14 January 2012 Modeling and optimizing electrodischarge machine process (EDM) with an approach based on genetic algorithm
Author Affiliations +
Proceedings Volume 8349, Fourth International Conference on Machine Vision (ICMV 2011): Machine Vision, Image Processing, and Pattern Analysis; 83493C (2012); doi: 10.1117/12.921077
Event: Fourth International Conference on Machine Vision (ICMV 11), 2011, Singapore, Singapore
Abstract
Electro Discharge Machine (EDM) is the commonest untraditional method of production for forming metals and the Non-Oxide ceramics. The increase of smoothness, the increase of the remove of filings, and also the decrease of proportional erosion tool has an important role in this machining. That is directly related to the choosing of input parameters.The complicated and non-linear nature of EDM has made the process impossible with usual and classic method. So far, some methods have been used based on intelligence to optimize this process. At the top of them we can mention artificial neural network that has modelled the process as a black box. The problem of this kind of machining is seen when a workpiece is composited of the collection of carbon-based materials such as silicon carbide. In this article, besides using the new method of mono-pulse technical of EDM, we design a fuzzy neural network and model it. Then the genetic algorithm is used to find the optimal inputs of machine. In our research, workpiece is a Non-Oxide metal called silicon carbide. That makes the control process more difficult. At last, the results are compared with the previous methods.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Iman Zabbah, "Modeling and optimizing electrodischarge machine process (EDM) with an approach based on genetic algorithm", Proc. SPIE 8349, Fourth International Conference on Machine Vision (ICMV 2011): Machine Vision, Image Processing, and Pattern Analysis, 83493C (14 January 2012); doi: 10.1117/12.921077; https://doi.org/10.1117/12.921077
PROCEEDINGS
7 PAGES


SHARE
KEYWORDS
Neural networks

Genetic algorithms

Fuzzy logic

Process modeling

Silicon carbide

Fuzzy systems

Carbon

RELATED CONTENT


Back to Top