Fuzzy regression method for prediction and control the bead width in the robotic arc-welding process

In the manufacturing industry, quality is of prime importance and a reduction in rejects rates is desirable to reduce costs, improve productivity as well as satisfy customer quality requirements. Many quality problems arise due to poorly set process variables, and this is often due to lack of welding skill on the part of the mechanised or robotic system suppliers. However, the automated welding system has not been achieved much because of difficulties of the mathematical model and sensor technologies. In this paper, the possibilities of the fuzzy regression method in modeling the bead width in the robotic arc-welding process are presented. Fuzzy regression is a well-known method to deal with the problems with a high degree of fuzziness so that the approach is employed to build the relationship between four process variables and the quality characteristic, such as bead width. Using the developed model, the proper prediction of the process variables for obtaining the optimal bead width can be determined.