First Contact: Data-driven Friction-Stir Process Control

2025-07-03

AI Summary

To address the slow thermal response and poor controllability of tool temperature during the plunging phase in friction stir processing (FSP), this paper proposes a data-driven open-loop setpoint control method integrating neural networks with lumped-parameter differential equations (LPDEs). For the first time, a differentiable and interpretable temperature prediction model is constructed by embedding a neural network into the LPDE framework—bridging purely data-driven and purely physics-based modeling. Based on this hybrid model, an optimal plunging trajectory is synthesized to enable rapid and precise convergence of tool temperature to the target value. Experimental validation demonstrates that the method reduces steady-state temperature error to within ±2.3°C and shortens convergence time by 40%, significantly enhancing thermal–mechanical consistency. This work establishes a closed-loop paradigm of “interpretable modeling → dynamic prediction → open-loop control,” offering a novel pathway for adaptive thermal management in FSP.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

This study validates the use of Neural Lumped Parameter Differential Equations for open-loop setpoint control of the plunge sequence in Friction Stir Processing (FSP). The approach integrates a data-driven framework with classical heat transfer techniques to predict tool temperatures, informing control strategies. By utilizing a trained Neural Lumped Parameter Differential Equation model, we translate theoretical predictions into practical set-point control, facilitating rapid attainment of desired tool temperatures and ensuring consistent thermomechanical states during FSP. This study covers the design, implementation, and experimental validation of our control approach, establishing a foundation for efficient, adaptive FSP operations.

Problem

• Validates Neural Lumped Parameter Differential Equations for FSP control
• Integrates data-driven and heat transfer methods for temperature prediction
• Enables rapid tool temperature control for consistent FSP outcomes

Innovation

• Neural Lumped Parameter Differential Equations for control
• Data-driven framework with heat transfer techniques
• Practical set-point control for temperature stability