This thesis presents an on-line quality monitoring and control method to obtain uniform weld quality in gas metal are welding (GMAW) processes. The geometrical parameters of the weld pool such as the top bead width and the penetration depth plus half back width are utilized to assess the integrity of the weld quality. Monitoring of these geometrical parameters is very important for on-line process control as well as for on-line quality evaluation. It is, however, an extremely difficult problem because of the inherent characteristics of the welding process.
The monitoring variables used are the surface temperatures measured at various points on the top surface of the weldment which are strongly related to the formation of the weld pool. The surface temperatures are measured using infra-red temperature sensing system. The relationship between the measured temperatures and the weld pool size is implemented on the multilayer perceptrons which are powerful for realization of complex mapping characteristics. The main task of the neural network is to realize the mapping characteristics from the point temperatures to the weld pool sizes through training. After training, the neural estimator can estimate the weld pool sizes from the leamed mapping characteristics. The design parameters of the neural network estimator such as the number of hidden layers and the number of nodes in a layer, are chosen based upon an estimation error analysis. A series of bead-on-plate welding experiments was performed to assess the performance of the neural network estimator. The experimental results show that the proposed neural network estimator can estimate the weld pool sizes with satisfactory accuracy.
To estimate the weld pool sizes in the region of transient states, the time history of the surface temperatures is used as the input to the neural estimator. Various types of the input to the neural estimator such as the type of input information and the number of terms in the time sequence of the surface temperatures are tested. Estimation of the weld pool sizes in the region of transient states is very important for on-line quality monitoring and control.
The control purpose is to obtain uniform weld quality in GMA welding process. In this research, the weld pool size is directly regulated to a desired one. The proposed controller is composed of the neural pool size estimator, the neural feedforward controller and the feedback controller. The pool size estimator predicts the weld pool size under growing. The feedforward controller compensates for the nonlinear characteristics of the welding process. The simulation study and experimental results show that the proposed control method improves the overall system response in the presence of changes in torch travel speed during GMA welding and guarantees the uniform weld quality.