Study On Adaptive Fuzzy Control System Of CO₂ Arc Welding Inverter Source

CO2 arc welding has been widely applied to production practice due to high efficiency, energy saving, and low cost. But it exists big welding spatter and poor weld appearance. When the speed increases to 0.5m/min above. These two issues become more prominent and hinder the play of its high-speed and high-quality characteristics. To solve this problem, based on the characteristics of drople transfer, this paper puts forward a waveform control method in which the dual-slope control is used in short circuiting period and dual constant power control is used in arc period. A set of CO2 welding inverter waveform control system was designed based on the digital controller dsPIC30F6010 and experiments was conducted.In this paper, IGBT was used as the power switching devices, the main circuit is consist of full-bridge inverter and in which the limited bipolar control mode was used. Four PWM waveforms are generated by DSP chip using the internal PWM module, and incentives IGBT to turn-on. The dynamic characteristics of inverter power inverter is controlled by adjusting the duty cycle of inverter. Using adaptive fuzzy PI controller, current closed-loop control and Power Feedback Control were achieved separately in short circuiting period and in arc period. The control system was simulated by using MATLAB software, and the parameters of adaptive fuzzy PI controller are optimized by using genetic algorithm to make the system be in a state of optimal controlled status. Drive circuit, protection circuit and other peripheral circuit design were described in detail in the hardware. Welding main program, short/arcing subroutine, adaptive fuzzy PI controlling subroutine, A/D sampling subroutines as well as the initialization subroutine were programmed, and modular programming is adapted in software design.Debugging the welder offline and online, test results show that the welder basically meets the design requirements and achieves the external characteristics of constant power output. Building the experimental platform, an experimental program was adopted in which the dual-slope control is used in a short circuiting period and dual constant power control and constant voltage arc are used separately in arc period. Combination of weld shape, electrical signals both in the welding arc stability and energy were statisticsed by using Hanover Quality Analyzer and MATLAB software. The rule that the welding process stability and weld quality were affected by the two kinds of control mode was analyzed. Experimental results show that: when the welding speed of CO2 welding reaches 0.8m/min, because of the inverter's small output inductor, both the welding stability and the weld appearance are not very satisfied by using constant voltage control, but the method of power feedback can effectively improve the stability of output energy, so the uniformity of the transition period and welding quality are improved.