The aim of this research was to develop an adaptive quality control strategy for roboticgas metal arc welding of thin steel sheets. Statistical methods were used to monitor andcontrol the quality of welds produced.The quality of welds cannot be directly measured during welding. It can however beestimated by correlating weld quality parameters to relevant process variables. It wasfound sufficient to do this using welding current and voltage transient signals only.The strategy developed was problem solving oriented with emphasis on qualityassurance, defect detection and prevention. It was based on simple algorithms developedusing multiple regression models, fuzzy regression models and subjective rules derivedfrom experimental trials.The resulting algorithms were used tocontrol weld bead geometry;prevent inadequate penetration;detect and control metal transfer;assess welding arc stability;optimise welding procedure;prevent undercut;detect joint geometry variations.Modelling was an integral part of this work, and as a feasibility study, some of themodels developed for process control were remodelled using "Backpropagation"Artificial Neural Networks. The neural network models were found to offer nosignificant improvement over regression models when used for estimating weld qualityfrom welding parameters and predicting optimum welding parameter.As a result of the work a multilevel quality control strategy involving preweld parameteroptimisation, on line control and post weld analysis was developed and demonstratedin a production environment. The main emphasis of the work carried out was ondeveloping control models and means of monitoring the process on-line; theimplementation of robotic control was outside the scope of this work. The controlstrategy proposed was however validated by using post weld analysis and simulation insoftware.
展开▼
