Laser cladding powder flow field detection system based on ISR optimization algorithm

Abstract

In coaxial powder feeding laser cladding, the morphology of the powder flow field is crucial for the forming quality. Therefore, this study utilizes high-speed imaging technology and an Image Super Resolution algorithm to create a laser cladding powder flow field detection system that is capable of detecting and tracking powder particles in the laser cladding environment. The experiment shows that the optimized algorithm has significant improvement in structural similarity indicators, with an improvement rate of nearly 11 %. For powder particle tracking, the distance accuracy of the optimized model is 1.5 lower than that of the unimproved model. In addition, by combining with the Kalman filtering algorithm, the tracking effect of powder particles has been further improved. This paper also found a relationship between powder transfer rate and powder utilization rate. In summary, the powder flow field analysis based on visual detection and image processing technology designed in this study can effectively reflect and predict the trend of changes in cladding quality.