AI-assisted reverse engineering and process optimization of a titanium-based hybrid alloy (Ti–Al–Nb Composite) turbine blade using response surface methodology

The increasing demand for high-performance turbine blade materials that can withstand severe thermo-mechanical conditions necessitates the optimization of Ti–Al-Nb hybrid alloys with superior mechanical properties and oxidation resistance. Hence, the primary aim of this study is to investigate and optimize the thermo-mechanical behavior, mechanical properties, and oxidation response of Ti–Al-Nb hybrid alloys for advanced turbine blade materials. This research focuses on the thermo-mechanical behavior, mechanical properties, and oxidation response of Ti–Al-Nb hybrid alloys, which can be used for advanced turbine blades. FEA has revealed significant temperature gradients and stress concentrations across the blade, indicating that materials with high thermal stability, mechanical properties, and oxidation resistance are essential. The materials were evaluated experimentally for tensile strength, fracture toughness, hardness, and oxidation weight gain across Al and Nb compositions, heat treatment temperatures, and cooling rates. Response Surface Methodology (RSM) was used to quantify the effects of each factor, and a specialized artificial intelligence model, PNBM-SHAP-ZOA, was used to enhance the predictive accuracy of the results. The proposed AI model demonstrated superior prediction accuracy compared with the previously used ANN-GA, CNN-PSO, and traditional RSM, achieving the lowest prediction error for each response. Using RSM, the multi-response optimization produced optimal levels of 26.3% Al, 11.9% Nb, 850 °C maximum temperature, and 9.93 °C/s cooling rates, which all yielded excellent mechanical properties, good mechanical stability, and minimal weight gain from oxidation. The optimized alloy achieved a tensile strength of 1098.39 MPa, fracture toughness of 24.13 MPa√m, hardness of 380.52 HV, and oxidation weight gain of 0.16 mg/cm 2 . Overall, the results on the Ti–Al–Nb alloys showed they are promising materials for advanced turbine components that can withstand high-temperature loads, and the new intelligence framework demonstrated the ability to predict and optimize processes for materials with multiple properties.