Research On Correlation Between Manufacturing Precision And Engine Friction Performance

Engine friction performance, which is influenced by themanufacturing precision of engine friction pairs as well as the performanceof engine lubrication system, determines the performance and life of theengine directly. As the quick development of Chinese vehicle market, ithas been a tendency that more and more car companies were established inChina. However, even produced in the same manufacturing line, domesticengines always have big performance fluctuation, as the design andmanufacturing can’t match well. For one reason, the level ofmanufacturing cannot follow the requirement of design, thus the toleranceof many components are not well assigned, resulting in too high scrap rate.For another, many producers have little knowledge of relationship betweentolerance and engine performance fluctuation. So they arbitrarily changethe tolerances to reduce the costs which they only care about. Therefore,this paper worked on the project to investigate how to build a model thatconnects the components tolerances and engine friction performancefluctuation. Then the engine performance uniformity can be improved byoptimizing the key dimension tolerance.In this paper, engine performance model is firstly built to investigatethe influence of friction loss changes on engine performance, includingoutput torque, power and oil consumption. Then friction performance ofpiston system and crankshaft system, which are two most important partsof friction loss, were investigated using multi-body dynamics (MBD).Latin Hypercube Sampling (LHS) was used to conduct Design ofExperiment (DOE).5from piston system and2from crankshaft system, parameters were chosen from lots of parameters as design variables toattend the optimization. Some analyses on dimensional chains were madeand key parameters tolerances grades were given. Kriging interpolationmethod was implemented in this paper, to establish approximation modelsof engine piston and crankshaft system friction fluctuation. Besides, costand tolerance model was also built using the method of fuzzycomprehensive evaluation. Optimization of the key parameters tolerancesbased on approximation models was conducted using Genetic Algorithm(GA). The best tolerances design of key parameters was got, and thefriction fluctuation of the engine piston system dropped42.85%,crankshaft systems32.09%.