Abstract: Overview This dataset supports the manuscript *BubbleID-Flow: Machine-Vision Quantification of Vapor Area Fraction in Subcooled Flow Boiling*. It contains high-speed optical images, manually annotated images, thermal workbooks, acoustic-emission (AE) records, and the fine-tuned Mask R-CNN checkpoint used in the reported analysis. The experiments used FC-72 in a rectangular flow passage that was 2.5 mm wide and 5.0 mm high. The bottom wall was electrically heated over 114.6 mm. High-speed images were acquired with a FASTCAM SA-Z at 1024 x 1024 pixels and 3000 frames/s. The manuscript reports a two-dimensional, near-wall vapor area fraction measured from the image-plane union of predicted vapor masks. It is not a volumetric vapor fraction or void fraction. Contents ```text AnnotatedImages/ 130 paired BMP images and LabelMe JSON annotation files. ModelWeights/ model_final.pth Fine-tuned one-class Mask R-CNN checkpoint. MultimodalData/ Images/ Raw high-speed images and camera metadata. FlowLoopThermalData/ Thermal workbooks and derived thermal plots. AcousticDatta/ AE event, continuous, waveform, and plot files. Videos/ BubbleID-Flow_45V_raw_vs_segmented.mp4 Derived 16-second presentation video. BubbleID-Flow_45V_raw_vs_segmented.json Video frame-selection and inference manifest. BubbleID-Flow_45V_raw_top_segmented_bottom.mp4 Derived 16-second presentation video. BubbleID-Flow_45V_raw_top_segmented_bottom.json Video frame-selection and inference manifest. ``` `AnnotatedImages` contains 130 valid LabelMe JSON files paired with 130 BMP images. The annotation labels are `bubble` and `bubble_cluster`; the BubbleID-Flow model merges both labels into one vapor-region class for union-mask vapor-area-fraction analysis. `MultimodalData/Images` contains 3,977 BMP images and 37 `.cihx` camera metadata files. `MultimodalData/FlowLoopThermalData` contains three Excel workbooks per case: `.xlsx`, `_IN_with_units.xlsx`, and `_u_with_units.xlsx`. The PNG files under `Plot Images` and the AE `Plot_Hit` and `Plot_Time` folders are derived visualizations, not primary measurement inputs. `Videos/BubbleID-Flow_45V_raw_vs_segmented.mp4` is a derived presentation output. It shows 20 uniformly sampled frames from the 45 V state of each case at 5 frames/s (16 seconds total), with a raw near-wall view and the corresponding Mask R-CNN union-mask overlay. Its JSON sidecar records the selected filenames, checkpoint checksum, ROI, score threshold, and detection cap. Case Map | Canonical case ID | Nominal mass flow | Nominal inlet subcooling | Optical image folder | Thermal folder | AE folder | | --- | ---: | ---: | --- | --- | --- | | `5gs_22C` | 5 g/s | 22 degC | `MultimodalData/Images/5gs_22C` | `MultimodalData/FlowLoopThermalData/5gs_22C` | `MultimodalData/AcousticDatta/5gs_22C` | | `10gs_22C` | 10 g/s | 22 degC | `MultimodalData/Images/10gs_22C` | `MultimodalData/FlowLoopThermalData/10gs_22C` | `MultimodalData/AcousticDatta/10gs_22C` | | `15gs_20C` | 15 g/s | 20 degC | `MultimodalData/Images/15gs_20C` | `MultimodalData/FlowLoopThermalData/15gs_20C` | `MultimodalData/AcousticDatta/15gs_22C` (archive alias) | | `25gs_20C` | 25 g/s | 20 degC | `MultimodalData/Images/25gs_20C` | `MultimodalData/FlowLoopThermalData/25gs_20C` | `MultimodalData/AcousticDatta/25gs_22C` (archive alias) | Archive naming notes - `AcousticDatta` is the current directory name in this archive. - The AE directories `15gs_22C` and `25gs_22C` are aliases only. Their contained HIT, TIME, DTA, and WFS filenames identify the canonical `15gs_20C` and `25gs_20C` cases. Use the canonical IDs in analyses and citations. - Optical folder labels such as `25_ONSET`, `55 CHF`, and `59CHF` are source acquisition labels. They are not independently validated onset or critical-heat-flux labels. Modality Details Optical images The image folders are organized as `/`. The study uses a fixed near-wall region of interest (ROI) beginning at pixel coordinate `(0, 485)` with size `1024 x 70` pixels. Spatial calibration and registration between this ROI, the heated surface, and the seven temperature locations were not available. Therefore, this archive does not support calibrated bubble-size measurements, local