AI-Powered Dent Analysis of Aircraft Engine Nacelles Using Altair PhysicsAI: A Lightweight Digital Twin Demonstrator

This paper investigates enhancing aircraft structural integrity and operational efficiency by addressing the challenge of rapidly evaluating impact-related dents in engine nacelle lip skins. The traditional Finite Element Analysis (FEA) process can be time-consuming due to a variety of factors including model setup, computational intensity, and post-processing. Our objective was to develop a fast, high-confidence method for structural diagnostics and stress prediction. We present a lightweight digital twin demonstrator model, developed using PhysicsAI®, which leverages Artificial Intelligence (AI) to serve as a surrogate model, enabling rapid stress estimation directly from Computer-Aided Design (CAD) inputs. The model, trained on an FEA dataset of forty dent configurations, successfully captures stress distribution patterns and correlates stress magnitudes with dent characteristics. The model can deliver predictions in under 20 seconds—a significant improvement over traditional FEA's typical 4-hour process. This demonstrator's compatibility with 3D scanning pipelines bridges digital inspection with virtual structural assessment, establishing a foundational step towards AI-powered Maintenance, Repair, and Overhaul (MRO) decision-making and realizing a digital twin vision for aviation maintenance.