49 results
Now showing1 - 10 of 49
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Bridging ocean wave physics and deep learning: physics-informed neural operators for nonlinear wavefield reconstruction in real-time(American Institute of Physics (AIP), 2025-10-07); ; Accurate real-time reconstruction of phase-resolved ocean wave fields remains a critical yet largely unsolved problem, primarily due to the absence of practical data assimilation methods for obtaining initial conditions from sparse or indirect wave measurements. While recent advances in supervised deep learning have shown potential for this purpose, they require large labeled datasets of ground truth wave data, which are infeasible to obtain in real-world scenarios. To overcome this limitation, we propose a physics-informed neural operator (PINO) framework for reconstructing spatially and temporally phase-resolved, nonlinear ocean wave fields from sparse measurements, without the need for ground truth data during training. This is achieved by embedding residuals of the free surface boundary conditions of ocean gravity waves into the loss function, constraining the solution space in a soft manner. In the current implementation, the framework is demonstrated for long-crested, unidirectional wave surfaces, where the wave propagation direction is aligned with the radar scanning direction. Within this setting, we validate our approach using highly realistic synthetic wave measurements by demonstrating the accurate reconstruction of nonlinear wave fields from both buoy time series and radar snapshots. Our results indicate that PINOs enable accurate, real-time reconstruction and generalize robustly across a wide range of wave conditions, thereby paving the way for future extensions of this framework toward multidirectional sea states and thus operational wave reconstruction in realistic marine environments.Publicationtype: Journal Article29 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Dynamics-informed reservoir computing with visibility graphsAccurate prediction of complex and nonlinear time series remains a challenging problem across engineering and scientific disciplines. Reservoir computing (RC) offers a computationally efficient alternative to traditional deep learning by training only the readout layer while employing a randomly structured and fixed reservoir network. Despite its advantages, the largely random reservoir graph architecture often results in suboptimal and oversized networks with poorly understood dynamics. Addressing this issue, we propose a novel Dynamics-Informed Reservoir Computing (DyRC) framework that systematically infers the reservoir network structure directly from the input training sequence. This work proposes to employ the visibility graph (VG) technique, which converts time series data into networks by representing measurement points as nodes linked by mutual visibility. The reservoir network is constructed by directly adopting the VG network from a training data sequence, leveraging the parameter-free visibility graph approach to avoid expensive hyperparameter tuning. This process results in a reservoir that is directly informed by the specific dynamics of the prediction task under study. We assess the DyRC-VG method through prediction tasks involving the canonical nonlinear Duffing oscillator, evaluating prediction accuracy and consistency. Compared to an Erdős-Rényi (ER) graph of the same size, spectral radius, and fixed density, we observe higher prediction quality and more consistent performance over repeated implementations in the DyRC-VG. An ER graph with density matched to the DyRC-VG can in some conditions outperform both approaches.Publicationtype: Journal Article18 14 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Physics-informed neural networks for phase-resolved data assimilation and prediction of nonlinear ocean waves(American Physical Society (APS), 2025-09-02); ; ; ; Cao, RuiThe assimilation and prediction of phase-resolved surface gravity waves are critical challenges in ocean science and engineering. Potential flow theory (PFT) has been widely employed to develop wave models and numerical techniques for wave prediction. However, traditional wave prediction methods are often limited. For example, most simplified wave models have a limited ability to capture strong wave nonlinearity, while fully nonlinear PFT solvers often fail to meet the speed requirements of engineering applications. This computational inefficiency also hinders the development of effective data assimilation techniques, which are required to reconstruct spatial wave information from sparse measurements to initialize the wave prediction. To address these challenges, we propose a solver method that leverages physics-informed neural networks (PINNs) that parametrize PFT solutions as neural networks. This provides a computationally inexpensive way to assimilate and predict wave data. The proposed PINN framework is validated through comparisons with analytical linear PFT solutions and experimental data collected in a laboratory wave flume. The results demonstrate that our approach accurately captures and predicts irregular, nonlinear, and dispersive wave surface dynamics. Moreover, the PINN can infer the fully nonlinear velocity potential throughout the entire fluid volume solely from surface elevation measurements, enabling the calculation of fluid velocities that are difficult to measure experimentally.Publicationtype: Journal Article16 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Pull-off force prediction in viscoelastic adhesive Hertzian