Online reconstruction of 3D temperature field fused with POD-based reduced order approach and sparse sensor data

Jiang, Geng-Hui; Kang, Ming; Cai, Zhenwei; Wang, Han; Liu, Yingzheng; Wang, Weizhe

doi:10.1016/j.ijthermalsci.2022.107489

Cited by 28 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The goal of sparse data reconstruction is to take a few sensor values from a space that cannot be fully observed, and use them to reconstruct the global field [1]. Reconstructing a field from a few observations is considered a 'grand challenge' [2][3][4] in many fields, including: industry [5], medicine [6], and environmental sciences [7][8][9][10]. Sparse data reconstruction is especially challenging in applications involving fluid dynamics, which tend to present highly non-linear and chaotic phenomena [11].…”

Section: Introductionmentioning

confidence: 99%

Journey over destination: dynamic sensor placement enhances generalization

Marcato,

Guiltinan,

Viswanathan

et al. 2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

Reconstructing complex, high-dimensional global fields from limited data points is a challenge across various scientific and industrial domains. This is particularly important for recovering spatio-temporal fields using sensor data from, for example, laboratory-based scientific experiments, weather forecasting, or drone surveys. Given the prohibitive costs of specialized sensors and the inaccessibility ofcertain regions of the domain, achieving full field coverage is typically not feasible. Therefore, the development of machine learning algorithms trained to reconstruct fields given a limited dataset is of critical importance. In this study, we introduce a generalapproach that employs moving sensors to enhance data exploitation during the training of an attention based neural network, thereby improving field reconstruction. The training of sensor locations is accomplished using an end-to-end workflow, ensuringdifferentiability in the interpolation of field values associated to the sensors, and is simple to implement using differentiable programming. Additionally, we have incorporated a correction mechanism to prevent sensors from entering invalid regions within the domain. We evaluated our method using two distinct datasets; the results show that our approach enhances learning, as evidenced by improved test scores.

show abstract

Section: Introductionmentioning

confidence: 99%

Journey over destination: dynamic sensor placement enhances generalization

Marcato,

Guiltinan,

Viswanathan

et al. 2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…In addition, Kaneko et al (2021) attempted to reconstruct the propagation time and attenuation rate distributions of each sound source grid point to a microphone, which are required for beamforming, using the reduced-order model with sparse sampling for the acceleration of the computation. Furthermore, it should be noted that the data-driven sparse sampling has recently been applied to the online temperature fields reconstruction of steam turbine casing (Jiang et al, 2022), to the fast computation of inverse transient analysis for pipeline condition assessment (Wang, 2022), and to the improvement of signal-to-noise ratio in noisy measurement (Inoue et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Optimization of Sparse Sensor Placement for Estimation of Wind Direction and Surface Pressure Distribution Using Time-Averaged Pressure-Sensitive Paint Data on Automobile Model

Inoba,

Uchida,

Iwasaki

et al. 2022

Preprint

View full text Add to dashboard Cite

This study proposes a method for predicting the wind direction against the simple automobile model (Ahmed model) and the surface pressure distributions on it by using data-driven optimized sparse pressure sensors. Positions of sparse pressure sensor pairs on the Ahmed model were selected for estimation of the yaw angle and reconstruction of pressure distributions based on the time-averaged surface pressure distributions database of various yaw angles, whereas the symmetric sensors in the left and right sides of the model were assumed. The surface pressure distributions were obtained by pressure-sensitive paint measurements. Three algorithms for sparse sensor selection based on the greedy algorithm were applied, and the sensor positions were optimized. The sensor positions and estimation accuracy of yaw angle and pressure distributions of three algorithms were compared and evaluated. The results show that a few optimized sensors can accurately predict the yaw angle and the pressure distributions.

show abstract

“…This is so-called "full state recovery." This framework has already been used in the real (experimental) data in the fluid dynamics (Kanda et al, 2021;Zhou et al, 2021;Loiseau et al, 2018;Inoue et al, 2021;Jiang et al, 2022;Inoba et al, accepted). Usage of this framework is not limited to the time-series data.…”

Section: Introductionmentioning

confidence: 99%

Seismic Wavefield Reconstruction based on Compressed Sensing using Data-Driven Reduced-Order Model

Nagata,

Nakai,

Yamada

et al. 2022

Preprint

View full text Add to dashboard Cite

Reconstruction of the distributions of ground motion due to an earthquake is one of the key technologies for the prediction of seismic damage to infrastructure. Particularly, immediate reconstruction of spatially continuous wavefield is valuable for decision-making of disaster response decisions in the initial phase. For a fast and accurate reconstruction, utilization of prior information and effective reduction of observation is essential. In fluid mechanics, full-state recovery, which recovers the full state from sparse observation using a data-driven model reduced-order model, is actively used. In the present study, the framework developed in the field of fluid mechanics is applied to seismic wavefield reconstruction. A seismic wavefield reconstruction framework based on compressed sensing using the data-driven reduced-order model (ROM) is proposed and its characteristics are investigated through numerical experiments. The data-driven ROM is generated from the dataset of the wavefield using the singular value decomposition. The spatially continuous seismic wavefield is reconstructed from the sparse and discrete observation and the datadriven ROM. The observation sites used for reconstruction are effectively selected by the sensor optimization method for linear inverse problems based on a greedy algorithm. The proposed framework was applied to synthetic waveforms without and with observation noise. The validity of the proposed method was confirmed by the reconstruction based on the noise-free observation. Since the ROM of the wavefield is used as prior information, the reconstruction error is reduced to an approximately lower error bound of the present framework, even though the number of the sensors used for reconstruction is limited and randomly selected. In addition, the reconstruction error obtained by the proposed framework is much smaller than that obtained by the Gaussian process regression. For the numerical experiment with noise-contaminated observation, the reconstructed wavefield is degraded due to the observation noise, but the reconstruction error obtained by the present framework with all available observation sites is close to a lower error bound, even though the reconstructed wavefield using the Gaussian process regression is fully collapsed. Although the reconstruction error is larger than that obtained using all observation sites, the number of observation sites used for reconstruction can be reduced while minimizing the deterioration and scatter of the reconstructed data by combining it with the sensor optimization method. Hence, the subset of the optimized observation sites selected by the greedy method provides a better and more stable reconstruction of the wavefield than randomly selected observation sites, even if the reconstruction is carried out with a smaller number of observations with observation noise.

show abstract

Online reconstruction of 3D temperature field fused with POD-based reduced order approach and sparse sensor data

Cited by 28 publications

References 24 publications

Journey over destination: dynamic sensor placement enhances generalization

Journey over destination: dynamic sensor placement enhances generalization

Optimization of Sparse Sensor Placement for Estimation of Wind Direction and Surface Pressure Distribution Using Time-Averaged Pressure-Sensitive Paint Data on Automobile Model

Seismic Wavefield Reconstruction based on Compressed Sensing using Data-Driven Reduced-Order Model

Contact Info

Product

Resources

About