Pattern recognition of micro and macro grinding phenomenon with a generic strategy to machine process monitoring

Griffin, James (2008) Pattern recognition of micro and macro grinding phenomenon with a generic strategy to machine process monitoring. PhD thesis, University of Nottingham.

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (14MB) | Preview

Abstract

Abstract

In modern manufacturing environments waste is an issue of great importance. Specifically the research in this thesis looks at issues in establishing the initial steps to gain a generic process monitoring system that ensures that grinding is both optimised but not the determent where costly malfunctions mean the scrapping and re-melting of expensive quality intensive materials.

The research conducted in this thesis investigates the process of cutting, ploughing and rubbing during single grit scratch tests. These investigations meant the correlation between physical material removal phenomenon and the emitted material dislocations gained from acoustic emission extraction. The initial work looked at different aerospace materials and the distinction of cutting, ploughing and rubbing during single grit radial scratch tests. This initial work provided novel results not seen in this area before and paved the way for more robust results in investigating the same phenomena during horizontal single grit scratch tests. This work provided more robust classification of cutting, ploughing and rubbing and transferred directly to grinding pass cuts from 1um and 0.1mm depth cuts respectively. In using robust classifiers such as the Neural Network and novel classifiers such as non-linear data paradigms, Fuzzy-c clustering with Genetic Algorithm optimisation, cutting, ploughing and rubbing phenomenon was investigated. These investigations showed that more cutting occurs when there is moreinteraction between grit and workpiece based on the increase depth of cut. Other thesis results investigated a generic classifier using Genetic Programming to classify multiple anomaly phenomena. This work can be bridged together with the unit event grit classification work.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Chen, Xun
Keywords: Grinding, Single Grit, STFT, WPT, Neural Networks, Fuzzy Clustering, Genetic Algorithms, Genetic Programming, Acoustic Emission, Force, Burn, Chatter, Cutting, Ploughing, Rubbing, CMSX4, Inconel 718, EN8, MARM-002, Titanium-64,
Subjects: T Technology > TJ Mechanical engineering and machinery
Faculties/Schools: UK Campuses > Faculty of Engineering > Department of Mechanical, Materials and Manufacturing Engineering
Item ID: 10487
Depositing User: EP, Services
Date Deposited: 27 Jun 2008
Last Modified: 14 Oct 2017 16:40
URI: https://eprints.nottingham.ac.uk/id/eprint/10487

Actions (Archive Staff Only)

Edit View Edit View