In-situ multi-eye monitoring of melt pool temperature field in laser additive manufacturing by light field camera

Laser direct energy deposition (LDED) and laser powder bed fusion (LPBF) are two typical metal laser-based additive manufacturing (AM) processes used in critical fields such as aerospace and aviation. However, the stability of their part quality remains challenging. The temperature of the melt pool during the AM process significantly influences the quality of the manufactured parts. Therefore, breakthroughs in in-situ monitoring technology for high-temperature, small-area melt pools keep an urgent need. To address this challenge, this paper proposes a multi-eye monitoring method using a light field (LF) camera for in-situ melt pool temperature field monitoring. Initially, a LF sub-aperture Bayer model (LFSBM) is established to extract melt pool images at red, green, and blue (R, G, and B) wavelengths. By calibrating the LF camera’s relative spectral response ratio using the blackbody furnace, the melt pool’s temperature field is derived based on dual-wavelength theory from two images at R, G, and B channels. Linear fitting of the relative spectral response ratio for channel combinations of B and G, R and B, and R and G yielded root mean square errors of 76.34 K, 62.24 K, and 78.66 K, respectively. The mean error for maximum temperature was verified to be 1.03 %, and less than 3 % for temperature filed at temperatures of 2973.15 K, 3073.15 K, and 3273.15 K by the blackbody furnace. The influence of coaxial system of LPBF on wavelength intensity was calibrated. The contour error between the temperature map and the blackbody furnace was found to be less than 1.4 %. Experiments were conducted on high-entropy alloy, and Ti6Al4V alloys manufactured by both LDED and LPBF equipment, and evolution of length, width, and maximum temperature were analyzed. The proposed method simplifies the measurement process and allows for an unlimited temperature range, providing a groundbreaking approach for in-situ melt pool temperature monitoring during the AM process.