optical-thermal registration, or volumetric void-fraction calculation. Thermal data The thermal workbooks contain time, mass flow, voltage, electrical power, inlet subcooling, inlet/outlet temperature, pressure, and seven test-section surface-temperature streams. The manuscript derives heat flux from electrical power and the documented heated area and derives mass flux from the measured flow area. Reported standard deviations characterize temporal variability within selected windows; they are not instrument uncertainty estimates. For each canonical case, the BubbleID-Flow manuscript pipeline uses both: - `.xlsx` for thermal variables; and - `_IN_with_units.xlsx` to obtain voltage and define/match state windows. Acoustic-emission data Each AE case directory contains: - `HIT_.TXT`: event-level hit data; - `TIME_.TXT`: continuous time-domain summary data; - `Raw Data/.wfs`: raw waveform file; - `Raw Data/.DTA`: acquisition/export file; - `notes.txt`: available operator notes; and - `Plot_Hit` and `Plot_Time`: derived diagnostic figures. The optical, thermal, and AE systems do **not** have a verified common hardware trigger. The manuscript's supplementary 10 g/s AE screen uses a 25.395 s archive-start timestamp offset for descriptive state comparison only. These data do not establish optical-acoustic synchronization, causal coupling, source localization, flow-regime classification, or calibrated AE energy. Annotations and model checkpoint The model was fine-tuned on 130 manually annotated images. The published reconstruction evaluation uses a deterministic same-sequence split: every fifth sorted image is assigned to holdout, yielding 104 training images and 26 holdout images. This is not an experiment-held-out generalization test; the source-sequence proximity between training and holdout images must be considered when interpreting performance. The included checkpoint is: ```text ModelWeights/model_final.pth SHA-256: 6f60969ce876f57a78b53fc61895c99b92cb2e1b6f0e0b30f01fa301236042cd ```
more »
« less
BoilingBench-Multimodal Lite: Multimodal Boiling and Thermal-Management Benchmark Dataset
Abstract: BoilingBench-Multimodal Lite is a versioned, machine-learning-oriented release from the NED³ Laboratory at the University of Arkansas. It contains seven related benchmark tracks covering pool boiling, bubble-image annotation, acoustic sensing, and infrared immersion cooling. BoilingBench-1 through BoilingBench-4 contain multimodal boiling experiments; BoilingBench-5 contains human-annotated boiling images; BoilingBench-6 is a hydrophone-focused pool-boiling dataset; and BoilingBench-7 is an infrared immersion-cooling dataset. This Lite profile omits original Phantom `.cine` recordings to reduce distribution size. It retains the released raw files, processed analysis products, annotations, manifests, plots, and machine-readable exports included in the directory. The complete archive and any omission details are documented in the accompanying README and release manifest. Raw acquisition files are measurements or source observations. Heat flux, reconstructed surface temperature, synchronization records, spectra, spectrograms, event times, hysteresis metrics, and other analysis products are derived data unless explicitly identified otherwise. BoilingLab-generated heat-flux products require users to inspect thermocouple locations, geometry, calibration, material assumptions, time alignment, sign convention, and quality fields before treating them as labels. BoilingBench-3 and BoilingBench-4 use legacy MISTRAS USB AE acquisition and do not contain continuous AE waveform exports. BoilingBench-3 contains microphone data. BoilingBench-1 and BoilingBench-2 contain continuous AE waveform products where documented. Users must split examples by independent experiment/run or physical condition rather than randomly splitting adjacent temporal windows. The dataset is intended for reproducible research in heat-transfer prediction, multimodal sensing, boiling-regime and transition analysis, bubble-image analysis, acoustic heat-flux estimation, sequence regression, and cross-domain generalization. It is not a clinical, personal, or safety-certified dataset.
