A Deep Learning Perspective on Dropwise Condensation

2020

In this article we report the atypical and anomalous evaporation kinetics of saline sessile droplets on surfaces with elevated temperatures. In a previous we showed that saline sessile droplets evaporate faster compared to water droplets when the substrates are not heated. In the present study we discover that in the case of heated surfaces, the saline droplets evaporate slower than the water counterpart, thereby posing a counter-intuitive phenomenon. The reduction in the evaporation rates is directly dependent on the salt concentration and the surface wettability. Natural convection around the droplet and thermal modulation of surface tension is found to be inadequate to explain the mechanisms. Flow visualisations using particle image velocimetry PIV reveals that the morphed advection within the saline droplets is a probable reason behind the arrested evaporation. Infrared thermography is employed to map the thermal state of the droplets. A thermosolutal Marangoni based scaling ana...

Sensors

Liquid crystal (LC)-based materials are promising platforms to develop rapid, miniaturised and low-cost gas sensor devices. In hybrid gel films containing LC droplets, characteristic optical texture variations are observed due to orientational transitions of LC molecules in the presence of distinct volatile organic compounds (VOC). Here, we investigate the use of deep convolutional neural networks (CNN) as pattern recognition systems to analyse optical textures dynamics in LC droplets exposed to a set of different VOCs. LC droplets responses to VOCs were video recorded under polarised optical microscopy (POM). CNNs were then used to extract features from the responses and, in separate tasks, to recognise and quantify the vapours exposed to the films. The impact of droplet diameter on the results was also analysed. With our classification models, we show that a single individual droplet can recognise 11 VOCs with small structural and functional differences (F1-score above 93%). The o...

Energies

The methods used in chemical engineering are strongly reliant on having a solid grasp of the thermodynamic features of complex systems. It is difficult to define the behavior of ions and molecules in complex systems and to make reliable predictions about the thermodynamic features of complex systems across a wide range. Deep learning (DL), which can provide explanations for intricate interactions that are beyond the scope of traditional mathematical functions, would appear to be an effective solution to this problem. In this brief Perspective, we provide an overview of DL and review several of its possible applications within the realm of chemical engineering. DL approaches to anticipate the molecular thermodynamic characteristics of a broad range of systems based on the data that are already available are also described, with numerous cases serving as illustrations.

SSRN Electronic Journal

Powder-based additive manufacturing has transformed the manufacturing industry over the last decade. In Laser Powder Bed Fusion, a specific part is built in an iterative manner in which two-dimensional cross-sections are formed on top of each other by melting and fusing the proper areas of the powder bed. In this process, the behavior of the melt pool and its thermal field has a very important role in predicting the quality of the manufactured part and its possible defects. However, the simulation of such a complex phenomenon is usually very time-consuming and requires huge computational resources. Flow-3D is one of the software packages capable of executing such simulations using iterative numerical solvers. In this work, we create three datasets of single-trail processes using Flow-3D and use them to train a convolutional neural network capable of predicting the behavior of the three-dimensional thermal field of the melt pool solely by taking three parameters as input: laser power, laser velocity, and time step. The CNN achieves a relative Root Mean Squared Error of 2% to 3% for the temperature field and an average Intersection over Union score of 80% to 90% in predicting the melt pool area. Moreover, since time is included as one of the inputs of the model, the thermal field can be instantly obtained for any arbitrary time step without the need to iterate and compute all the steps.

A B S T R A C T This paper explores the use of time resolved infrared (IR) thermography combined with high-speed imaging to describe the liquid-surface interfacial heat transfer phenomena occurring at droplet/wall interactions. A first set of experiments is used to infer on the validation of custom made calibration and post-processing methods to evaluate the potential to obtain accurate data, useful to describe the observed phenomena. In this context, the technique showed good potential to capture very well particular details on droplet dynamics and heat transfer, allowing to identify air bubble trapping at the impact region, as well as the temperature variations at the formation of the rim. The combined analysis of droplet dynamics (e.g. the spreading factor) with the radial temperature profiles, heat flux and cooling effectiveness allowed establishing qualitative and quantitative trends on the effect of various parameters on the heat transfer occurring at droplet/wall interactions. Particularly, details on the temperature distribution within the lamella, reflected on the surface temperature fields, which are related to the spreading dynamics, are explored and discussed. Furthermore, extreme wetting scenarios, such as superhydrophobicity are studied in detail. The analysis performed show that they limit the heat transfer between the droplet and the surface. However, the thermal analysis coupled with the spreading dynamics reveals that a major reason for this is not just related to the reduced contact time of the droplet (due to rebound) or air entrapment, but is also associated to the reduced wetted area caused by low wettability, which is not obvious from the analysis of the spreading diameter alone, but becomes evident when this is coupled with the evaluation of the temperature fields on the heated surface.

