Niell Elvin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Niell Elvin has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Andrei Fershalov
Department of Mechanical Engineering, The City College of New York , New York, New York 10031
3 shared publications - Yang Liu
Department of Mechanical Engineering, The City College of New York , New York, New York 10031
3 shared publications Alex Elvin
School of Civil and Environmental Engineering, University of the Witwatersrand, Johannesburg, South Africa
2 shared publications- Daniel Struk
Department of Mechanical Engineering, The City College of New York , New York, New York 10031
2 shared publications - Feridun Delale
Department of Mechanical Engineering, The City College of New York
2 shared publications - Pieter Orlandini
Department of Mechanical Engineering, The City College of New York , New York, New York 10031
2 shared publications
Latest Publications
- 2 September 2026
- Published by Taylor & Francis in Journal of Adhesion Science and Technology
- p. 1-21
Performance of multi-material adhesively bonded joints under quasi-static shear loading
- 2 September 2026
- Published by Taylor & Francis in Journal of Adhesion Science and Technology
- p. 1-21
Performance of multi-material adhesively bonded joints under quasi-static shear loading
- Open Access
- 3 August 2026
- Published by MDPI AG in Journal of Composites Science
- Vol. 10 (8) , 412
Machine Learning-Based Static Performance Prediction of Bonded Structural Patch Repairs
This study investigates adhesively bonded composite patch repair to enhance the load-carrying capacity of damaged metallic structures, introducing a novel FE-augmented machine learning (ML) framework that addresses the limited availability of experimental data in structural repair applications. To evaluate this approach, aluminum and steel specimens with central fatigue cracks were repaired using glass-fiber/epoxy and carbon-fiber/epoxy composite patches and tested under quasi-static loading at room (), high (), and low (−) temperatures. Finite element (FE) models were then developed in ABAQUS© to predict the failure loads of the patched specimens under varying temperature conditions, showing excellent agreement with the experimental data. The high accuracy of the FE predictions enabled their use as additional training data, effectively augmenting the limited experimental dataset and allowing the development of more robust regression models. Ten machine learning (ML) regression models, including linear regression (LR), polynomial regression (PR), support vector regression (SVR), random forest (RF), gradient boosting (GB), XGBoost (XGB), LightGBM (LGBM), Gaussian process (GP) regression, artificial neural networks (ANNs), and Kolmogorov–Arnold networks (KANs), were trained to predict the failure load of both unpatched and patched specimens as a function of material type, temperature, specimen thickness, crack length, and, for patched specimens, patch type and thickness. The datasets combined a limited set of physical results (75 patched samples: 63 experimental and 12 finite-element; 72 unpatched samples: 27 experimental and 45 theoretical) with Gaussian-mixture-model synthetic samples used only to augment the training data up to 300 samples per case. Under a configuration-grouped, leakage-free nested cross-validation (entire configurations held out for testing, hyperparameters tuned on inner folds only), the best models predicted the failure load of unseen configurations with mean absolute percentage errors of 2.78% (Gradient Boosting, patched, ) and 3.33% (Gaussian Process, unpatched, ). A paired ablation showed that Gaussian-mixture-model augmentation did not improve accuracy and, for several models, actually reduced it; the final models therefore rely on the real multi-source (experimental, FE, and theoretical) data, with the synthetic pipeline reported as a validated but non-beneficial component for these datasets. Overall, this study provides a novel, data-efficient framework combining experimental testing, FE simulation, and validated regression modeling to predict the performance of adhesively bonded composite patch repairs under varying thermal and mechanical conditions.- 3 August 2026
- Published by MDPI AG in Journal of Composites Science
- Vol. 10 (8) , 412
