We presented a theoretical study for the structural, mechanical, and thermophysical properties of the precipitates in 2xxx series aluminum alloy by applying the widely used density functional theory of Perdew-Burke-Ernzerhof (PBE). The results indicated that the most thermodynamically stable structure refers to the Al₃Zr phase in regardless of its different polymorphs, while the formation enthalpy of Al₅Cu₂Mg₈Si₆ is only -0.02 eV (close to zero) indicating its metastable nature. The universal anisotropy index of A^U follows the trend of: Al₂Cu > Al₂CuMg ≈ Al₃Zr_D0₂₂ ≈ Al₂₀Cu₂Mn₃ > Al₃Fe ≈ Al₆Mn > Al₃Zr_D0₂₃ ≈ Al₃Zr_L1₂ > Al₇Cu₂Fe > Al₃Fe₂Si. The thermal expansion coefficients (TECs) were calculated based on Quasi harmonic approximation (QHA); Al₂CuMg shows the highest linear thermal expansion coefficient (LTEC), followed by Al₃Fe, Al₂Cu, Al₃Zr_L1₂ and others, while Al₃Zr_D0₂₂ is the lowest one. The calculated data of three Al₃Zr polymorphs follow the order of L1₂ > D0₂₃ > D0₂₂, all of them show much lower LTEC than Al substance. For multi-phase aluminum alloys, when the expansion coefficient of the precipitates is quite different from the matrix, it may cause a relatively large internal stress, or even produce cracks under actual service conditions. Therefore, it is necessary to discuss the heat misfit degree during the material design. The discrepancy between a-Al and Al₂CuMg is the smallest, which may decrease the heat misfit degree between them and improve the thermal shock resistant behaviors.

Vision-based laser penetration control has become an important research area in the field of welding quality control. Due to the complexity and large number of parameters in the monitoring model, control of the welding process based on deep learning and the reliance on long-term information for penetration identification are challenges. In this study, a penetration recognition method based on a two-stage temporal convolutional network is proposed to realize the online process control of laser welding. In this paper, a coaxial vision welding monitoring system is built. A lightweight segmentation model, based on channel pruning, is proposed to extract the key features of the molten pool and the keyhole from the clear molten pool keyhole image. Using these molten pool and keyhole features, a temporal convolutional network based on attention mechanism is established. The recognition method can effectively predict the laser welding penetration state, which depends on long-term information. In addition, the penetration identification experiment and closed-loop control experiment of unequal thickness plates are designed. The proposed method in this study has an accuracy of 98.96% and an average inference speed of 20.4 ms. The experimental results demonstrate that the proposed method exhibits significant performance in recognizing the penetration state from long sequences of welding image signals, adjusting welding power, and stabilizing welding quality.