| Abstract: |
It is well known that, surface roughness plays an important role on the contactbehavior of surfaces in contact. Therefore, in order to study any tribologica1phenomena (such as friction and wear) between any two rough surfaces, it isnecessary to characterize each surface by using a scale-independent mathematicalmethod that gives unique values for the characterized surface. Thus, the fractalgeometry approach is utilized, through the present work, instead of theconventional surface roughness characterization method in order to characterizethe roughness profiles of machined surfaces by two fractal parameters: fractaldimension ”D” and vertical scaling parameter ”G”.In addition, the final geometry of surface roughness is influenced by variousmachining conditions such as spindle speed, feed, depth of cut, pre-tool flankwear, and vibration level. Artificial neural networks models (ANNs) and fractalgeometry approach were utilized to correlate between the different cuttingconditions and the corresponding surface roughness profiles. Modeling of thesurface roughness of machined specimens using computer-based programsenables the manufacturers to examine the predicted machined surface profilesvia computer running process, that include surface roughness prediction.Thus, the present work is intended to manufacture a number of specimens atdifferent machining conditions. The machined specimens were characterizedusing the conventional roughness parameters ”R,”, ”R;’, ”R,”, and the fractalparameters ”D” and ”G”. A statistical comparison was made between the twocharacterization methods for all machined surfaces. It was found that the fractalcharacterization method is better than the conventional surface roughnesscharacterization method.
|
|
|