Open AccessMachine Learning-Based Static Performance Prediction of Bonded Structural Patch Repairs
This study investigates adhesively bonded composite patch repair to enhance the load-carrying capacity of damaged metallic structures, introducing a novel FE-augmented machine learning (ML) framework that addresses the limited availability of experimental data in structural repair applications. To evaluate this approach, aluminum and steel specimens with central fatigue cracks were repaired using glass-fiber/epoxy and carbon-fiber/epoxy composite patches and tested under quasi-static loading at room (), high (), and low (−) temperatures. Finite element (FE) models were then developed in ABAQUS© to predict the failure loads of the patched specimens under varying temperature conditions, showing excellent agreement with the experimental data. The high accuracy of the FE predictions enabled their use as additional training data, effectively augmenting the limited experimental dataset and allowing the development of more robust regression models. Ten machine learning (ML) regression models, including linear regression (LR), polynomial regression (PR), support vector regression (SVR), random forest (RF), gradient boosting (GB), XGBoost (XGB), LightGBM (LGBM), Gaussian process (GP) regression, artificial neural networks (ANNs), and Kolmogorov–Arnold networks (KANs), were trained to predict the failure load of both unpatched and patched specimens as a function of material type, temperature, specimen thickness, crack length, and, for patched specimens, patch type and thickness. The datasets combined a limited set of physical results (75 patched samples: 63 experimental and 12 finite-element; 72 unpatched samples: 27 experimental and 45 theoretical) with Gaussian-mixture-model synthetic samples used only to augment the training data up to 300 samples per case. Under a configuration-grouped, leakage-free nested cross-validation (entire configurations held out for testing, hyperparameters tuned on inner folds only), the best models predicted the failure load of unseen configurations with mean absolute percentage errors of 2.78% (Gradient Boosting, patched, ) and 3.33% (Gaussian Process, unpatched, ). A paired ablation showed that Gaussian-mixture-model augmentation did not improve accuracy and, for several models, actually reduced it; the final models therefore rely on the real multi-source (experimental, FE, and theoretical) data, with the synthetic pipeline reported as a validated but non-beneficial component for these datasets. Overall, this study provides a novel, data-efficient framework combining experimental testing, FE simulation, and validated regression modeling to predict the performance of adhesively bonded composite patch repairs under varying thermal and mechanical conditions. - 1 July 2026
- Published by AIP Publishing in Physics of Fluids
- Vol. 38 (7)
Aerodynamic coupling and vibration modes in vortex-induced vibrations of side-by-side cylinder arrays
Wind-tunnel experiments were conducted to investigate vortex-induced vibrations (VIVs) of side-by-side arrays of two and three identical circular cylinders mounted on flexible beams. The gap-to-diameter ratio was varied over 0.025≤G/D≤4. At large spacings, the response approached that of an isolated oscillator: 1≤G/D≤4 for two cylinders and 2≤G/D≤4 for three cylinders. At smaller spacings, wake coupling modified the lock-in response and produced distinct collective vibration modes. In the two-cylinder array, the cylinders oscillated either synchronously or 180° out of phase. In the three-cylinder array, additional states appeared, including cases where the two outer cylinders moved together while the middle cylinder moved oppositely and cases where the middle cylinder remained nearly stationary while the outer cylinders vibrated 180° out of phase. With increasing reduced velocity, these states transitioned toward in-phase motion. Decreasing G/D below 0.5 substantially increased both vibration amplitude and lock-in bandwidth, with maximum amplitudes reaching approximately three and four times the isolated-cylinder value for the two- and three-cylinder arrays, respectively. These results demonstrate that spacing-controlled wake coupling can amplify VIV response and broaden the operational bandwidth of multi-cylinder oscillator arrays for flow-energy harvesting applications.