more »
« less
- PAR ID:
- 10712570
- Publisher / Repository:
- Zenodo
- Date Published:
- Edition / Version:
- 0.1.0
- Subject(s) / Keyword(s):
- Right(s):
- Creative Commons Attribution 4.0 International
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Boiling is a high-performance heat dissipation process that is central to electronics cooling and power generation. The past decades have witnessed significantly improved and better-controlled boiling heat transfer using structured surfaces, whereas the physical mechanisms that dominate structure-enhanced boiling remain contested. Experimental characterization of boiling has been challenging due to the high dimensionality, stochasticity, and dynamicity of the boiling process. To tackle these issues, this paper presents a coupled multimodal sensing and data fusion platform to characterize boiling states and heat fluxes and identify the key transport parameters in different boiling stages. Pool boiling tests of water on multi-tier copper structures are performed under both steady-state and transient heat loads, during which multimodal, multidimensional signals are recorded, including temperature profiles, optical imaging, and acoustic signals via contact acoustic emission (AE) sensors, hydrophones immersed in the liquid pool, and condenser microphones outside the boiling chamber. The physics-based analysis is focused on i) extracting dynamic characteristics of boiling from time lags between acoustic-optical-thermal signals, ii) analyzing energy balance between thermal diffusion, bubble growth, and acoustic dissipation, and iii) decoupling the response signals for different physical processes, e.g., low-to-midfrequency range AE induced by thermal expansion of liquids and bubble ebullition. Separate multimodal sensing tests, namely a single-phase liquid test and a single-bubble-dynamics test, are performed to reinforce the analysis, which confirms an AE peak of 1.5 kHz corresponding to bubble ebullition. The data-driven analysis is focused on enabling the early fusion of acoustic and optical signals for improved boiling state and flux predictions. Unlike single-modality analysis or commonly-used late fusion algorithms that concatenate processed signals in dense layers, the current work performs the fusion process in the deep feature domain using a multi-layer perceptron regression model. This early fusion algorithm is shown to lead to more accurate and robust predictions. The coupled multimodal sensing and data fusion platform is promising to enable reliable thermal monitoring and advance the understanding of dominant transport mechanisms during boiling.more » « less
-
Abstract: This paper presents Hit2Flux, a machine learning framework for boiling heat flux prediction using acoustic emission (AE) hits generated through threshold-based transient sampling. Unlike continuously sampled data, AE hits are recorded when the signal exceeds a predefined threshold and are thus discontinuous in nature. Meanwhile, each hit represents a waveform at a high sampling frequency (∼1 MHz). In order to capture the features of both the high-frequency waveforms and the temporal distribution of hits, Hit2Flux involves i) feature extraction by transforming AE hits into the frequency domain and organizing these spectra into sequences using a rolling window to form “sequences-of-sequences,” and ii) heat flux prediction using a long short-term memory (LSTM) network with sequences of sequences. The model is trained on AE hits recorded during pool boiling experiments using an AE sensor attached to the boiling chamber. Continuously sampled acoustic data using a hydrophone were also collected as a reference data set for this study. Results demonstrate that the proposed AE-based method achieves performance comparable to hydrophones, validating its potential for heat flux monitoring. Additionally, it is shown that the inclusion of multiple acoustic emission hits as model inputs leads to higher performance. The Hit2Flux model is also compared to methods pairing various signal preparation techniques with state-of-the-art models. This comparison further highlighted the superior accuracy of the proposed approach. The developed Hi2Flux algorithm can be applied to other transient sampling events, such as structural health monitoring, detection of electromagnetic pulses, among others. TechnicalInfo: # Data for: Hit2flux: A machine learning framework for boiling heat flux prediction using hit-based acoustic emission sensing Dataset DOI: [10.5061/dryad.g79cnp628](10.5061/dryad.g79cnp628) ## Description of the data and file structure This dataset includes acoustic emission hit data and waveforms, hydrophone data, pressure, and temperature data from transient pool boiling tests on copper microchannels. The pool boiling test facility includes (a) a heating element that consists of a copper block and cartridge heaters; (b) a closed chamber with flow loops for a chiller (Thermo Scientific Polar ACCEL 500 Low/EA) connecting Graham condenser (Ace glass 5953-106) with an adapter (Ace glass 5838-76) and an in-house built coiled copper condenser; and (c) a synchronized multimodal sensing system. The copper block is submerged in deionized water and heated by nine cartridge heaters (Omega Engineering HDC19102), each with a power rating of 50 W, inserted from the bottom. The cartridge heaters are connected to the DC power supply (MagnaPower SL200-7.5) for controlling the heating power. The copper tube coiled condenser is used to