Digital twin-driven design for elevator fairings via multi-objective optimization

Traditional geometry optimization of elevator fairings is only based on computational fluid dynamics simulations to find optimal structure parameters, and a large volume of data generated during the elevator operation is not utilized to optimize elevator fairings collaboratively. This paper proposes a digital twin-driven design framework to design the elevator fairing of the next generation. A digital twin model corresponding to the real elevator is first established via a computing platform, and a multi-objective optimization method like neighborhood cultivation genetic algorithms is employed to optimize the elevator fairing design. The effectiveness of the digital twin-driven design framework is demonstrated by the elevator fairing design, and the results show that compared with the unoptimized elevator fairing, the air drag and the lateral force are lowered by 18.1% and 11.2%, respectively, and the turbulence coefficient decreases from 0.478 to 0.451 after optimizing the elevator fairing.