Response surface analysis of statistical features of voltage and current in a GMAW powersource on welding v-groove joints

Past studies have suggested that welding current and voltage are signature variables that have potential information on welding quality and can be used for real-time monitoring while welding. This work explores the interaction between gas metal arc welding parameters such as stick out distance, gas flow rate and transverse speed to the mean and standard deviation of welding voltage and current. SA 516 gr.70 boiler plates with v-grooves were welded using KUKA KR robot and Fronius TPS 5000 power source. Data acquisition system with a sampling rate of 8 kHz was used to acquire current and voltage data during welding. Experimental tests were organized using central composite design. Mean and standard deviation of voltage and current are analyzed using response surface analysis. The measured voltage was confirmed previously to be an effective parameter for detecting porosity using control chart and probability density distribution. Therefore, in this research, the effects and interactions of welding parameters are statistically analyzed. Stick out distance is concluded to be the most effective factor on to the responses being investigated. A change in the gas flow rate can affect the voltage significantly but will not alter the welding current. In addition, mathematical models are developed to forecast mean and normal voltage and current deviations in real-time.