(More Coming Soon)
(More Coming Soon)
This project develops simulation-informed machine learning frameworks for ultrasonic weld characterization under limited data conditions. By coupling reduced-order Lamb wave models with high-fidelity simulations and diffusion-based generative alignment, the framework enables robust defect inference even when real measurements differ from simulated training data. The approach emphasizes physics-constrained learning and uncertainty-aware inversion for deployable nondestructive evaluation pipelines.
Representative Papers:
Tempelman, J.R., Wachtor, A.J., Flynn, E.B. (2025). Machine learning-based ultrasonic weld characterization using hierarchical wave modeling and diffusion-driven distribution alignment. arXiv:2510.13023.
Maxwell, C., Tempelman, J.R., et al. (2025). Spatially-informed deep learning for full-field ultrasonic nondestructive evaluation. Under review.
This work formulates hyperspectral emissivity retrieval as a probabilistic inverse problem grounded in physics-based radiative transfer models. Using variational inference and generative modeling, the framework produces calibrated posterior distributions rather than point predictions, explicitly accounting for model–measurement mismatch. The result is interpretable, uncertainty-aware inference for remote sensing and materials characterization.
Representative Papers:
Tempelman, J.R., Mitchell, K., Wachtor, A.J., Flynn, E.B. (2025). Probabilistic emissivity retrieval from hyperspectral data via physics-guided variational inference. arXiv:2508.08291.
My doctoral research advanced nonlinear wave propagation and topological protection in mechanical lattices and metamaterials. I developed theoretical and experimental frameworks showing how strong nonlinearity modifies topological interface modes and enables intermodal targeted energy transfer. These results connect dynamical systems theory with next-generation vibration and acoustic metamaterials.
Representative Papers:
Tempelman, J.R., Matlack, K.H., Vakakis, A.F. (2021). Topological protection in a strongly nonlinear interface lattice. Physical Review B.
Tempelman, J.R., Vakakis, A.F., Matlack, K.H. (2023). Spectral Scattering and Inter-band Targeted Energy Transfer in Phononic Lattices. Physical Review E.
Gzal, M., Tempelman, J.R., et al. (2024). Subwavelength topological interface modes in vibroacoustic metamaterials. Frontiers in Acoustics.
This project integrates acoustic emissions, pyrometry, and machine learning to detect keyhole pore formation during laser powder-bed fusion. By combining signal processing with multimodal sensor fusion, the framework identifies defect signatures in real time and links process dynamics to resulting microstructure. The work contributes toward physics-guided, inline qualification pipelines for advanced manufacturing.
Representative Papers:
Tempelman, J.R., Wachtor, A.J., Flynn, E.B., et al. (2022). Detection of keyhole pore formations in LPBF using acoustic process monitoring. Additive Manufacturing.
Tempelman, J.R., Wachtor, A.J., Flynn, E.B., et al. (2022). Sensor Fusion of Pyrometry and Acoustic Measurements for Keyhole Pore Identification. Journal of Materials Processing Technology.
Ahmed, B., Tempelman, J.R., et al. (2024). Deep Learning to Predict Pore Formation during Additive Manufacturing. Scientific Reports.
This work investigates targeted energy transfer (TET) and nonlinear wave scattering in vibro-impact and strongly nonlinear systems. Through harmonic balance formulations and experimental validation, I demonstrated intermodal energy transfer mechanisms that enable broadband vibration mitigation without tuning to resonance. The research bridges nonlinear dynamics theory with experimentally validated mechanical systems.
Representative Papers:
Tempelman, J.R., Mojahed, A., et al. (2022). Experimental inter-modal targeted energy transfer in a cantilever beam. Journal of Sound and Vibration.
Tempelman, J.R., Matlack, K.H., Vakakis, A.F. (2024). Harmonic balance formulation for nonlinear wave-scattering clusters. International Journal of Nonlinear Mechanics.
This project applies topological data analysis (TDA) to identify bifurcations and dynamical regime changes in noisy nonlinear systems. By leveraging persistent homology and geometric signatures, the framework detects qualitative transitions without requiring explicit model forms. The work connects dynamical systems theory with modern data-driven analysis methods.
Representative Papers:
Tempelman, J.R., Khasawneh, F.A. (2020). A look into chaos detection through topological data analysis. Physica D.
Tanweer, S., Khasawneh, F.A., Munch, E., Tempelman, J.R. (2024). A Topological Framework for Identifying Phenomenological Bifurcations. Nonlinear Dynamics.