EventFlow: Real-time neuromorphic event-driven classification of two-phase boiling flow regimes

Abstract

Flow boiling is an efficient heat transfer mechanism capable of dissipating high heat loads with minimal temperature variation, making it an ideal thermal management method. However, sudden shifts between flow regimes can disrupt thermal performance and system reliability, highlighting the need for accurate and low-latency real-time monitoring. Conventional optical imaging methods are limited by high computational demands and insufficient temporal resolution, making them inadequate for capturing transient flow behavior. To address this, we propose a real-time framework based on signals from neuromorphic sensors for flow regime classification. Neuromorphic sensors detect changes in brightness at individual pixels, which typically correspond to motion at edges, enabling fast and efficient detection without full-frame reconstruction and providing event-based information. We develop five classification models using both traditional image data and event-based data, demonstrating that models leveraging event data outperform frame-based approaches due to their sensitivity to dynamic flow features. Among these models, the event-based long short-term memory model provides the best balance between accuracy and speed, achieving 97.6% classification accuracy with a processing time of 0.28 ms. Our asynchronous processing pipeline supports continuous, low-latency predictions and delivers stable output through a majority voting mechanisms, enabling reliable real-time feedback for experimental control and intelligent thermal management.

References

[1] Agonafer D, Spector MS, Miljkovic N. Materials and interface challenges in high-vapor-quality two-phase flow boiling research. IEEE Trans Compon Packaging Manuf Technol. 2021; 11: 1583-1591.

[2] Mousa MH, Yang C-M, Nawaz K, Miljkovic N. Review of heat transfer enhancement techniques in two-phase flows for highly efficient and sustainable cooling. Renew Sustain Energy Rev. 2022; 155:111896.

[3] Qu W, Mudawar I. Flow boiling heat transfer in two-phase micro-channel heat sinks. I. Experimental investigation and assessment of correlation methods. Int J Heat Mass Transf. 2003; 46: 2755-2771.

[4] Kharangate CR, O'Neill LE, Mudawar I, et al. Flow boiling and critical heat flux in horizontal channel with one-sided and double-sided heating. Int J Heat Mass Transf. 2015; 90: 323-338.

[5] Pop E. Energy dissipation and transport in nanoscale devices. Nano Res. 2010; 3: 147-169.

[6] Thome JR. The new frontier in heat transfer: microscale and nanoscale technologies. Heat Transfer Eng. 2006; 27: 1-3.

[7] Suh Y, Chandramowlishwaran A, Won Y. Recent progress of artificial intelligence for liquid‒vapor phase change heat transfer. NPJ Comput Mater. 2024; 10: 65.

[8] Suh Y, Bostanabad R, Won Y. Deep learning predicts boiling heat transfer. Sci Rep. 2021; 11: 5622.

[9] Ni S, Pan C, Hibiki T, Zhao J. Applications of nucleate boiling in renewable energy and thermal management and recent advances in modeling—a review. Energy. 2024; 289:129962.

[10] Azarifar M, Arik M, Chang J-Y. Liquid cooling of data centers: a necessity facing challenges. Appl Therm Eng. 2024; 247:123112.

[11] Wu X, Yang J, Liu Y, et al. Investigations on heat dissipation performance and overall characteristics of two-phase liquid immersion cooling systems for data center. Int J Heat Mass Transf. 2025; 239:126575.

[12] Mudawar I. Assessment of high-heat-flux thermal management schemes. IEEE Trans Compon Packag Technol. 2001; 24: 122-141.

[13] Huang C-N, Chang S, Suh Y, Mudawar I, Won Y, Kharangate CR. Machine learning boiling prediction: from autonomous vision of flow visualization data to performance parameter theoretical modeling. Int J Multiphase Flow. 2024; 179:104928.

[14] Kim SM, Mudawar I. Review of databases and predictive methods for pressure drop in adiabatic, condensing and boiling mini/micro-channel flows. Int J Heat Mass Transfer. 2014; 77: 74-97.

