Scholarly Works, Mechanical Engineering

Permanent URI for this collection

https://hdl.handle.net/10919/23262

Research articles, presentations, and other scholarship

Browse

Now showing 1 - 20 of 572

Electrostatic Defrosting
Lolla, Venkata Yashasvi; Zhang, Hongwei; Socha, Beckett Z.; Qiao, Rui; Boreyko, Jonathan B. (Wiley-VCH, 2025-11-11)
Electrification of ice has been studied for over half a century, mostly in the context of atmospheric science. Here, the polarizability and natural thermovoltage of a substrate-bound frost sheet are exploited for frost removal by placing an actively charged electrode overhead. This new technique, which we term electrostatic defrosting (EDF), can remove up to 75% of the frost’s mass from its substrate over a time scale of only minutes. A one-dimensional numerical model is developed to rationalize the effective electrostatic force exerted by the electrode on the warm end of the frost sheet. Experimentally, the effectiveness of EDF is shown to depend on the applied voltage, relative humidity of the ambient air, the gap height between the frosted substrate and the electrode plate, and the type of substrate. Although EDF primarily removes the dendritic frost structures rather than the underlying frozen condensate, this selective removal can still offer significant advantages for applications requiring improved visibility or reduced surface roughness. EDF can effectively remove frost without the application of heat, chemicals, or mechanical forces, rendering it a promising new construct for defrosting.
Airborne Acoustic Vortex End Effector-Based Contactless, Multi-Mode, Programmable Control of Object Surfing
Li, Teng; Li, Jiali; Bo, Luyu; Brooks, Michael R.; Du, Yingshan; Cai, Bowen; Pei, Zhe; Shen, Liang; Sun, Chuangchuang; Cheng, Jiangtao; Pan, Y. Albert; Tian, Zhenhua (Wiley, 2024-09-01)
Tweezers based on optical, electric, magnetic, and acoustic fields have shown great potential for contactless object manipulation. However, current tweezers designed for manipulating millimeter-sized objects such as droplets, particles, and small animals exhibit limitations in translation resolution, range, and path complexity. Here, a novel acoustic vortex tweezers system is introduced, which leverages a unique airborne acoustic vortex end effector integrated with a three-degree-of-freedom (DoF) linear motion stage, for enabling contactless, multi-mode, programmable manipulation of millimeter-sized objects. The acoustic vortex end effector utilizes a cascaded circular acoustic array, which is portable and battery-powered, to generate an acoustic vortex with a ring-shaped energy pattern. The vortex applies acoustic radiation forces to trap and spin an object at its center, simultaneously protecting this object by repelling other materials away with its high-energy ring. Moreover, The vortex tweezers system facilitates contactless, multi-mode, programmable object surfing, as demonstrated in experiments involving trapping, repelling, and spinning particles, translating particles along complex paths, guiding particles around barriers, translating and rotating droplets containing zebrafish larvae, and merging droplets. With these capabilities, It is anticipated that the tweezers system will become a valuable tool for the automated, contactless handling of droplets, particles, and bio-samples in biomedical and biochemical research. A novel acoustic vortex tweezers system is reported, which leverages a unique acoustic vortex end effector based on a portable, battery-powered, cascaded circular acoustic array. The system enables contactless, multi-mode, programmable object surfing, as demonstrated in experiments involving trapping, repelling, and spinning particles, translating particles along complex paths, guiding particles around barriers, and translating and rotating droplets containing zebrafish larvae. image
Scale modeling of thermo-structural fire tests of multi-orientation wood laminates
Gangi, Michael J.; Lattimer, Brian Y.; Case, Scott W. (Springer, 2024-07-01)
The stacking sequence of laminated wood significantly impacts the composite mechanical behavior of the material, especially when scaling down thermo-mechanical tests on plywood. In previous research, we developed a scaling methodology for thermo-structural tests on samples with similar cross sections, however this paper focused on testing plywood samples with different stacking sequences between the scales. Plywood samples at 1/2 -scale and 1/4 -scale were subjected to combined bending and thermal loading, with the loading scaled to have the same initial static bending stresses. While the 1/4 -scale 4-layer [0 degrees/90 degrees]s laminate and the 1/2 -scale 8-layer [0 degrees/90 degrees/90 degrees/0 degrees]s laminate had an equal number of 0 degrees and 90 degrees layers, as the char front progresses, the sections behave differently. Thus, modeling becomes essential to extrapolating the data from the smaller 1/4 -scale test to predict the behavior of the larger 1/2 -scale test. Reduced cross-sectional area models (RCAM) incorporating classical laminated plate theory were used to predict the mechanical response of the composite samples as the char front increased. Three methods were proposed for calibrating the RCAM models: Fourier number scaling, from detailed kinetics-based pyrolysis GPyro models, and fitting to data from fire exposure thermal response tests. The models calibrated with the experimental char measurements produced the most accurate predictions. The experimental char models validated to predict the behavior of the 1/4 -scale tests within 2.5%, were then able to predict the 1/2 -scale test behavior within 4.5%.
