Bezier Curve-Based Shape Knowledge Acquisition and Fusion for Surrogate Model Construction

Surrogate model technology is a key technology in the field of engineering design with limited data. Fusion of engineering knowledge into surrogate models is an effective method to improve the prediction accuracy. However, engineering knowledge in this field describes the complex relationship between variables, which makes it difficult to obtain quantitative knowledge. Therefore, the engineering knowledge acquisition and fusion technology based on Bezier Curve for complex equipment design was proposed, which covered the entire process from knowledge acquisition to filtering and fusion. Finally, through the verification of the Unmanned Vehicle Truss design case and test functions, the experimental results show that the technology can achieve the effective acquisition of complex curve knowledge and represent multi-knowledge information effectively.