Research on MCVE piston machining and process parameter optimization

A piston is an important part of an engine. Its shape is designed into middle-convex and varying ellipse (MCVE) to adapt to the complex working environment. The main requirements of MCVE piston machining are high frequency response, small range tool motion, and high precision. In this article, an MC...

Full description

Saved in:
Bibliographic Details
Published in:International journal of advanced manufacturing technology 2017-12, Vol.93 (9-12), p.3955-3966
Main Authors: Zhang, Yong, Huang, Yu, Shao, WenJun, Ming, Wuyi
Format: Article
Language:English
Subjects:
Artificial neural networks
Back propagation
Back propagation algorithms
Back propagation networks
CAE) and Design
Computer-Aided Engineering (CAD
Data models
Ellipticity
Engineering
Frequency response
Genetic algorithms
Industrial and Production Engineering
Machining
Mathematical models
Mechanical Engineering
Media Management
Neural networks
Optimization
Original Article
Pistons
Process parameters
Surface roughness
Turning (machining)
Working conditions
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A piston is an important part of an engine. Its shape is designed into middle-convex and varying ellipse (MCVE) to adapt to the complex working environment. The main requirements of MCVE piston machining are high frequency response, small range tool motion, and high precision. In this article, an MCVE data model is established for the piston profile design, and the turning principle and control procedure are discussed to develop a fast tool servo (FTS) system for piston turning. In the end, back propagation neural network (BPNN) and genetic algorithm (GA) are combined to optimize the process parameters in the MCVE piston machining, which includes general turning parameters and special MCVE turning parameters. Through the experiments and BPNN-GA optimization, the ellipticity error ( E ) and surface roughness ( Ra ) of all pistons met the design requirements. According to verification experiments, the optimization results of E and Ra are 3.04 and 1.204 μm, respectively, and their relative errors are 10.13 and 4.27%, respectively. It has been proved that the MCVE data model and the control design of FTS are feasible and can effectively produce MCVE piston; the BPNN-GA optimization method is obviously effective and can improve processing effect and machining efficiency.
ISSN:0268-3768
1433-3015
DOI:10.1007/s00170-017-0838-4