- 1 July 2026
- Published by AIP Publishing in Physics of Fluids
- Vol. 38 (7)
Aerodynamic coupling and vibration modes in vortex-induced vibrations of side-by-side cylinder arrays
Wind-tunnel experiments were conducted to investigate vortex-induced vibrations (VIVs) of side-by-side arrays of two and three identical circular cylinders mounted on flexible beams. The gap-to-diameter ratio was varied over 0.025≤G/D≤4. At large spacings, the response approached that of an isolated oscillator: 1≤G/D≤4 for two cylinders and 2≤G/D≤4 for three cylinders. At smaller spacings, wake coupling modified the lock-in response and produced distinct collective vibration modes. In the two-cylinder array, the cylinders oscillated either synchronously or 180° out of phase. In the three-cylinder array, additional states appeared, including cases where the two outer cylinders moved together while the middle cylinder moved oppositely and cases where the middle cylinder remained nearly stationary while the outer cylinders vibrated 180° out of phase. With increasing reduced velocity, these states transitioned toward in-phase motion. Decreasing G/D below 0.5 substantially increased both vibration amplitude and lock-in bandwidth, with maximum amplitudes reaching approximately three and four times the isolated-cylinder value for the two- and three-cylinder arrays, respectively. These results demonstrate that spacing-controlled wake coupling can amplify VIV response and broaden the operational bandwidth of multi-cylinder oscillator arrays for flow-energy harvesting applications.
- 1 September 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (9)
The effect of endplates on vortex-induced vibration of finite-length cylinders
This study experimentally investigates the three-dimensional (3-D) effects of a cylinder oscillator energy harvester system in terms of vortex-induced vibration (VIV). The 3-D effects were modulated by varying endplate size. Specifically, these effects were evaluated by examining the influence of different endplate sizes on aerodynamic forces and vibrational amplitudes of the finite-length cylinder oscillator. Two configurations were studied: a rigidly mounted stationary cylinder as a reference, and a flexibly mounted cylinder oscillating transversely in airflow, both with the same diameter. The Reynolds number range was ∼5300–21 300. Adding circular endplates to the stationary cylinder significantly increased lift fluctuations, while an endplate diameter three times the cylinder was sufficient to suppress spanwise end effects. These findings informed the VIV experiments, where bare and endplate (various diameters) configurations were evaluated under matched structural parameters. While the addition of endplates did not alter the lock-in bandwidth, they strongly influenced aerodynamic forces and vibration amplitude within the lock-in regime. Smaller endplates (two times diameter) enhanced peak vibration amplitude, despite added mass, indicating a net gain in energy harvesting. In contrast, larger endplates (three times diameter) generally reduced vibration amplitude and induced earlier transition to the lower VIV branch, despite producing the highest lift coefficients. Phase and frequency analysis revealed configuration-specific differences, especially beyond an amplitude ratio of 0.3. In the post-lock-in regime, organized vortex shedding persisted only for the larger endplate cylinder, but was absent for both the bare and smaller-endplate cylinders.
- 1 September 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (9)
The effect of endplates on vortex-induced vibration of finite-length cylinders
This study experimentally investigates the three-dimensional (3-D) effects of a cylinder oscillator energy harvester system in terms of vortex-induced vibration (VIV). The 3-D effects were modulated by varying endplate size. Specifically, these effects were evaluated by examining the influence of different endplate sizes on aerodynamic forces and vibrational amplitudes of the finite-length cylinder oscillator. Two configurations were studied: a rigidly mounted stationary cylinder as a reference, and a flexibly mounted cylinder oscillating transversely in airflow, both with the same diameter. The Reynolds number range was ∼5300–21 300. Adding circular endplates to the stationary cylinder significantly increased lift fluctuations, while an endplate diameter three times the cylinder was sufficient to suppress spanwise end effects. These findings informed the VIV experiments, where bare and endplate (various diameters) configurations were evaluated under matched structural parameters. While the addition of endplates did not alter the lock-in bandwidth, they strongly influenced aerodynamic forces and vibration amplitude within the lock-in regime. Smaller endplates (two times diameter) enhanced peak vibration amplitude, despite added mass, indicating a net gain in energy harvesting. In contrast, larger endplates (three times diameter) generally reduced vibration amplitude and induced earlier transition to the lower VIV branch, despite producing the highest lift coefficients. Phase and frequency analysis revealed configuration-specific differences, especially beyond an amplitude ratio of 0.3. In the post-lock-in regime, organized vortex shedding persisted only for the larger endplate cylinder, but was absent for both the bare and smaller-endplate cylinders.
