Collaborative control of deposition width and height for thick-walled parts in arc-directed energy deposition via integrating arc voltage and vision sensing

Gas tungsten arc-directed energy deposition (GTA-DED) offers significant potential for producing large-size metal components. However, the precise automatic control of process stability and dimensional accuracy of components with multi-layer multi-bead characteristics in GTA-DED remains a huge challenge. This study develops a collaborative control system for deposition width and height in GTA-DED of thick-walled parts. An arc voltage (AV) and vision collaborative sensing framework is established. The peak AV signal is filtered via an ant colony optimization-based wavelet threshold algorithm to monitor the deposition height stability. The image processing algorithm is applied to extract the deposition width from molten pool images captured at base current periods. Three peak AV-arc length-current relationship models are constructed based on the arc morphologies under different deposition positions. The AV signal is used to calibrate the vision sensing system for the accurate monitoring of deposition width, and the width monitoring error is less than 0.04 mm. A model reference adaptive controller is developed to adjust the wire feeding speed for improving the deposition height stability, while a self-tuning fuzzy controller is designed to regulate the peak current for maintaining deposition width uniformity. The total height and width deviations of the thick-walled parts are no more than 0.5 mm. The present work provides important guidance and feasible solutions for promoting the automated manufacturing level and forming accuracy in GTA-DED of thick-walled parts.