Return to search

Numerical simulation and optimization of micro-EDM using geometrical methods and machine learning

As the need for smaller, more compact and integrated products has evolved, it is no surprise that manufacturing technologies have significantly evolved in order to make miniaturisation to smaller scales possible. More specifically non-conventional machining technologies, relative newcomers in the field of machining, have proven well suited to the task at hand. Among those technologies is micro-EDM (short for Electrical Discharge Machining) that has been the subject of numerous developments. A certain number of variants of micro-EDM exists among which are wire micro-EDM, die-sinking micro-EDM, micro-EDM milling and micro-EDM drilling. While die-sinking macro-EDM is quite common, its micro counterpart isn’t due to problematic tool wear. In order to optimise the die-sinking micro-EDM process in terms of time and cost and make its use more interesting and viable, the present work aims at optimizing the initial tool shape so that it compensates for future wear. The first step was to design a simulation tool effectively able to predict the location and magnitude of wear during the simulation process. An iterative geometrical method was developed, first using NURBS as support geometries then voxels embedded in an octree data structure in order to improve speed and accuracy.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:669181
Date January 2015
CreatorsSurleraux, Anthony
PublisherCardiff University
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Sourcehttp://orca.cf.ac.uk/80776/

Page generated in 0.0021 seconds