arXiv (Cornell University), 2024

We examine five machine learning-based architectures to estimate the droplet size distributions obtained using digital inline holography. The architectures, namely, U-Net, R2 U-Net, Attention U-Net, V-Net, and Residual U-Net are trained using synthetic holographic images. Our assessment focuses on evaluating the training, validation, and prediction performance of these architectures. We found that U-Net and R2 U-Net to be the most proficient, displaying consistent performance trends and achieving the highest Intersection Over Union (IOU) scores compared to the other three architectures. We employ additional training using experimental holographic images for the two top-performing architectures to validate their efficacy further. Subsequently, they are employed to segment an experimental dataset illustrating the bag breakup phenomenon, facilitating the extraction of size distribution. The extracted size distribution from U-Net and R2 U-Net segmentation is then compared with the analytical model proposed by Jackiw and Ashgriz (2022) by employing the gamma and log-normal distributions. Our findings indicate that the gamma distribution provides a more accurate prediction of the multi-modal size distribution than the log-normal distribution owing to its long exponential tail. The present study

2020

This study provides an insight to the analysis of time-averaged Heat Transfer Coefficient(HTC), water collection rate and pattern of drop-size distribution, employing IR thermography and specialized imaging software, under different surface wettabilityconditions. We performed an experimental analysis under free convection, with two different sets of surface conditions (hydrophilic and hydrophobic) on a smooth, vertical glass surface, exposed to a quiescent environment containing humid air. Experimental results showed that the HTC (time averaged) for hydrophobic surface is greater than that on the hydrophilic surface under same set of test conditions. Hydrophobic surface yielded faster water collection rate compared to the hydrophilic one. Distribution of droplet population over the range of 0.1 – 1 mm diameter is seen to have a decreasing trend with increasing drop size, while droplets smaller than 0.1mm diameter covered the major fraction of the heat transfer

2021

This paper presents an investigation of the rapid variations in the temperature of metal melt pool for Additive Manufacturing (AM) processes. The melt pool is created by scanning a high-power laser beam across a metal powder bed. Rapid heating and cooling processes are involved in the layer-by-layer fabrication of the metal part. Recent advances in Machine Learning and Deep Learning algorithms provide efficient ways to analyze large sets of data in search of correlations that would otherwise be extremely time-consuming. The use of Machine Learning and Deep Learning algorithms to understand temperature variations in AM fabrication process will allow to predict the formation of porosity before it occurs. The objective of this research is to advance the AM technology using enhanced Deep Learning techniques to provide in-situ analysis of the melt pool temperature that can lead to a reliable manufacturing of Three-Dimensional (3D) metal parts/components. In specific, Deep Learning based ...

Scientific Reports, 2022

One major difficulty to achieve such goals is to be able to analyze in a robust way a large number of objects in order to draw out conclusions with a sufficient statistical meaning. A basic and simple approach is to count and measure manually or semi-automatically a few particles during, typically, calcination or reduction treatments in temperature: this has been performed in some cases to study the growth of NPs during coalescence and/or Ostwald ripening ETEM 35-38. Such a manual processing quickly becomes impractical when dealing with a large population in a dynamic approach in order to track realistically the evolution of the catalysts in time using video recording. The purpose of the present work is to develop algorithms and methodologies in order to treat automatically sequences of images recorded during in situ ETEM observations of metal-based nanocatalysts supported on an oxide media under gas at high temperatures. We first survey briefly the main algorithmic approaches employed so far.

Physical Review E

The dynamics of the condensation process on nanostructured surfaces can be modulated substantially by tuning the surface architecture. Present study uses the mesoscopic framework of lattice Boltzmann method (LBM) to explore the role of surface morphology and cold spot temperature in determining the visual state of the condensate droplet, mode of nucleation, and associated rates of energy and mass interactions. A multiple relaxation time-(MRT)-based LBM solver, coupled with pseudopotential model, has been developed to simulate a rectangular domain of saturated vapor, housing a cold spot on the bottom rough surface. Superhydrophobicity has been achieved for certain combinations of surface parameters, with the intercolumn spacing being the most influential one. Gradual increase in the spacing modifies the nucleation mode from top through side to bottom, while the droplet changes from Cassie to Wenzel state. The Cassie drop in top nucleation mode exhibits the largest contact angle and least rate of surface heat transfer. Both types of Wenzel drops display large rate of condensation and two peaks in heat transfer, along with very short nucleation time in comparison with Cassie drops. Couple of phase diagrams have been developed combining all four scenarios of condensation predicted by the present model. One important novelty of the present study is the consideration of nonisothermal condition within LB structure. Enhancement in the degree of subcooling at the cold spot encourages greater condensation and Cassie-to-Wenzel transition.