Influence of fuel inhomogeneity on detonation wave propagation in a rotating detonation combustor
Raj, P.; Meadows, Joseph (Springer, 2024-10-01)
Rotating detonation combustor (RDC) is a form of pressure gain combustion, which is thermodynamically more efficient than the traditional constant-pressure combustors. In most RDCs, the fuel-air mixture is not perfectly premixed and results in inhomogeneous mixing within the domain. Due to discrete fuel injection locations, local pockets of rich and lean mixtures are formed in the refill region. The objective of the present work is to gain an understanding of the effects of reactant mixture inhomogeneity on detonation wave structure, wave velocity, and pressure profile. To study the effect of mixture inhomogeneity, probability density functions of fuel mass fractions are generated with varying standard deviations. These distributions of fuel mass fractions are incorporated in 2D reacting simulations as a spatially/temporally varying inlet boundary condition. Using this methodology, the effect of mixture inhomogeneity is independently investigated to determine the effects on detonation wave propagation and RDC performance. As mixture inhomogeneity is increased, detonation wave speed, detonation efficiency, and potential for pressure gain all decrease, ultimately leading to the separation of the reaction zone from the shock wave.
Inhibiting Shuttle Effect and Dendrite Growth in Sodium-Sulfur Batteries Enabled by Applying External Acoustic Field
Zhang, Qipeng; Bo, Luyu; Li, Hao; Shen, Liang; Li, Jiali; Li, Teng; Xiao, Yunhao; Tian, Zhenhua; Li, Zheng (American Chemical Society, 2024-08-21)
The room-temperature sodium-sulfur (RT Na-S) battery is a promising alternative to traditional lithium-ion batteries owing to its abundant material availability and high specific energy density. However, the sodium polysulfide shuttle effect and dendritic growth pose significant challenges to their practical applications. In this study, we apply diverse disciplinary backgrounds to introduce a novel method to stimulate polarized BaTiO3 (BTO) nanoparticles on the separator. This approach generates more charges due to the piezoelectric effect under stronger driving forces produced by applying a controllable acoustic field at the outer edge of the cell. The acoustically stimulated BTO attracts more polysulfides, thus reducing the shuttling effect from the cathode to the anode and ultimately enhancing the battery performance. Meanwhile, the acoustic waves create additional streaming flows, improving the uniformity of the sodium ion dispersion, enhancing the sodium ion transport and reducing the possibility of sodium dendrite development. We believe that this work offers a new strategy for the development of high-performance Na-S batteries.