keep the vapor pressure in check during the boiling test. This is done by controlling the flow rate passing through it manually using the flow metering valve with a Vernier handle (Swagelok SS-SS4-VH). Two screw-plug immersion heaters (McMaster-Carr 4668T54) are immersed in the liquid pool for degassing and auxiliary heating. A variable-voltage transformer (McMaster-Carr 6994K17) is used for adjusting the voltage required for the immersion heaters within the given range. The test facility is installed on a Thorlab Nexus optical table (T46V) with passive legs (PTP703) for vibration isolation. The following sensors are installed in the pool boiling facility. * A high-accuracy pressure transducer (Omega Engineering PX409030A5V) with an NI-9239 module is used to measure the vapor pressure in the boiling chamber. * Four T-type thermocouples (Omega Engineering Tj36-CPSS-032U-6) are inserted near the boiling surface of the copper block to measure the temperature gradient and calculate the heat flux and boiling surface temperature. Two T-type threaded thermocouples (McMaster-Carr 1245N16) are placed inside the boiling chamber to measure liquid and vapor temperatures. The thermocouples are connected to an NI-9210 module for data acquisition. * Two hydrophones (High Tech HTI-96-Min) are immersed in the liquid pool for acoustic sensing. They are connected to an NI-9230 module. * An acoustic emission sensor (MISTRAS R3a) is mounted to the outer surface of the bottom plate of the boiling chamber. It is connected to a MISTRAS DAQ (1283 USB AE Node) for data collection. The detailed experimental procedures can be found in [Pandey et al. 2024](https://www.sciencedirect.com/science/article/pii/S1359431123018367), [Pandey et al. 2024](https://www.sciencedirect.com/science/article/pii/S2352340924005493), and [Pandey et al. 2025](https://www.sciencedirect.com/science/article/pii/S1359431124029272). These data are used for training [SeqReg](https://github.com/cldunlap73/SeqReg), a machine learning framework for sequence regression. This dataset is used for acoustic-based heat flux prediction, detailed in this [tutorial](https://colab.research.google.com/drive/13TsHtLGvv9iiV82KZQYsbkmeTUubQO8c?usp=sharing). ### Files and variables #### File: ned3-005\_Hit2Flux.zip **Description:** This zip includes four folders, i.e., Boiling-124a, Boiling-124b, Boiling-125, and Boiling-126. Each folder includes five files and a subfolder. In each folder, labeled as a generic test ID, Boiling-xxx, * The Boiling-xxx_AE.DTA file is the source file of acoustic emission signals captured using a MISTRAS R3a sensor and a MISTRAS USB AE Node. This file can be opened using the MISTRAS AEWin software. * The Boiling-xxx_AEHit.txt file is the hit data exported from the DTA file. * The Boiling-xxx_AE_waveform folder contains the waveforms for each hit in csv format, also exported from the DTA file. * The Boiling-xxx_Hydrophone.lvm file includes the continuously sampled acoustic data using two HTI-96-MIN hydrophones with an NI-9230 module. * The Boiling-xxx_Temperature.lvm includes the temperature measurements using probe thermocouples (Omega Engineering Tj36-CPSS-032U-6) in the copper block and pool thermocouples in the water and vapor phases (McMaster-Carr 1245N16). These data are collected using an NI-9210 module. * The Boiling-xxx_Pressure.lvm file includes pressure measurements in the vapor phase using a pressure transducer (Omega Engineering PX409030A5V) and an NI-9239 module. Boiling-124a and Boiling-124b are closed chamber tests, where the Graham condenser is closed after degassing, and the pressure is regulated through the balance between boiling and condensation by tuning the flow rate of chiller water through the copper coil condenser. Boiling-125 and Boiling-126 are open-chamber tests, where the Graham condenser is open throughout the tests, and the vapor pressure inside the chamber is the same as the ambient pressure. ## Code/software All files except for the DTA files can be opened using programming languages such as Python and MATLAB. They can also be opened using Excel, Notepad, or other sheet or text readers. Python users can follow the procedures documented in the tutorial of [SeqReg](https://github.com/cldunlap73/SeqReg) to read and use the dataset. The DTA file needs to be opened using the MISTRAS AEWin software, but it is not very often needed. For most analyses, the AEHit and waveforms are sufficient. The DTA file is provided in case users may want to export the hit data for different variables or in a different format.more » « less
-
Abstract: Two-phase cooling offers superior heat dissipation compared to conventional single-phase cooling methods. Nevertheless, the occurrence of critical heat flux (CHF) during boiling introduces reliability concerns, potentially leading to system failure. To improve system reliability, optical imaging is employed to analyze and monitor cooling systems without disrupting the boiling dynamics. These methods involve analyzing images of the boiling process to identify boiling regimes and evaluate heat flux. However, current optical-based methods are limited to static images, thereby missing out on the valuable temporal information captured by high-speed imaging. Inspired by the successful integration of temporal information in other fields, this work aims to exploit the temporal information from transient pool boiling captured via high-speed imaging for enhanced heat flux monitoring. For this purpose, two frameworks, comprising