contact by physics augmented machine learningPredicting the adhesive properties of viscoelastic Hertzian contacts is crucial for diverse engineering applications, including robotics, biomechanics, and advanced material design. This study introduces a novel physics-augmented machine learning (PA-ML) framework as a hybrid approach to study the maximum adherence force of a Hertzian indenter unloaded from a viscoelastic substrate, bridging the gap between analytical models and data-driven solutions. The PA-ML model is capable of rapidly predicting the pull-off force in an Hertzian profile unloaded from a broad band viscoelastic material, with varying Tabor parameter, preload and retraction rate. Compared to previous models, the PA-ML approach provides fast yet accurate predictions in a wide range of conditions, properly predicting the effective surface energy and the work-to-pull-off. The integration of the analytical model provides critical guidance to the PA-ML framework, supporting physically consistent predictions. We demonstrate that physics augmentation enhances predictive accuracy, reducing mean squared error (MSE) while increasing model interpretability. We provide data-driven and PA-ML models for real-time predictions of the adherence force in soft materials like silicons and elastomers opening to the possibility to integrate PA-ML into materials and interface design. The models are openly available on Zenodo and GitHub.Publicationtype: Journal Article13 18 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Lithium-ion battery degradation forecasting using data-driven time series models(American Society of Mechanical Engineers (ASME), 2025-06); ; The maritime industry faces significant challenges as it adapts from a major carbon emitter to a low-emission sector, intending to eventually achieve zero emissions. This transition requires innovative solutions for both new and old vessels, lithium-ion batteries show promise in achieving these goals. Battery management systems improve reliability and safety by monitoring voltage, current, and temperature through sensors. These parameters enable the prediction of remaining usable life, allowing for prompt maintenance and replacement before failure occurs. Publicly accessible lithium battery datasets provide a useful starting point for predictive degradation model development. This study investigates time series modeling methodologies for lithium-ion battery degradation, utilizing NASA’s battery degradation dataset. Three models viz. Autoregressive, Autoregressive Integrated Moving Average, and its extension using seasonality parameters were developed. They were tested with four train/test ratios to predict the remaining useful life values and assess the accuracy of the predicted degradation curve against experimental results. From the results, it was observed that the Autoregressive Integrated Moving Average model had the least combined average Root Mean Square Error values, resulting in a good overall degradation curve fitting, whereas the Seasonal Autoregressive Integrated Moving Average model was able to predict the End of Life values more accurately.Publicationtype: Conference Paper28 14 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Data-driven, non-linear ship response prediction based on time series of irregular, long-crested sea states amidships(Elsevier, 2025-02-01); ; ; ; The accurate prediction of vessel responses in waves is crucial for decision-making and contribute to the operational safety and risk minimization. Short-term predictions can be carried out by estimating the vessel's motions and loads based on incident waves. Existing model-based approaches either require computationally intensive simulations that compromise real-time capability or use simplified models affecting the accuracy of the prediction. Therefore, this study explores the feasibility of using neural networks for mapping time signals of surface elevation data and a set of corresponding ship responses, i.e. the heave and pitch motions as well as the vertical bending moment. The approach followed here is built on the assumption that the wave profile amidships is known. A synthetic dataset was generated using a time-domain strip theory solver with considerations of non-linear effects on motions and loads due to large amplitude waves in a variety of irregular, long-crested sea state conditions. We propose two different neural network models, a multi-layer perceptron (MLP) and a fully convolutional neural network (FCNN), and compare their performances on measurement data obtained from model tests in a seakeeping basin. The evaluations also include the freak wave reproduction of the ‘new year wave’. The proposed networks are able to estimate the motions and bending moment accurately for a wide range of sea state conditions, surpassing current state-of-the-art models on the given data sets.Publicationtype: Journal Article80 103 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , In Situ Operando Indicator of Dry Friction Squeal(Multidisciplinary Digital Publishing Institute (MDPI), 2024-12-08); ; ; In various applications, dry friction could induce vibrations. A well-known example is frictional braking systems in ground transportation vehicles involving a sliding contact between a rotating and a stationary part. In such scenarios, the emission of high-intensity noise, commonly known as squeal, can present human health risks based on the noise’s intensity, frequency, and occurrences. Despite the importance of squeal in the context of advancing urbanization, the parameters determining its occurrence remain