Development Of Soft Switch Inverter Main Circuit And Dircet Short Signal Detection System Based On Wavelet Transform

It made a digitized waveform control soft-switching CO₂ weld machine in this paper. Soft switch inverter main circuit and the short circuit signal detection system of CO₂ welding based on the wavelet transform were made.Soft switch inverter main circuit of 20KHz in frequency based on IGBT is made. Firstly the equation among the component parameters of main circuit is calculated according to the requirement that the zero voltage in shifted leg is switched and the zero current in static leg is switched. The equation is the foundation of testifying the correction of the component parameters in main circuit. Under the background of MATLAB, the simulated model of inverter power is built based on the topology of full bridge inverter main circuit. From the equation, the parameters of component in simulated model were set, such as snubber capacitance, IGBT, series inductor, transformer, block capacitance. The rationality of the topology of soft switch full bridge inverter main circuit and the correction of the component parameters were testified by the simulation. And the experiment testified that soft switch in range of load is achieved and the main circuit has switch losses and less switch stress. All of that is the foundation of improvement the inverter frequency.Due to the specialty of the components, saturable inductor and transformer with little leakage were made, and they were tested met with the requirement by experiment.The short circuit and liquid bridge pinch signals can not be detected precisely, reliable and duly, so the DSP system based on wavelet detect short circuit signal was developed to improve. The probability and advantage of the short circuit signal detection of CO₂ welding system by wavelet were tested theoretically, which is the theoretical foundation of the DSP system. Based on FFT, the DSP spectrum analysis system of arc voltage is set. Pretreatment circuit of the short circuit signal detection of CO₂ welding is designed. The noise is filtered and the detail arc voltage signal is reserved with the circuit. Then the optimum digital filter group analyzing arc voltage is derived by the way wavelet tools in MATLAB. Finally the DSP detection system of short circuit based on wavelet is constructed, and the precision, reliability and real-time of the system is verified by experiment. The system can distinguish true and false short circuit signals, and provides precondition of wave control.