- 1 April 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (4)
Harvesting sustainable energy through vortex-induced vibrations of finite length cylinder
Vortex-induced vibration (VIV) has emerged as a promising method for small-scale energy harvesting. This research explores the key parameters affecting VIV in a cylinder-cantilever beam system within a Reynolds number range of 400–7500. The investigation focused on identifying the airflow velocity thresholds that initiate vibrations, measuring peak vibration amplitudes, and determining the critical airflow velocities where vibrations are maximized. By systematically varying mass, stiffness, and cylinder diameter, we examined their distinct effects on system behavior. Key outcomes indicate that larger cylinder diameters lead to increased vibration amplitudes and broader operational bandwidths, while adding mass reduces the bandwidth. Higher stiffness boosts both the maximum amplitude and bandwidth, shifting these to higher airflow velocities. The lock-in regime was observed to initiate at a Strouhal number (St) between 0.175 and 0.197, with vibration cessation occurring at an approximately consistent Strouhal number for each cylinder diameter. The peak vibration amplitude occurred at St ≈ 0.16, with fluctuations of less than 5% across all models. Additionally, the wake structure behind the cylinder and its behavior across the vibration bandwidth were analyzed using flow visualization techniques. A hot-wire anemometer positioned downstream measured velocity fluctuations from vortex shedding. These findings offer practical insights for optimizing VIV-based energy harvesting, linking wake behavior to amplitude response and power output. This study contributes to the broader understanding of VIV energy harvesters and provides a foundation for validating numerical models and enhancing the efficiency of sustainable energy systems.
- 1 April 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (4)
Harvesting sustainable energy through vortex-induced vibrations of finite length cylinder
Vortex-induced vibration (VIV) has emerged as a promising method for small-scale energy harvesting. This research explores the key parameters affecting VIV in a cylinder-cantilever beam system within a Reynolds number range of 400–7500. The investigation focused on identifying the airflow velocity thresholds that initiate vibrations, measuring peak vibration amplitudes, and determining the critical airflow velocities where vibrations are maximized. By systematically varying mass, stiffness, and cylinder diameter, we examined their distinct effects on system behavior. Key outcomes indicate that larger cylinder diameters lead to increased vibration amplitudes and broader operational bandwidths, while adding mass reduces the bandwidth. Higher stiffness boosts both the maximum amplitude and bandwidth, shifting these to higher airflow velocities. The lock-in regime was observed to initiate at a Strouhal number (St) between 0.175 and 0.197, with vibration cessation occurring at an approximately consistent Strouhal number for each cylinder diameter. The peak vibration amplitude occurred at St ≈ 0.16, with fluctuations of less than 5% across all models. Additionally, the wake structure behind the cylinder and its behavior across the vibration bandwidth were analyzed using flow visualization techniques. A hot-wire anemometer positioned downstream measured velocity fluctuations from vortex shedding. These findings offer practical insights for optimizing VIV-based energy harvesting, linking wake behavior to amplitude response and power output. This study contributes to the broader understanding of VIV energy harvesters and provides a foundation for validating numerical models and enhancing the efficiency of sustainable energy systems.
- Open Access
- 12 August 2020
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 31 (20) , 2378-2389
Non-linear piezoelectric fluidic energy harvesters: The mutual interaction of two oscillating cylinders
The presence of a bluff body upstream of a cantilever beam promotes persistent, aero-elastic vibrations of the beam. Vortex-induced vibration in an array of two mutually interacting bluff bodies in such configurations undergoing two-degrees of freedom transverse oscillation has not been investigated before. In the present work, we have studied experimentally, the unsteady response of an array of two similar rigid cylinders, positioned side-by-side in reference to the freestream velocity, each one mounted on the upstream end of an elastic cantilever beam. By fitting the beams with piezoelectric layers, these configurations are converted to piezoelectric fluid energy harvesters (PFEH) that can extract small amounts of energy from the flow. Comparing the performance of linear (L-PFEH), non-linear (NL-PFEH), and a non-linear array (NLA-PFEH) of harvesters show that NLA-PFEH has the widest broadband operating velocity range and the greatest generated power followed by NL-PFEH and then L-PFEH. The maximum electric power output of NLA-PFEH was ~1000% greater than for NL-PFEH with a corresponding ~250% increase in the operating velocity range. Different cylinder configurations reveal the presence of hysteresis in the behavior of NLA-PFEH when the distance between the cylinders (so-called cylinder gap to diameter ratio), G/D < 0.5. At large distances from each other (G/D ≥ 4), the two cylinders behave like independent, isolated harvester units with rather weak mutual interaction.