[15] Gungor KE, Winterton RHS. A general correlation for flow boiling in tubes and annuli. Heat Mass Transfer. 1986; 29: 351-358.

[16] Tran TN, Wambsganss MW, France DM. Small circular- and rectangular-channel boiling with two refrigerants. Pergamon Int J Multiphase Flow. 1996; 22: 485-498.

[17] Lee HJ, Lee SY. Heat transfer correlation for boiling flows in small rectangular horizontal channels with low aspect ratios. Int J Multiphase Flow. 2001; 27: 2043-2062.

[18] Inanlu MJ, Ganesan V, Upot NV, et al. Unveiling the fundamentals of flow boiling heat transfer enhancement on structured surfaces. Sci Adv. 2024; 10:eadp8632.

[19] Upot NV, Bakhshi A, Fazle Rabbi K, Lu F, Jacobi AM, Miljkovic N. Enhanced refrigerant flow boiling heat transfer in microstructured finned surfaces. Int J Heat Mass Transf. 2023; 207:123999.

[20] Upot NV, Mahvi A, Fazle Rabbi K, Li J, Jacobi AM, Miljkovic N. Scalable and resilient etched metallic micro- and nanostructured surfaces for enhanced flow boiling. ACS Appl Nano Mater. 2021; 4: 6648-6658.

[21] Reid RS, Pate MB, Bergles AE. Evaporation of refrigerant 113 flowing inside smooth tubes. J Heat Transf. 1987; 109: 685-702.

[22] Carey VP, Tervo P, Shollenberger K. Partial dryout in enhanced evaporator tubes and its impact on heat transfer performance. J Heat Transf. 1992; 114: 635-643.

[23] Hernández L, Julia JE, Ozar B, Hibiki T, Ishii M. Flow regime identification in boiling two-phase flow in a vertical annulus. J Fluids Eng. 2011; 133:091304.

[24] Huo X, Chen L, Tian YS, Karayiannis TG. Flow boiling and flow regimes in small diameter tubes. Appl Therm Eng. 2004; 24: 1225-1239.

[25] Ooi ZJ, Zhu L, Bottini JL, Brooks CS. Identification of flow regimes in boiling flows in a vertical annulus channel with machine learning techniques. Int J Heat Mass Transf. 2022; 185:122439.

[26] Nnabuife SG, Pilario KES, Lao L, Cao Y, Shafiee M. Identification of gas‒liquid flow regimes using a non-intrusive Doppler ultrasonic sensor and virtual flow regime maps. Flow Meas Instrum. 2019; 68:101568.

[27] Hervieu E. Identification of gas‒liquid flow regimes from a space-frequency representation by use of an impedance probe and a neural network. Proceedings of the ASME Fluids Engineering Division Summer Meeting, Montreal, Canada, 2002; 36169: 685-691.

[28] Jombo G, Zhang Y. Acoustic-based machine condition monitoring—methods and challenges. Eng. 2023; 4: 47-79.

[29] Hobold GM, da Silva AK. Machine learning classification of boiling regimes with low speed, direct and indirect visualization. Int J Heat Mass Transf. 2018; 125: 1296-1309.

[30] Kadish S, Schmid D, Son J, Boje E. Computer vision-based classification of flow regime and vapor quality in vertical two-phase flow. Sensors. 2022; 22: 996.

[31] Nie F, Wang H, Song Q, Zhao Y, Shen J, Gong M. Image identification for two-phase flow patterns based on CNN algorithms. Int J Multiphase Flow. 2022; 152:104067.

[32] Schepperle M, Junaid S, Woias P. Computer-vision- and deep-learning-based determination of flow regimes, void fraction, and resistance sensor data in microchannel flow boiling. Sensors. 2024; 24: 3363.

[33] Chang S, Suh Y, Shingote C, et al. BubbleMask: autonomous visualization of digital flow bubbles for predicting critical heat flux. Int J Heat Mass Transf. 2023; 217:124656.

[34] Suh Y, Chang S, Simadiris P, et al. VISION-iT: a framework for digitizing bubbles and droplets. Energy AI. 2024; 15:100309.

