Information Analysis And Decoupling Control For Pulse MIG Welding Process Of Aluminum Alloy

In recent years, with the wide application of aluminum alloy in automobiles, trains and other means of vehicle, requirements of automation,intelligence and welding quality stability in aluminum alloy welding structures are higher and higher. In this paper, according to the key problems such as obvious heat accumulation phenomenon at constant parameters in pulsed MIG welding of aluminum alloy and unstable welding process when parameters do not match, further study on aluminum alloy welding control was done through sensing and analysis of muti-infornmation, process modeling and simulation and real-time decoupling control in welding process.Due to many uncertain factors in pulsed MIG welding process of aluminum alloy, real-time detection and control is an important method to ensure the stability of welding process and welding quality. On the basis of vision sensing realized through combination of LabVIEW and COM technology, a rapid prototyping control platform was built with xPC target real-time environment.In pulsed MIG welding of aluminum alloy, an extraction algorithm for wire extension was proposed by method of vision sensing and welding pool edge extraction was also studied using ant colony algorithm and genetic algorithm aiming at the deficiency of traditional differential operator. Analysis on stability of aluminum alloy pulsed MIG weldig process was made, and the correlation between probability density, approximate entropy value of arc voltage and the stability of welding process was obtained. In order to overcome the characterization deficiency of existing signals for welding process, sound signal in welding process was acquired, and the characteristic of arc sound signal was analyzed under different metal transfer mode to identify the mode of metal transfer, and then the relationship between welding arc sound and subsidence of aluminum alloy MIG welding bead was also studied using wavelet analysis method. Furthermore, muti-information fusion was preliminarily studied, and the correlation between characteristics after fusion of welding current, voltage and stability of welding process was established by method of U-I two-dimensional phase space, two-dimensional approximate entropy and algorithm of multi-information fusion.In this paper, analysis on MIMO control model was made on the basis of existing process identification for pulsed MIG welding of aluminum alloy and a decoupling control scheme using pool width and wire extension as control object while using wire feed speed and duty cycle as controlled variable was selected. Simulation using classical theory of decoupling control and neural network object inverse model theory was carried out. Aiming at the characteristics of arc system and on the basis of considering the droplet transfer, dynamic arc model of wire melting was built, and the actual welding process was well reflected by simulation of the model. The reasons for the instability of arc system was also revealed. Further, wire extension control simulation was made on the built model and provided a reference for arc stability control.On the basis of sensing and simulation, firstly, wire extension was independently controlled by adjusting wire feed speed to make arc system stable on the rapid prototyping platform. Then with double-pulse technology, heat input was changed for pool width control by regulating double-pulse duty factor. On this basis, MIMO real-time decoupling control was carried out using wire extension and pool width as control object while using wire feed speed and double-pulse duty factor as controlled variable. The control effect under different sensing methods and different controllers was also studied based on the rapid prototyping control platform. Adopting vision sensing for pool width and wire extension, decoupling control with fuzzy PID controller was carried out and obtained a stable welding process and a perfect weld, whose pool width was uniform and formation was beautiful. Therefore, the heat accumulation problem was resolved well by overcoming the strong coupling among parameters when controlling in pulsed MIG welding process.