six different machine-learning models, have been developed for a comparative analysis. Specifically, the first framework includes two models that use static images for monitoring, serving as a representation of existing methodologies and a benchmark against which the second framework is measured. The remaining four models within the dynamic image-based framework (the 2nd framework) leverage sequences of images to capture temporal information. To evaluate the advantage of incorporating temporal information, transient boiling experiments were conducted to construct the dataset. A comparative analysis confirmed that temporal information significantly enhances the accuracy of the developed heat flux monitoring models. Among these models, the developed principal components (PCs)-convolutional neural network (CNN) stands out with a superior determination coefficient of 97.4% and a mean absolute percentage error of 7.0%, achieving an excellent balance between monitoring accuracy and computational efficiency. TechnicalInfo: # Data for: A temporal–spatial framework for efficient heat flux monitoring of transient boiling Dataset DOI: [10.5061/dryad.6m905qgc7](10.5061/dryad.6m905qgc7) ## Description of the data and file structure This dataset includes the temperature, acoustic (hydrophone), and videos of transient pool boiling tests of water on flat copper surfaces. The pool boiling test facility includes (a) a heating element that consists of a copper block and cartridge heaters; (b) a closed chamber with flow loops for a chiller (Thermo Scientific Polar ACCEL 500 Low/EA) connecting Graham condenser (Ace glass 5953-106) with an adapter (Ace glass 5838-76) and an in-house built coiled copper condenser; and (c) a synchronized multimodal sensing system. The copper block is submerged in deionized water and heated by nine cartridge heaters (Omega Engineering HDC19102), each with a power rating of 50 W, inserted from the bottom. The cartridge heaters are connected to the DC power supply (MagnaPower SL200-7.5) for controlling the heating power. The copper tube coiled condenser is used to keep the vapor pressure in check during the boiling test. This is done by controlling the flow rate passing through it manually using the flow metering valve with a Vernier handle (Swagelok SS-SS4-VH). Two screw-plug immersion heaters (McMaster-Carr 4668T54) are immersed in the liquid pool for degassing and auxiliary heating. A variable-voltage transformer (McMaster-Carr 6994K17) is used for adjusting the voltage required for the immersion heaters within the given range. The test facility is installed on a Thorlab Nexus optical table (T46V) with passive legs (PTP703) for vibration isolation. ### Files and variables #### File: ned3-006\_BoilingImageSequence.zip **Description:** This zip file includes eleven folders, i.e., Boiling-87, Boiling-89, Boiling-90, Boiling-91, Boiling-92, Boiling-93, Boiling-94, Boiling-96a, Boiling-96b, Boiling-96c, and Boiling-96d. Each folder, labeled with a test ID, Boiling-xxx, includes three files: * The Boiling-xxx_Temperature.lvm includes the temperature measurements using probe thermocouples (Omega Engineering Tj36-CPSS-032U-6) in the copper block and pool thermocouples in the water and vapor phases (McMaster-Carr 1245N16). These data are collected using an NI-9210 module. * The Boiling-xxx_Hydrophone.lvm file includes the continuously sampled acoustic data using two HTI-96-MIN hydrophones with an NI-9230 module. * The Boiling-xxx_video.mp4 file is a video for the boiling test. The original video (.cine) was captured using a Phantom VEO 710L high-speed camera. The .mp4 file was converted from the .cine file using the [Phantom Camera Control (PCC) software](https://www.phantomhighspeed.com/resourcesandsupport/phantomresources/pccsoftware). ## Code/software * The .txt and .lvm files can be opened and analyzed using programming languages such as Python and MATLAB. They can also be opened using Excel, Notepad, or other sheet or text readers. * The .mp4 file can be opened using a video player.more » « less
-
Abstract Power intensification and miniaturization of electronics and energy systems are causing a critical challenge for thermal management. Single-phase heat transfer mechanisms including natural and forced convection of air and liquids cannot meet the ever-increasing demands. Two-phase heat transfer modes, such as evaporation, pool boiling, flow boiling, have much higher cooling capacities but are limited by a variety of practical instabilities, e.g., the critical heat flux (CHF), aka departure from nucleate boiling (DNB) in the nuclear industry, flow maldistribution, flow reversal, among others. These instabilities are often triggered suddenly during normal operation, and if not identified and mitigated in time, will lead to overheating issues and detrimental device failures. For example, when CHF is triggered during pool boiling, the device temperature can ramp up in the order of 150 °C/min. It is thus critical to implement real-time detection and mitigation algorithms for two-phase cooling. In the present work, we have developed an accurate and reliable technology for fault detection of high-performance two-phase cooling systems by coupling acoustic emission (AE) with multimodal fusion using deep learning. We have leveraged the contact AE sensor attached to the heater and hydrophones immersed in the working fluid to enable non-invasive fault detection.more » « less
An official website of the United States government
