الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 129 |  |  |
| | | from | 129 | | |
Abstract
Plasma arc cutting Process (PAC) is a thermal cutting operation which uses as a restricted, high-speed jet of extremely high temperature dissociated, ionized inert gas known as plasma (4th state of matter) formed from the electrical arc between the nozzle and the surface to melt and cut for electrically conductive materials.
This thesis deals with the potential of using PAC technology to cut copper plates and experimental investigations of the effect of cutting variables on the quality of the cut surface monitored by the arithmetic average roughness Ra, and perpendicularity error (conicity error). Material removal rate (MRR) and Heat effected zones (HAZ) are also monitored through the investigations in order to reach an approach to their optimal values.
Cutting current intensity, speed, standoff distance and cutting gas pressure were selected as the cutting variables. The experiments were all carried out on 5 mm specimens of copper. Ra was measured using Surtronic-3, Taylor-Hobson surface texture equipment, MRR calculated by the difference between samples weight after and before cutting operation. Conicity error using scanned pictures of the samples after cutting the operation and HAZ were monitored by a metrological microscope.
Analysis of Variance (ANOVA) by statistical investigation is employed after the experiments to calculate the percentage contribution of each variable on the output parameters and determine the most dominating variables to be controlled through the future research work. An Artificial Neural Network (ANN) is developed, trained and tested to predict the output parameter responses for copper materials under investigation.
Results from this investigation shows the effect of changing those variables on the measured parameters, and their contribution in the overall effect. The trained feed forward-back propagation artificial neural network was able to accurately predict the values of the input variables for the desired values of the output parameters and could be used to choose the optimal values for those cutting variables to achieve lower values of Ra, HAZ, conicity errors and higher values for MRR.