LIBRARY ı6ıul 0) - Cranfield University

3.3.2 Off-line programming module
In the off-line programming module, the welded joints are located and the
quality requirements input. The module then predicts a list of welding parameters
which are expected to produce the required quality and from which the user should
choose the best suited set. Further to predicting the welding parameters, this module
extracts geometrical information from the joint and generates the co-ordinates of the
start and end of the weld as well as torch approaching and withdrawing points
(positions and orientations). The geometrical and welding data are stored together
with the CAD model data, as extended entity data' . Therefore, the designed welding
data is kept as long as the CAD drawing exists.
The off-line programming module was developed using AutoLISP®2 and
consists of two main branches, the welding parameters generator and the robot
programming branch (see Figure 3.2).
3.3.2.1 Welding parameters generator
The welding parameters generator (dot-dashed line in Figure 3.2) outputs
optimised parameters based on the geometry required for each weld bead. The
parameters are calculated based on the empirical welding models developed by
Ogunbiyi [ref. 51] for the gas metal arc welding of thin sheet steel.
Three types of prediction models are used: a) models for predicting process
ideal features under stable conditions; b) models for predicting bead geometry, and c)
models for predicting process stability. These models were developed by applying
multiple regression and fuzzy regression analysis. The welding data used were
collected for a stable process carried out with travel speeds between 0.4 m/min and
1.6 m/min, wire feed speeds between 4 m/min and 16 m/min and contact tip-to-
workpiece distances (stand-off) between 12 mm and 20 mm. For each travel speed,
wire feed speed and stand-off combination, the voltage was chosen to give the most
stable process. The models are outlined below in equations (3.1) to (3.10).
a) Models for predicting process ideal features
Expected mean current from procedure:
I.. =a, +ß, WFS+S, SO"IVFS (3.1)
Expected maximum to mean current ratio (TST):
'max =a2+ß21e, +62SO
'mean
(3.2)
The concept of extended entity data and its structure can be found in the AutoCAD Release 12 -
Customization Manual
2A description of the AutoLISP functions and its sintax can be found in the AutoCAD Release 12
-
AutoLISP Programmer's Reference
68