[35] Wang Y, Lv K, Huang R, Song S, Yang L, Huang G. Glance and focus: a dynamic approach to reducing spatial redundancy in image classification. Adv Neural Inf Process Syst. 2020; 33: 2432-2444.

[36] Furmonas J, Liobe J, Barzdenas V. Analytical review of event-based camera depth estimation methods and systems. Sensors. 2022; 22: 1201.

[37] Lu D, Suh Y, Won Y. Rapid identification of boiling crisis with event-based visual streaming analysis. Appl Therm Eng. 2024; 239:122004.

[38] Rusch A, Rösgen T. TrackAER: real-time event-based quantitative flow visualization. Exp Fluids. 2023; 64: 136.

[39] Willert CE, Klinner J. Event-based imaging velocimetry: an assessment of event-based cameras for the measurement of fluid flows. Exp Fluids. 2022; 63: 101.

[40] Hewitt G. Pressure drop, two-phase flow. Thermopedia. Begel House Inc.; 2011.

[41] Kim S-M, Mudawar I. Theoretical model for local heat transfer coefficient for annular flow boiling in circular mini/micro-channels. Int J Heat Mass Transf. 2014; 73: 731-742.

[42] Kong R, Kim S. Characterization of horizontal air‒water two-phase flow. Nucl Eng Des. 2017; 312: 266-276.

[43] Emamzadeh M, Issa RI. A model for predicting the transition between stratified and annular flow in horizontal pipes. Multiph Sci Technol. 2013; 25: 79-100.

[44] Wang Y, Yu Y, Liu Z, Chang Y, Zhao X, Wang Q. Experimental study on gas‒liquid two-phase stratified flow at high pressure in a horizontal pipe. Energies (Basel). 2024; 17: 1056.

[45] O'Neill LE, Mudawar I. Review of two-phase flow instabilities in macro- and micro-channel systems. Int J Heat Mass Transf. 2020; 157:119738.

[46] Gehrig D, Scaramuzza D. Low-latency automotive vision with event cameras. Nature. 2024; 629: 1034-1040.

[47] O'Shea K, Nash R. An introduction to convolutional neural networks. arXiv preprint. 2015; arXiv:1511.08458.

[48] Peterson LE. K-nearest neighbor. Scholarpedia. 2009; 4: 1883.

[49] Ghosh-Dastidar S, Adeli H. Spiking neural networks. Int J Neural Syst. 2009; 19: 295-308.

[50] Shi X, Chen Z, Wang H, Yeung D-Y, Wong W-K, Woo W-C. Convolutional LSTM network: a machine learning approach for precipitation nowcasting. Adv Neural Inf Process Syst. 2015; 28: 802-810.

[51] Eshraghian JK, Ward M, Neftci EO, et al. Training spiking neural networks using lessons from deep learning. Proc IEEE. 2023; 111: 1016-1054.

[52] Zhemchuzhnikova D, Lebyodkin M, Lebedkina T, Mogucheva A, Yuzbekova D, Kaibyshev R. Peculiar spatiotemporal behavior of unstable plastic flow in an AlMgMnScZr alloy with coarse and ultrafine grains. Metals (Basel). 2017; 7: 325.

[53] Manneville P. Understanding the sub-critical transition to turbulence in wall flows. Pramana. 2008; 70: 1009-1021.

[54] Mousa MH, Rao R, Yang C-M, et al. Saturated water flow boiling heat transfer and pressure drop in scalably microstructured plain and finned copper tubes. Int J Heat Mass Transf. 2024; 222:125138.

[55] Mousa MH, Roni MRH, Rao R, et al. Water flow boiling heat transfer and pressure drop in smooth, etched, and herringbone aluminum tubes. Appl Therm Eng. 2024; 257:124426.

[56] Hu Y, Liu S-C, Delbruck T. v2e: from video frames to realistic DVS events. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA, 2021: 1312-1321.

Rights & permissions

2026 The Author(s). Droplet published by Jilin University and John Wiley & Sons Australia, Ltd.