Integrated Fatigue Evaluation of As-Built WAAM Steel Through Experimental Testing and Finite Element Simulation
Gothivarekar, Sanjay; Brains, Steven; Raeymaekers, Bart; Talemi, Reza (MDPI, 2025-10-11)
Additive Manufacturing (AM) has attracted considerable interest over the past three decades, driven by growing industrial demand. Among metal AM techniques, Wire and Arc Additive Manufacturing (WAAM), a Directed Energy Deposition (DED) variant, has emerged as a prominent method for producing large-scale components with high deposition rates and cost efficiency. However, WAAM parts typically exhibit rough surface profiles, which can induce stress concentrations and promote fatigue crack initiation under cyclic loading. This study presents an integrated experimental and numerical investigation into the fatigue performance of as-built WAAM steel. Fatigue specimens extracted from a WAAM-fabricated wall were tested under cyclic loading, followed by fractography to assess the influence of surface irregularities and subsurface defects on fatigue behaviour. Surface topography analysis identified critical stress-concentration regions and key surface roughness parameters. Additionally, 3D scanning was used to reconstruct the specimen topography, enabling detailed 2D and 3D finite element (FE) modelling to analyze stress distribution along the as-built surface and predict fatigue life. A Smith-Watson-Topper (SWT) critical plane-based approach was applied for multiaxial fatigue life estimation. The results reveal a good correlation between experimental fatigue data and numerically predicted results, validating the proposed combined methodology for assessing durability of as-built WAAM components.
Parameter Identification of Soil Material Model for Soil Compaction Under Tire Loading: Laboratory vs. In-Situ Cone Penetrometer Test Data
Shokanbi, Akeem; Jasoliya, Dhruvin; Untaroiu, Costin D. (MDPI, 2025-10-15)
Accurate numerical simulations of soil-tire interactions are essential for optimizing agricultural machinery to minimize soil compaction and enhance crop yield. This study developed and compared two approaches for identifying and validating parameters of a LS-Dyna soil model. The laboratory-based approach derives parameters from triaxial, consolidation, and cone penetrometer tests (CPT), while the optimization-based method refines them using in-situ CPT data via LS-OPT to better capture field variability. Simulations employing Multi-Material Arbitrary Lagrangian–Eulerian (MM-ALE), Smoothed Particle Hydrodynamics (SPH), and Hybrid-SPH methods demonstrate that Hybrid-SPH achieves the optimal balance of accuracy (2% error post-optimization) and efficiency (14-h runtime vs. 22 h for SPH). Optimized parameters improve soil–tire interaction predictions, including net traction and tire sinkage across slip ratios from −10% to 30% (e.g., sinkage of 12.5 mm vs. 11.1 mm experimental at 30% slip, with overall mean-absolute percentage error (MAPE) reduced to 3.5% for sinkage and 4.2% for traction) and rut profiles, outperforming lab-derived values. This framework highlights the value of field-calibrated optimization for sustainable agriculture, offering a cost-effective alternative to field trials for designing low-compaction equipment and reducing yield losses from soil degradation. While sandy loam soil at 0.4% moisture content was used in this study, future extensions to different soil types with varied moisture are recommended.
Holographic thermal mapping in volumes using acoustic lenses
Cengiz, Ceren; Shahab, Shima (IOP Publishing, 2024-09-13)
Acoustic holographic lenses (AHLs) show great potential as a straightforward, inexpensive, and reliable method of sound manipulation. These lenses store the phase and amplitude profile of the desired wavefront when illuminated by a single acoustic source to reconstruct ultrasound pressure fields, induce localized heating, and achieve temporal and spatial thermal effects in acousto-thermal materials like polymers. The ultrasonic energy is transmitted and focused by AHL from a transducer into a particular focal volume. It is then converted to heat by internal friction in the polymer chains, causing the temperature of the polymer to rise at the focus locations while having little to no effect elsewhere. This one-of-a-kind capability is made possible by the development of AHLs to make use of the translation of attenuated pressure fields into programmable heat patterns. However, the impact of acousto-thermal dynamics on the generation of AHLs is largely unexplored. We use a machine learning-assisted single inverse problem approach for rapid and efficient AHLs' design to generate thermal patterns. The process involves the conversion of thermal information into a holographic representation through the utilization of two latent functions: pressure phase and amplitude. Experimental verification is performed for pressure and thermal measurements. The volumetric acousto-thermal analyses of experimental samples are performed to offer a knowledge of the obtained pattern dynamics, as well as the applicability of holographic thermal mapping for precise volumetric temperature control. Finally, the proposed framework aims to provide a solid foundation for volumetric analysis of acousto-thermal patterns within thick samples and for assessing thermal changes with outer surface measurements.
