Prediction algorithm for springback of frame-rib parts in rubber forming process by incorporating Sobol within improved grey relation analysis

Abstract

A precise springback prediction is of great significance to reduce the manufacturing time and also to improve the forming quality for frame-rib parts. Based on Sobol and improved grey relation analysis, a novel prediction algorithm for springback of frame-rib parts in rubber forming process is proposed in this paper. First of all, the mapping model between springback angle and geometric parameters of the flange is established by using the response surface method and experiment. With the obtained mapping model, the weight of each parameter on springback is calculated by Sobol global sensitivity analysis method. Afterwards, a novel method for springback prediction is presented by incorporating k nearest-neighbor within improved grey relation analysis. A case study on springback angle prediction for the flange of rib part is given to validate the applicability and effectiveness of the proposed approach. Furthermore, by comparing the predicted values with the actual value of 18 cases, the high accuracy and stability of the proposed algorithm are further demonstrated.

Keywords

• Sensitivity analysis
• Springback angle prediction
• Grey relation analysis
• Hybrid similarity algorithm
• Frame-rib parts