Intelligent prediction of forming deformations for marine titanium alloy curved plates based on the beluga whale optimization algorithm

View at publisher Summarise with VIRAT
Authors: Shun Wang, Jiayan Wang, Ji Wang, Rui Li
Cited by: 0
References: 20

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

To predict the forming of titanium alloy more accurately, the deformation prediction of titanium alloy based on convolutional neural network (CNN) and backpropagation neural network (BPNN) were established. The feasibility of the titanium alloy line heating model is verified by comparing with the simulation and experiment of mild steel. Based on titanium alloy simulation model, 101 sets of TA5 titanium alloy forming parameters were obtained and solved. CNN and BPNN were applied to predict shrinkage. The prediction performance of BPNN is better than that of CNN. To improve the prediction accuracy of the BPNN, the BPNN was optimized using the Beluga Whale Optimization (BWO) algorithm. The MAPE of BWO-BPNN in shrinkage and deflection prediction were 2.82% and 2.57%, respectively. Compared to the BPNN, the BWO-BPNN reduced the MAPE of the shrinkage and deflection predicted values by 61.55% and 68.89%, respectively. The relative errors of the BWO-BPNN are all less than 5%.

Fields of Study & Subjects

Laser and Thermal Forming Techniques Metal Forming Simulation Techniques Metallurgy and Material Forming

Publication details

Journal / source
Ships and Offshore Structures
Pages
1–16
Publication date
30 March 2026
Language
EN