Simultaneous optimal system and controller design for multibody systems with joint friction using direct sensitivities
Verulkar, Adwait; Sandu, Corina; Sandu, Adrian; Dopico, Daniel (Springer, 2025-05-01)
Real-world multibody systems are often subject to phenomena like friction, joint clearances, and external events. These phenomena can significantly impact the optimal design of the system and its controller. This work addresses the gradient-based optimization methodology for multibody dynamic systems with joint friction using a direct sensitivity approach. The Brown-McPhee model has been used to characterize the joint friction in the system. This model is suitable for the study due to its accuracy for dynamic simulation and its compatibility with sensitivity analysis. This novel methodology supports codesign of the multibody system and its controller, which is especially relevant for applications like robotics and servo-mechanical systems, where the actuation and design are highly dependent on each other. Numerical results are obtained using a software package written in Julia with state-of-the-art libraries for automatic differentiation and differential equations. Three case studies are provided to demonstrate the attractive properties of simultaneous optimal design and control approach for certain applications.
Origami-/kirigami-inspired structures: from fundamentals to applications
Li, Suyi; Daqaq, Mohammed (Royal Society, 2024-10-07)
The Scutulum and the Pre-Auricular Aponeurosis in Bats
Pedersen, Scott C.; Snipes, Chelsie C. G.; Carter, Richard T.; Muller, Rolf (Wiley, 2024-11-01)
The external ear in eutherian mammals is composed of the annular, auricular (pinna), and scutellar cartilages. The latter extends between the pinnae, across the top of the head, and lies at the intersection of numerous auricular muscles and is thought to be a sesamoid element. In bats, this scutulum consists of two distinct regions, (1) a thin squama that is in contact with the underlying temporalis fascia and (2) a lateral bossed portion that is lightly tethered to the medial surface of the pinna. The planar size, shape, and proportions of the squama vary by taxa, as does the relative size and thickness of the boss. The origins, insertions, and relative functions of the auricular muscles are complicated. Here, 30 muscles were tallied as to their primary attachment to the pinnae, scutula, or a pre-auricular musculo-aponeurotic plate that is derived from the epicranius. In contrast to Yangochiroptera, the origins and insertions of many auricular muscles have shifted from the scutulum to this aponeurotic plate, in both the Rhinolophidae and Hipposideridae. We propose that this functional shift is a derived character related primarily to the rapid translations and rotations of the pinna in high-duty-cycle rhinolophid and hipposiderid bats.
Practical Application of Passive Air-Coupled Ultrasonic Acoustic Sensors for Wheel Crack Detection
Shaju, Aashish; Kumar, Nikhil; Mantovani, Giovanni; Southward, Steve; Ahmadian, Mehdi (MDPI, 2025-10-03)
Undetected cracks in railroad wheels pose significant safety and economic risks, while current inspection methods are limited by cost, coverage, or contact requirements. This study explores the use of passive, air-coupled ultrasonic acoustic (UA) sensors for detecting wheel damage on stationary or moving wheels. Two controlled datasets of wheelsets, one with clear damage and another with early, service-induced defects, were tested using hammer impacts. An automated system identified high-energy bursts and extracted features in both time and frequency domains, such as decay rate, spectral centroid, and entropy. The results demonstrate the effectiveness of UAE (ultrasonic acoustic emission) techniques through Kernel Density Estimation (KDE) visualization, hypothesis testing with effect sizes, and Receiver Operating Characteristic (ROC) analysis. The decay rate consistently proved to be the most effective discriminator, achieving near-perfect classification of severely damaged wheels and maintaining meaningful separation for early defects. Spectral features provided additional information but were less decisive. The frequency spectrum characteristics were effective across both axial and radial sensor orientations, with ultrasonic frequencies (20–80 kHz) offering higher spectral fidelity than sonic frequencies (1–20 kHz). This work establishes a validated “ground-truth” signature essential for developing a practical wayside detection system. The findings guide a targeted engineering approach to physically isolate this known signature from ambient noise and develop advanced models for reliable in-motion detection.
