Study On Deformation Prediction And Parameter Optimizationof Key Processes Of Marine Forged Aluminum Piston Skirt

Piston is one of the most important parts of marine diesel engine.By connecting the piston pin with connecting rod,the linear motion of piston obtained after gas burst pressure is transformed into the rotary motion of crankshaft.Its processing quality directly affects the power and service life of diesel engine.Therefore,the production and processing of piston has strict shape and position tolerance and size requirements.Forged aluminium piston is prone to irregular deformation of skirt due to its complex processing and poor stiffness,which can not meet the production requirements and is difficult to guarantee the processing quality.Therefore,in view of the processing process of forged aluminium piston,it is very important to study the methods of processing deformation prediction and optimization of processing parameters for improving the quality of forged aluminium piston production and reducing the rejection rate of products.In this paper,the processing deformation of forged aluminium piston is studied.Firstly,the simulation model of the processing process is esTable lished by using finite element analysis technology.Then,the prediction model of the processing deformation of forged aluminium piston is esTable lished based on the prediction method.Then,based on the prediction model,the genetic algorithm is used to optimize the cutting parameters.Finally,based on the above research results,the cutting parameters are optimized.A simulation system for deformation prediction and optimization of marine diesel engine piston processing is developed.Specific research contents are as follows:(1)Based on ANSYS and DEFORM simulation software,the simulation model of the key process of forged aluminium piston was esTable lished,and the deformation of piston skirt length and short axle under the influence of clamping force,cutting tool and cutting parameters was obtained.Finally,the clamping and cutting experiments of forged aluminium piston were designed to verify the reliability of the finite element theoretical model.(2)Using the theory of factorial design,the factorial design of the influence factors(cutting speed,cutting depth,feed rate,tool tip radius and tool wear length)of piston processing was carried out,and the most significant influence factors on the deformation of piston after cutting were screened out.(3)Using BP neural network modeling and multiple regression modeling,cutting deformation prediction models of piston key processes are esTable lished respectively.Compared with the prediction results,the prediction model based on BP neural network has higher prediction accuracy.Finally,BP neural network prediction model is selected as the mapping relationship model between cutting parameters and piston long-short axis deformation.(4)Based on the multi-objective optimization principle of genetic algorithm and the neural network model,the cutting parameters of forged aluminium piston are optimized,and a set of optimal solutions are obtained.The validity of the optimization method is verified by optimizing the cutting parameters with the experience of the factory.Finally,it integrates the prediction technology of piston processing deformation and the optimization technology of cutting parameters into the optimization and prediction simulation system of piston processing parameters of marine diesel engine,so as to provide guidance for actual piston production and processing.