Energies

This study develops a geometry adaptive, physical field predictor for the combined forced and natural convection flow of a nanofluid in horizontal single or double-inner cylinder annular pipes with various inner cylinder sizes and placements based on deep learning. The predictor is built with a convolutional-deconvolutional structure, where the input is the annulus cross-section geometry and the output is the temperature and the Nusselt number for the nanofluid-filled annulus. Profiting from the proven ability of dealing with pixel-like data, the convolutional neural network (CNN)-based predictor enables an accurate end-to-end mapping from the geometry input and the desired nanofluid physical field. Taking the computational fluid dynamics (CFD) calculation as the basis of our approach, the obtained results show that the average accuracy of the predicted temperature field and the coefficient of determination R2 are more than 99.9% and 0.998 accurate for single-inner cylinder nanoflui...

ArXiv, 2018

Boiling heat transfer occurs in many situations and can be used for thermal management in various engineered systems with high energy density, from power electronics to heat exchangers in power plants and nuclear reactors. Essentially, boiling is a complex physical process that involves interactions between heating surface, liquid, and vapor. For engineering applications, the boiling heat transfer is usually predicted by empirical correlations or semi-empirical models, which has relatively large uncertainty. In this paper, a data-driven approach based on deep feedforward neural networks is studied. The proposed networks use near wall local features to predict the boiling heat transfer. The inputs of networks include the local momentum and energy convective transport, pressure gradients, turbulent viscosity, and surface information. The outputs of the networks are the quantities of interest of a typical boiling system, including heat transfer components, wall superheat, and near wall...

Macromolecules

Machine learning (ML) and artificial intelligence (AI) have the remarkable ability to classify, recognize, and characterize complex patterns and trends in large data sets. Here, we adopt a subclass of machine learning methods viz., deep learnings and develop a general-purpose AI tool-dPOLY for analyzing molecular dynamics trajectory and predicting phases and phase transitions in polymers. An unsupervised deep neural network is used within this framework to map a molecular dynamics trajectory undergoing thermophysical treatment such as cooling, heating, drying, or compression to a lower dimension. A supervised deep neural network is subsequently developed based on the lower dimensional data to characterize the phases and phase transition. As a proof of concept, we employ this framework to study coil to globule transition of a model polymer system. We conduct coarse-grained molecular dynamics simulations to collect molecular dynamics trajectories of a single polymer chain over a wide range of temperatures and use dPOLY framework to predict polymer phases. The dPOLY framework accurately predicts the critical temperatures for the coil to globule transition for a wide range of polymer sizes. This method is generic and can be extended to capture various other phase transitions and dynamical crossovers in polymers and other soft materials.

Applied Sciences, 2022

Much of the earth’s surface is covered by water. As was pointed out in the 2020 edition of the World Water Development Report, climate change challenges the sustainability of global water resources, so it is important to monitor the quality of water to preserve sustainable water resources. Quality of water can be related to the structure of water crystal, the solid-state of water, so methods to understand water crystals can help to improve water quality. As a first step, a water crystal exploratory analysis has been initiated with the cooperation with the Emoto Peace Project (EPP). The 5K EPP dataset has been created as the first world-wide small dataset of water crystals. Our research focused on reducing the inherent limitations when fitting machine learning models to the 5K EPP dataset. One major result is the classification of water crystals and how to split our small dataset into several related groups. Using the 5K EPP dataset of human observations and past research on snow cry...

Proceedings of the 5th World Congress on Mechanical, Chemical, and Material Engineering, 2019

Dropwise condensation is one of the regimes leading to the best heat performance for managing large heat fluxes. In order to model the transfers in this regime, knowledge of the drop size distribution is fundamental. In this paper, two different approaches are proposed and analyzed to determine this distribution. The first one is based on a statistical model, in which two populations of drops are considered and where the steady state is reached. The second model uses a more direct approach, modeling each drop on the studied surface. Assuming instantaneous coalescence when two droplets overlap, the distribution is calculated in transient regime. The mean value of this distribution in permanent regime is then compared to the distribution deduced from available statistical approach in the literature. Good adequation is obtained for the “big” droplets, while important discrepancies are highlighted for the “small” droplets.