High cycle performance of twisted and coiled polymer actuators
Tsai, Samuel; Wang, Qiong; Hur, Ohnyoung; Bartlett, Michael D.; King, William P.; Tawfick, Sameh (Elsevier, 2025-01-01)
Twisted and coiled polymer actuators (TCPA), also known as coiled artificial muscles, are gaining popularity in soft robotics due to their large contractile actuation and work capacity. However, while it has been previously claimed that the stroke of TCPA remains stable after thousands of cycles, their absolute length change has not been rigorously studied. Here, we constructed an isobaric cycling setup that relies on fast heating and cooling by water immersion. This enables testing for 10k cycles in a duration of 56 hours, where the muscle temperature is varied between 15 degrees C and 75 degrees C at a rate of 20 seconds per cycle. Surprisingly, while the stroke usually remains unchanged for the entire 10k cycles as previously claimed, the final muscle loaded length exhibits all the geometrical possibilities of creep behavior as it can remain unchanged, elongate (creep), or contract (reverse creep) at the end of the test. Based on a wide range of experiments, we derived an empirical law which captures the observed relationship between the final muscle length change Delta L, the stroke alpha, and the passive strain 80: 80 + alpha = Delta L. Using this relation, the final length change of the muscle can be predicted from the first 100 cycles only. We show that polyvinylidene fluoride (PVDF), which does not swell in water, and nylon, which swells, follow this empirical law by testing in water with and without a protective coating, respectively. These results offer practical design guidelines for predictive actuation over thousands of cycles.
Enhancing performance of lithium metal batteries through acoustic field application
Zhang, Qipeng; Bo, Luyu; Li, Hao; Li, Jiali; Li, Teng; Tian, Zhenhua; Qiao, Rui (Royal Society Chemistry, 2025-01-14)
Cost-effective strategies for enhancing performance of lithium metal batteries (LMB) are in high demand. Herein, we propose and demonstrate that applying an external acoustic field can significantly enhance LMB performance, offering a novel approach to advancing battery technology. Long-term electrochemical stability tests, along with SEM and XPS characterization, reveal that this enhancement may result from the increased lithium-ion diffusion at slip lines and kinks, which can enable a more uniform solid electrolyte interphase (SEI) layer. Without the acoustic field, lithium ions exhibit slower conduction through thicker SEI regions, influenced by slip lines and kinks. In contrast, the application of an acoustic field facilitates more uniform ion diffusion, thereby enhancing overall performance. This approach provides a valuable pathway for advancing battery technology beyond the traditional focus on material innovation.
Anisotropic and Heterogeneous Thermal Conductivity in Programmed Liquid Metal Composites Through Direct Ink Writing
Hur, Ohnyoung; Markvicka, Eric J.; Bartlett, Michael D. (Wiley-V C H Verlag, 2025-03-01)
Thermal management in electric vehicles, electronics, and robotics requires the systematic ability to dissipate and direct the flow of heat. Thermally conductive soft composites are promising for thermal management due to their high thermal conductivity and mechanical flexibility. However, composites typically have the same microstructure throughout a film, which limits directional and spatial control of thermal management in emerging systems that have distributed heat loads. Herein, directional and spatially tunable thermal properties are programmed into liquid metal (LM) soft composites through a direct ink writing (DIW) process. Through the local control of LM droplet aspect ratio and orientation this programmable LM microstructure has a thermal conductivity in the direction of LM elongation of 9.9 W m-1K-1, which is similar to 40 times higher than the unfilled elastomer (0.24 W m-1K-1). The DIW process enables LM droplets to be oriented in specific directions with tunable aspect ratios at different locations throughout a continuous film. This introduces anisotropic and heterogeneous thermal conductivity in compliant films to control the direction and magnitude of heat transfer. This methodology and resulting materials can provide designed thermal management solutions for rigid and soft devices.