- 12 August 2020
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 31 (20) , 2378-2389
Open AccessNon-linear piezoelectric fluidic energy harvesters: The mutual interaction of two oscillating cylinders
The presence of a bluff body upstream of a cantilever beam promotes persistent, aero-elastic vibrations of the beam. Vortex-induced vibration in an array of two mutually interacting bluff bodies in such configurations undergoing two-degrees of freedom transverse oscillation has not been investigated before. In the present work, we have studied experimentally, the unsteady response of an array of two similar rigid cylinders, positioned side-by-side in reference to the freestream velocity, each one mounted on the upstream end of an elastic cantilever beam. By fitting the beams with piezoelectric layers, these configurations are converted to piezoelectric fluid energy harvesters (PFEH) that can extract small amounts of energy from the flow. Comparing the performance of linear (L-PFEH), non-linear (NL-PFEH), and a non-linear array (NLA-PFEH) of harvesters show that NLA-PFEH has the widest broadband operating velocity range and the greatest generated power followed by NL-PFEH and then L-PFEH. The maximum electric power output of NLA-PFEH was ~1000% greater than for NL-PFEH with a corresponding ~250% increase in the operating velocity range. Different cylinder configurations reveal the presence of hysteresis in the behavior of NLA-PFEH when the distance between the cylinders (so-called cylinder gap to diameter ratio), G/D < 0.5. At large distances from each other (G/D ≥ 4), the two cylinders behave like independent, isolated harvester units with rather weak mutual interaction.
- 1 April 2020
- Published by Elsevier in International Journal of Adhesion and Adhesives
- Vol. 98, 102530
Characterization of adhesive joints under high-speed normal impact: Part II – Numerical studies
- 1 April 2020
- Published by Elsevier in International Journal of Adhesion and Adhesives
- Vol. 98, 102530
Characterization of adhesive joints under high-speed normal impact: Part II – Numerical studies
- 23 February 2018
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 29 (10) , 2267-2276
Freeze–thaw actuator and applications
Significant portions of the earth’s land mass undergo annual freeze–thaw cycles, and although water is abundant and practically a free resource, the possibility of using the water–ice phase transition for smart material applications and actuators for machines has not been studied. This article details some of the characteristics of a freeze–thaw actuator, compares it to other smart material actuators, and presents three experimental demonstrations of its potential for engineering applications. The first application is the conversion of the freeze cycle into electrical energy by coupling the freeze–thaw actuator with a bistable piezoelectric element. The second application demonstrates the ability to store energy mechanically and keep a count of multiple freeze–thaw cycles. This stored energy can then be released after a preset number of freeze–thaw cycles. The third application demonstrates a self-powered mechanism that is capable of moving itself one body length per freeze–thaw cycle.
- 23 February 2018
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 29 (10) , 2267-2276
Freeze–thaw actuator and applications
Significant portions of the earth’s land mass undergo annual freeze–thaw cycles, and although water is abundant and practically a free resource, the possibility of using the water–ice phase transition for smart material applications and actuators for machines has not been studied. This article details some of the characteristics of a freeze–thaw actuator, compares it to other smart material actuators, and presents three experimental demonstrations of its potential for engineering applications. The first application is the conversion of the freeze cycle into electrical energy by coupling the freeze–thaw actuator with a bistable piezoelectric element. The second application demonstrates the ability to store energy mechanically and keep a count of multiple freeze–thaw cycles. This stored energy can then be released after a preset number of freeze–thaw cycles. The third application demonstrates a self-powered mechanism that is capable of moving itself one body length per freeze–thaw cycle.
