A quality monitoring system in butt welding process is proposed by estimating weld pool sizes. Although many kind of welding control system were developed, it is difficult to measure the weld quality directly and to applicate the practical welding system.
The proposed system measured temperatures due to weldment surfase temperatures is related to weld pool size. the sensor system consist of two pyrometers which are located symmetric to torch center to get the information about the alignment of butt joint shape. The sensor detects infrared emission from the weldment and converts it to voltage signal. This was calibrated using by thermocouple to assure the reliability of sensor signal. The result of the calbration was showed the excellent characteristic for the high temperature. So the range of measured temperature is 400℃ to 1200℃. The output signal of the sensor was processed through the filter, Because measuring the temperatures can be affected by spatters and welding arc.
A neural network estimator is designed to estimate weld pool size from surface temperatures measured various points on the top surface of the weldment. Butt welding experiments were 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