Process parameter optimization in polymer powder bed fusion of final part properties in polyphenylene sulfide through design of experiments
Ho, Ian; Bryant, Jackson; Chatham, Camden; Williams, Christopher (Springernature, 2024-12-17)
The Additive Manufacturing (AM) modality of Laser-Based Powder Bed Fusion of Polymers (PBF-LB/P) is an established method for manufacturing semi-crystalline polymers. Like other AM processes, the selection of PBF-LB/P process parameters is critical as it has direct effect on final part properties. While prior research has been predominantly focused on polyamides (e.g., nylon 12), there exists a gap in exploring how process parameters affect higher performance polymers, such as polyphenylene sulfide (PPS). This work aims to explore the effects of PBF-LB/P process parameters on PPS parts printed via PBF-LB/P. While prior PBF-LB/P parameter research primarily relies on evaluating energy input to the system through a single numerical value of energy density, this study investigates the interplay of the print parameters within the energy density equation. To achieve these goals, an analysis was performed on the influence of the laser power, hatch spacing, and beam velocity on ultimate tensile strength (UTS), modulus, and crystallinity of printed parts. A Taguchi L8 array was used in balancing the print parameter combinations allowing for isolation of variance to the specific factors and interactions. Through this approach, print parameter combinations that improved UTS and modulus were identified. Additionally, the study revealed that numerically equivalent energy densities did not lead to equivalent performance, underscoring the significance for including the constitutive process parameters within the energy density equation when establishing process property relationships in printing with PBF-LB/P.
Electromigration of Aquaporins Controls Water-Driven Electrotaxis
Sáez, Pablo; Kale, Sohan (MDPI, 2025-09-10)
Cell motility is a process central to life and is undoubtedly influenced by mechanical and chemical signals. Even so, other stimuli are also involved in controlling cell migration in vivo and in vitro. Among these, electric fields have been shown to provide a powerful and programmable cue to manipulate cell migration. There is now a clear consensus that the electromigration of membrane components represents the first response to an external electric field, which subsequently activates downstream signals responsible for controlling cell migration. Here, we focus on a specific mode of electrotaxis: frictionless, amoeboid-like migration. We used the Finite Element Method to solve an active gel model coupled with a mathematical model of the electromigration of aquaporins and investigate the effect of electric fields on ameboid migration. We demonstrate that an electric field can polarize aquaporins in a cell and, consequently, that the electromigration of aquaporins can be exploited to regulate water flux across the cell membrane. Our findings indicate that controlling these fluxes allows modulation of cell migration velocity, thereby reducing the cell’s migratory capacity. Our work provides a mechanistic framework to further study the impact of electrotaxis and to add new insights into specific modes by which electric fields modify cell motility.
Modeling Hysteretically Nonlinear Piezoelectric Composite Beams
Alazemi, Abdulaziz H.; Kurdila, Andrew J. (MDPI, 2025-07-06)
This paper presents a modeling framework for hysteretically nonlinear piezoelectric composite beams using functional differential equations (FDEs). While linear piezoelectric models are well established, they fail to capture the complex nonlinear behaviors that emerge at higher electric field strengths, particularly history-dependent hysteresis effects. This paper develops a cascade model that integrates a high-dimensional linear piezoelectric composite beam representation with a nonlinear Krasnosel’skii–Pokrovskii (KP) hysteresis operator. The resulting system is formulated using a state-space model where the input voltage undergoes a history-dependent transformation. Through modal expansion and discretization of the Preisach plane, we derive a tractable numerical implementation that preserves essential nonlinear phenomena. Numerical investigations demonstrate how system parameters, including the input voltage amplitude, and hysteresis parameters significantly influence the dynamic response, particularly the shape and amplitude of limit cycles. The results reveal that while the model accurately captures memory-dependent nonlinearities, it depends on numerous real and distributed parameters, highlighting the need for efficient reduced-order modeling approaches. This work provides a foundation for understanding and predicting the complex behavior of piezoelectric systems with hysteresis, with potential applications in vibration control, energy harvesting, and precision actuation.