- 15 August 2017
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 29 (5) , 816-829
Capacitance changes in thin piezoelectric transducers embedded in isotropic host materials
Constraining a piezoelectric material from freely expanding causes a reduction in capacitance. This reduction in capacitance has implications in energy harvesting, active vibration control, and ultrasonic sensing. This article investigates the reduction in capacitance when a thin piezoelectric disk is embedded in an isotropic host material. Both simplified (one-dimensional axisymmetric) and computational (finite element two-dimensional axisymmetric) analyses have been presented and used to calculate the expected reduction in capacitance of thin embedded piezoelectric disks for a number of ceramic-based piezoelectric materials. The analyses show that there is a non-linear relationship between the capacitance and the elastic modulus of the host material; with increasing host modulus, capacitance initially decreases rapidly but then asymptotically converges to the fully constrained case. A simplified equivalent spring axisymmetric model has been derived from the finite element model and has been used to explain some of the key geometric and material properties that effect the reduction in embedded capacitance. Preliminary experimental results show some agreement with the derived models; however, substantial further validation is required. Some implications of the derived results for energy harvesting are also discussed.
- 15 August 2017
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 29 (5) , 816-829
Capacitance changes in thin piezoelectric transducers embedded in isotropic host materials
Constraining a piezoelectric material from freely expanding causes a reduction in capacitance. This reduction in capacitance has implications in energy harvesting, active vibration control, and ultrasonic sensing. This article investigates the reduction in capacitance when a thin piezoelectric disk is embedded in an isotropic host material. Both simplified (one-dimensional axisymmetric) and computational (finite element two-dimensional axisymmetric) analyses have been presented and used to calculate the expected reduction in capacitance of thin embedded piezoelectric disks for a number of ceramic-based piezoelectric materials. The analyses show that there is a non-linear relationship between the capacitance and the elastic modulus of the host material; with increasing host modulus, capacitance initially decreases rapidly but then asymptotically converges to the fully constrained case. A simplified equivalent spring axisymmetric model has been derived from the finite element model and has been used to explain some of the key geometric and material properties that effect the reduction in embedded capacitance. Preliminary experimental results show some agreement with the derived models; however, substantial further validation is required. Some implications of the derived results for energy harvesting are also discussed.
- 28 July 2016
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 27 (20) , 2837-2850
Comparison of the electromechanical properties of embedded and surface-mounted piezoelectric transducers
An experimental and computational comparison of the structural and electromechanical coupling properties of composite beams with surface-mounted and embedded piezoelectric disks shows that the embedded sensors tend to have higher electromechanical coupling coefficients due to improved strain transfer between the transducer and host material. Experimental testing of fiber–epoxy composite beams with embedded and surface-mounted piezoelectric disks under quasi-static three-point bending tests shows that bending elastic moduli and ultimate stresses are not significantly different between these two cases, while ultimate strains are greater for the surface-mounted cases compared with the embedded cases. Electromechanical coupling decreases with increasing bending loads due to either debonding between the sensor and host structure or due to damage of the brittle piezoelectric material. Embedded sensors tend to show a smaller degradation in electromechanical coupling with increasing stress level.
- 28 July 2016
- Published by SAGE Publications in Journal of Intelligent Material Systems and Structures
- Vol. 27 (20) , 2837-2850
Comparison of the electromechanical properties of embedded and surface-mounted piezoelectric transducers
An experimental and computational comparison of the structural and electromechanical coupling properties of composite beams with surface-mounted and embedded piezoelectric disks shows that the embedded sensors tend to have higher electromechanical coupling coefficients due to improved strain transfer between the transducer and host material. Experimental testing of fiber–epoxy composite beams with embedded and surface-mounted piezoelectric disks under quasi-static three-point bending tests shows that bending elastic moduli and ultimate stresses are not significantly different between these two cases, while ultimate strains are greater for the surface-mounted cases compared with the embedded cases. Electromechanical coupling decreases with increasing bending loads due to either debonding between the sensor and host structure or due to damage of the brittle piezoelectric material. Embedded sensors tend to show a smaller degradation in electromechanical coupling with increasing stress level.