This paper presents an analysis on heat flow behavior in the weldment, an experimentally proven temperature monitoring scheme and a control system designed to render uniform weld quality for the gas metal arc welding(GMAW) process parameters directly reflecting the weld quality. Here, the heat input generated by the electric arc is taken as the input and the temperature measured form the top surface of the weldment is taken as the output of the process, respectively.
To construct the feedback control system of the GMA welding process, the research presented in this paper consists of three parts: the heat flow analysis of the weldment, the analysis on the measurement location and the design of the process control system.
1) Heat flow analysis of the weldment:
A new analytical solution is introduced to predict the transient temperature distributions in a finite thickness plate during arc welding. This analytical solution is obtained by solving a three dimensional transient heat conduction equation with convection boundary conditions at the surfaces of the weldment. The heat source provided by electric arc is assumed to take a traveling Gaussian distribution. To prove the validity of the developed model, a series of bead on plate welding is performed on a medium carbon steel under various welding conditions. The results show that the finite thickness solution predicts the transient temperature distribution of the finite thickness or thin plates with good accuracy.
2) Analysis on the measurement location:
An infrared temperature sensing system is designed to measure the radiation emitted from the top surface of the weldment. The interference effect of the electric arc to the measurement is rejected by detecting the low peaks of the noisy signal. The influence of the measurement locations in monitoring the weld quality which is represented by weld bead width, depth of penetration and back bead width is analyzed based upon analytical solution. The theoretical results are compared with the experimental results obtained from a series of bead on plate welding performed on a medium carbon steel for various measurement locations. After a careful examination on these results, an optimizing criterion, in which the correlation between the weld quality and the measured temperature is maximized, is proposed to determine the optimal measurement location.
3) Design of the process control system:
The control purpose is to obtain uniform weld quality in GMA welding process and, in the research, this is accomplished by regulating the measured temperature into a constant desired temperature. Due to the complexity of the process and the bulkiness of the analytical solution, it is very difficult to design the control system based upon analytical process model. Thus, a self organizing fuzzy linguistic controller (SOFLC) which is not based upon the process model is proposed in this paper. The proposed SOFLC is processed in parallel with PD control with fixed gains. The fuzzy variables in SOFLC are the errors between the measured temperature and the desired temperature value and the integral of the error. The proposed SOFLC evaluates the extent of deviation of the process output form the desired input of the control system in terms of error and change in error, and it modifies the control rule base of the fuzzy controller. The experimental results show that the proposed technique improves the overall system response in the presence of step changes in torch travel speed or step changes in weldment thickness during GMA welding and, thus, guarantees the uniform weld quality.
