Key research themes
1. How do specific geometric features (convexities, concavities, intermediate points) influence mode shape recognition and perceptual coding of shapes?
This research area focuses on understanding which contour features of shapes—convexities, concavities, or intermediate points—are most salient or informative for shape recognition and mode shape coding. Investigations analyze how local curvature maxima and minima affect shape discriminability and whether certain features dominate in perceptual or neurological encoding. The outcome matters for theories of shape representation, computational modeling of shape recognition, and experimental design in vision sciences.
2. What computational and experimental methods enable accurate extraction and stitching of mode shapes from complex structures using optical and programmatic modeling approaches?
This theme examines advances in capturing, modeling, and reconstructing mode shapes of complex or large structures using computational frameworks, optical measurements, and programmatic shape representations. It addresses limitations of conventional accelerometer-based modal analysis, explores digital image correlation techniques for full-field, non-contact mode shape acquisition, and proposes procedural modeling with automated macro discovery to compactly represent mode shapes in 3D geometry programs. These methods have critical implications for structural health monitoring, design optimization, and generative modeling.
3. How do nonlocal elasticity theories and advanced mathematical modeling improve the computation of mode shapes and frequencies in complex engineered structures like nanobeams and plates?
This research theme centers on applying stress-driven nonlocal integral theories and analytical/numerical modeling techniques to accurately predict natural frequencies, mode shapes, and dynamic behaviors of engineered nanostructures and plates. It contrasts with classical local elasticity models by incorporating scale effects and material gradations, addressing ill-posedness in previous nonlocal approaches. The computational advances hold potential for optimizing design and performance in nanoelectromechanical systems and advanced materials.