uncertain due to the complexity of the involved phenomena. This study aims to identify a relevant operando indicator for predicting squeal occurrences. To this end, a pin-on-disc test rig was developed to replicate various contact conditions found in road profiles and investigate resulting squealing. Each test involves a multimodal instrumentation, complemented by surface observations. It is illustrated that the enhanced thermal indicator identified is relevant because it is sensitive to the thermomechanical and tribological phenomena involved in squealing.Publicationtype: Journal Article11 12 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Data assimilation and parameter identification for water waves using the nonlinear Schrödinger equation and physics-informed neural networks(Multidisciplinary Digital Publishing Institute (MDPI), 2024-10-01); ; ; The measurement of deep water gravity wave elevations using in situ devices, such as wave gauges, typically yields spatially sparse data due to the deployment of a limited number of costly devices. This sparsity complicates the reconstruction of the spatio-temporal extent of surface elevation and presents an ill-posed data assimilation problem, which is challenging to solve with conventional numerical techniques. To address this issue, we propose the application of a physics-informed neural network (PINN) to reconstruct physically consistent wave fields between two elevation time series measured at distinct locations within a numerical wave tank. Our method ensures this physical consistency by integrating residuals of the hydrodynamic nonlinear Schrödinger equation (NLSE) into the PINN’s loss function. We first showcase a data assimilation task by employing constant NLSE coefficients predetermined from spectral wave properties. However, due to the relatively short duration of these measurements and their possible deviation from the narrow-band assumptions inherent in the NLSE, using constant coefficients occasionally leads to poor reconstructions. To enhance this reconstruction quality, we introduce the base variables of frequency and wavenumber, from which the NLSE coefficients are determined, as additional neural network parameters that are fine tuned during PINN training. Overall, the results demonstrate the potential for real-world applications of the PINN method and represent a step toward improving the initialization of deterministic wave prediction methods.Publicationtype: Journal Article34 38 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Data-driven generation of tailored wave sequences(American Society of Mechanical Engineers (ASME), 2024-06); ; ; ; This paper explores the applicability of machine learning techniques for the generation of tailored wave sequences. For this purpose, a fully convolutional neural network was implemented for relating the target wave sequence at the target location in time domain to the respective control signal of the wave board. The database was generated by means of extensive wave tank tests. The experimental campaign focused on the generation of very steep wave groups including wave breaking which cannot be covered by the simplified wave generation methods. The extensive experimental campaign was performed in a small wave tank with an fully automated approach including determination and control of the wave maker motion as well as data measurement. The training data set features wave groups of short duration based on JONSWAP spectra, where the parameters wave steepness, peak wave period and enhancement factor were systematically varied. At the end of the training process, the trained models are able to predict the wave maker control signal based on time series of the target wave defined for a specific target location in the wave tank. The accuracy of the trained models are evaluated by means of unseen validation data. In addition, the predictive accuracy of the trained models is compared with the classical linear transformation approach.Publicationtype: Conference Paper57 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Building functional networks for complex response analysis in systems of coupled nonlinear oscillatorsSome aspects of engineering dynamics, such as nonlinearities and transient motion of many interconnected parts, remain difficult to handle today. To comply with increasing demands on resilience and safety, the dynamics of large machines need to be better understood. Complex network methods, already present in many scientific disciplines, provide a tool set complementary to conventional methods of system analysis. This work aims at providing a new, function-based view on mechanical systems by generating functional networks. To this end, a network algorithm is applied to sets of cyclically coupled Duffing oscillators as a common example of a complex nonlinear mechanical system. In the functional network, each node represents an oscillator while the direction of the network edges represents a functional coupling. Results show that the network method is capable of identifying dynamical transitions and synchronization between components, as well as determining the number of different states present within a system. Additionally, the time evolution of the component interactions, especially in response to a disturbance, is studied via a sliding-window approach. The results of this analysis might hopefully open new ways for a more efficient system analysis through optional sensor placement, and for effective countermeasures against unwanted dynamics through improved analysis of transient dynamics.Publicationtype: Journal Article20 61