adaptive computation and machine learning. MIT Press, Cambridge

Görtler J, Kehlbeck R, Deussen O (2019) A visual exploration of Gaussian processes. Distill. https://doi.org/10.23915/distill.00017

Rasmussen CE (2004) Gaussian processes in machine learning. In: Bousquet O, von Luxburg U, Rätsch G (eds) Advanced lectures on machine learning, vol 3176. Springer, Berlin, Heidelberg, pp 63–71 https://doi.org/10.1007/978-3-540-28650-9_4

Duvenaud D (2014) Automatic Model Construction with Gaussian Processes. Dissertation, University of Cambridge

Matthews AGdG, van der Wilk M, Nickson T et al. (2017) GPflow: a Gaussian process library using tensorflow. J Mach Learn Res 18(40):1–6

GPy: A Gaussian Process (GP) framework in Python. https://gpy.readthedocs.io/en/deploy/. Accessed 20 Jan 2023

Roh Y, Heo G, Whang SE (2021) A survey on data collection for machine learning: a big data—AI integration perspective. IEEE Trans Knowl Data Eng 33(4):1328–1347. https://doi.org/10.1109/TKDE.2019.2946162

Thông tin
Thông tin xuất bản

Forschung auf dem Gebiet des Ingenieurwesens A

Thông tin tác giả