a b s t r a c   t Keywords: Connecting rod Stepwise regression Dynamic lubrication Orthogonal experiment Support vector machine (SVM)

Dynamic lubrication analysis of connecting rod is a very complex problem。 Some factors have great effect on lubrication, such as clearance, oil viscosity, oil supplying hole, bearing elastic modulus, surface rough- ness, oil supplying pressure and engine speed and bearing width。 In this paper, ten indexes are used as the input parameters to evaluate the bearing performances: minimum oil film thickness (MOFT), friction loss, the maximum oil film pressure (MOFP) and average of the oil leakages (OLK)。 Two orthogonal exper- iments are combined to identify the factors dominating the bearing behavior。 The stepwise regression is used to establish the regression model without insignificant variables, and two most important variables are used as the input to carry out the surface response analysis for each model。 At last, the support vector machine (SVM) is used to identify the asperity contact。 Compared with SVM model, the particle swarm optimization-support vector machines (PSO–SVM) can predict the asperity contact more precise, espe- cially to the samples near piding line。 In future work, more soft computing methods with statistical characteristic are used to the tribology analyses。86008

1。Introduction

It is estimated that mechanical friction loss accounts for around 10% of the total energy in the fuel for a diesel engine, and about 40–55% of the friction losses are due to the power cylinder system, made up of the piston (25–47%), ring-pack (28–45%) and connecting-rod bearings (18–33%) [1], so connecting rod also contributes main friction loss to the engine。 A connecting rod assembly in an engine, which consists of rod, cap, bolts, big-end bearing and small-end bushing, the connecting rod serves as an important joint between the connecting rod big-end and crank pin。 The connect- ing rod bearing is one of the most highly stresses tribological component in an engine because of the complicated dynamic loading and surface rubbing motion it encoun- ters。 The knowledge required for this particular task implies a good understanding of the connecting rod lubrication performances。

Bearings are the most important but also the most difficult elements to design, it is the most difficult to design due to the changeable bearing loads from the cylinder pressure and dynamic balance。 From the literature reviewed, during the past two decades, many research articles had been written about the connecting rod bear- ing。 Due to the important of flexible deformation and too many parameters  which

affect the lubrication, classic hydrodynamic lubrication model was development to elasto-hydrodynamic (EHD) lubrication or thermal elasto-hydrodynamic (TEHD) lubrication, nearly all the parameters that affect the performances were researcher。 Fantino et al。 [2,3] studied the effect of the deformation of an elastic automotive con- necting rod on the oil film characteristics in the big-end bearing。 Fantino and Frêne [4] studied the influence of the engine type petrol and diesel on the same result, but no conclusion could be made about the impact of other parameters load and speed, he also focused on the effect of the viscosity on the minimum film thickness for a connecting rod big-end bearing。 However, as concerns the speed influence。 Aitken [5,6] found an EHD parametric result, which shows a decrease of the mini- mum film thickness in the range 100–700 rpm。 A more complete parametric study which involves the load, engine speed, and bearing stiffness on the lubrication is also reported。 Okamoto et al。 [7] reported the effects of bearing length and housing stiffness on the connecting rod big-end bearing。 The results proved that the decrease of the bearing length has a significant incidence on the minimum film thickness and maximum pressure due to the load capacity reduction。