Design Of Self-Adaptive Fuzzy Control Arc Sensor

In modern society, CO2 shielded welding can be seen everywhere, but hand welding can't meet the quality and quantity requirements of high-tech manufacturing process, it is necessary to develop and promote the automation of CO2 shielded welding. The process of short circuit transition is non-linear and with many influencing factors, and it is difficult to find an ideal model for control.As a branch of intelligent control, fuzzy control is based on human's thought and doesn't rely on mathematical model. And so it is a good method for auto-tracking in welding. In this thesis, an arc sensor based on fuzzy control is designed. Through the change of welding current in waving, the arc sensor judges the excursion and provides the adjusting output, meanwhile, it transmits the data acquired to a personal computer to be displayed and analyzed. Based on general fuzzy controller, the thesis gives how the controller's performance is influenced by factors of proportionality, membership function and defuzzification method. For overcoming the shortcomings of general fuzzy controller, rules self-adjusting fuzzy controller in whole domain is put forward, it can real-timely adjust the controlling rules, so it has some self-study ability, and relative simulation is done in MATLAB. Moreover, based on a MCU, the hardware and software are designed, and the feasibility and practicability of the self-adaptive fuzzy controller is proved by tests.