Post-Hurricane Debris and Community Flood Damage Assessment Using Aerial Imagery
Aggarwal, Diksha; Gautam, Suyog; Whitehurst, Daniel; Kochersberger, Kevin (MDPI, 2025-09-12)
Natural disasters often result in significant damage to infrastructure, generating vast amounts of debris in towns and water bodies. Timely post-disaster damage assessment is critical for enabling swift cleanup and recovery efforts. This study presents a combination of methods to efficiently estimate and analyze debris on land and on water. Specifically, analyses were conducted at Claytor Lake and Damascus, Virginia where flooding occurred as a result of Hurricane Helene on 27 September 2024. We use the Phoenix U15 motor glider equipped with the GoPro Hero 9 camera to collect aerial imagery. Orthomosaic images and 3D maps are generated using OpenDroneMap (ODM) software, version 3.5.6, providing a detailed view of the affected areas. For lake debris estimation, we employ a hybrid approach integrating machine learning-based tools and traditional techniques. Lake regions are isolated using segmentation methods, and the debris area is estimated through a combination of color thresholding and edge detection. The debris is classified based on the thickness and a volume range of debris is presented based on the data provided by the Virginia Department of Environmental Quality (VDEQ). In Damascus, debris estimation is achieved by comparing pre-disaster LiDAR data (2016) with post-disaster 3D ODM data. Furthermore, we conduct flood modeling using the Hydrologic Engineering Center’s River Analysis System (HEC-RAS) to simulate disaster impacts, estimate the flood water depth, and support urban planning efforts. The proposed methodology demonstrates the ability to deliver accurate debris estimates in a time-sensitive manner, providing valuable insights for disaster management and environmental recovery initiatives.
A Framework for an ML-Based Predictive Turbofan Engine Health Model
Jung, Jin-Sol; Son, Changmin; Rimell, Andrew; Clarkson, Rory J. (MDPI, 2025-08-14)
A predictive health modeling framework was developed for a family of turbofan engines, focusing on early detection of performance degradation. Turbine Gas Temperature (TGT) was employed as the primary indicator of engine health within the model, due to its strong correlation with core engine performance and thermal stress. The present research uses engine health monitoring (EHM) data acquired from in-service turbofan family engines. TGT is typically measured downstream of the high-pressure turbine stage and is regarded as the key thermodynamic variable of the gas turbine cycle. Three new training approaches were proposed using data segmentation based on time between major overhauls and compared with the conventional train–test split method. Detrending was employed to effectively remove trends and seasonality, enabling the ML-based model to learn more intrinsic relationships. Large generalized models based on the entire engine family were also investigated. Prediction performance was evaluated using selected machine learning (ML) models, including both linear and nonlinear algorithms, as well as a long short-term memory (LSTM) approach. The models were compared based on accuracy and other relevant performance metrics. The prediction accuracies of ML models depend on the selection of data size and segmentation for training and testing. For individual engines, the proposed training approaches predicted TGT with the accuracy of 4 ^∘C to 6 ^∘C in root mean square error (RMSE) by utilizing 65% less data than the train (80%)–test (20%) split method. Utilizing the data of each family engine, the large generalized model achieved similar prediction accuracy in RMSE with a smaller interquartile range. However, the amount of data required was 45–300 times larger than the proposed approaches. The sensitivity of prediction accuracy to the size of the training dataset offers valuable insights into the framework’s applicability, even for engines with limited data availability. Uncertainty quantification showed a coverage width criterion (CWC) between 29 ^∘C and 40 ^∘C, varying with different family engines.

Browse

Recent Submissions