Technology Category
Search and Optimization: Sampling/Simulation-based SearchMachine Learning: OptimizationReasoning under Uncertainty: Stochastic Optimization
When AI Bends Metal: AI-Assisted Optimization of Design Parameters in Sheet Metal Forming
🤖 AI Summary
In sheet metal forming, design parameter optimization suffers from heavy reliance on expert knowledge, high computational cost, and prolonged iteration cycles. To address these challenges, this paper proposes an AI-driven automated optimization workflow. Methodologically, it integrates deep learning—employed for initial parameter prediction and surrogate model construction—with active-learning-enhanced Bayesian optimization for efficient, uncertainty-guided sampling, all coupled with high-fidelity numerical simulation. The key contributions are: (1) the first integration of active learning into the Bayesian optimization framework to accelerate convergence; and (2) leveraging deep learning to provide robust initial parameter estimates, substantially reducing dependence on prior domain expertise. Evaluated on industrial-scale case studies, the method reduces the number of required simulations by over 40% and decreases computational resource consumption by approximately 35%, demonstrating its efficiency, robustness, and engineering scalability.
Technology Category
Application Category
📝 Abstract
Numerical simulations have revolutionized the industrial design process by reducing prototyping costs, design iterations, and enabling product engineers to explore the design space more efficiently. However, the growing scale of simulations demands substantial expert knowledge, computational resources, and time. A key challenge is identifying input parameters that yield optimal results, as iterative simulations are costly and can have a large environmental impact. This paper presents an AI-assisted workflow that reduces expert involvement in parameter optimization through the use of Bayesian optimization. Furthermore, we present an active learning variant of the approach, assisting the expert if desired. A deep learning model provides an initial parameter estimate, from which the optimization cycle iteratively refines the design until a termination condition (e.g., energy budget or iteration limit) is met. We demonstrate our approach, based on a sheet metal forming process, and show how it enables us to accelerate the exploration of the design space while reducing the need for expert involvement.
Problem
Research questions and friction points this paper is trying to address.
AI-assisted workflow reduces expert involvement in parameter optimization
Deep learning model provides initial parameter estimate for iterative refinement
Accelerates design space exploration while reducing need for expert knowledge
Innovation
Methods, ideas, or system contributions that make the work stand out.
AI-assisted workflow using Bayesian optimization
Deep learning model for initial parameter estimation
Active learning variant to assist experts
🔎 Similar Papers
No similar papers found.
