Spatiotemporal multi-sensor characterization of melt pool dynamics in laser directed energy deposition

Laser directed energy deposition (L-DED) technology has gained significant attention for its potential in manufacturing complex components. However, its industrial adoption is constrained by process fluctuations that cause variability in quality and mechanical properties. To address this challenge, a robot-assisted multi-sensor monitoring system (CMOS camera, NIR camera, microphone, and spectrometer) is developed, together with a software architecture for synchronized acquisition of heterogeneous signals. The key novelty of this research is a spatiotemporal feature fusion framework that enables mining of process data and comprehensive analysis of melt pool dynamic evolution. Correlation analysis is employed to elucidate the coupling mechanisms among multi-sensor signals, identifying key features and eliminating redundancies. The effects of laser scanning speed and power on melt pool geometry, temperature distribution, reaction intensity, and elemental excitation states are investigated, leading to quantitative correlations between process parameters and critical signal characteristics. The influence of process parameters on forming quality is assessed by scoring the single-bead deposits, from which a stable process window is established. This work reveals the quantitative relationships among process parameters, multi-sensor signal features, and L-DED formation quality. The findings provide prior knowledge for in-situ monitoring and future closed-loop control of the L-DED process.