Streamflow prediction is a long‐standing hydrologic problem. Development of models for streamflow prediction often requires incorporation of catchment physical descriptors to characterize the associated complex hydrological processes. Across different scales of catchments, these physical descriptors also allow models to extrapolate hydrologic information from one catchment to others, a process referred to as “regionalization”. Recently, in gauged basin scenarios, deep learning models have been shown to achieve state of the art regionalization performance by building a global hydrologic model. These models predict streamflow given catchment physical descriptors and weather forcing data. However, these physical descriptors are by their nature uncertain, sometimes incomplete, or even unavailable in certain cases, which limits the applicability of this approach. In this paper, we show that by assigning a vector of random values as a surrogate for catchment physical descriptors, we can achieve robust regionalization performance under a gauged prediction scenario. Our results show that the deep learning model using our proposed random vector approach achieves a predictive performance comparable to that of the model using actual physical descriptors. The random vector approach yields robust performance under different data sparsity scenarios and deep learning model selections. Furthermore, based on the use of random vectors, high‐dimensional characterization improves regionalization performance in gauged basin scenario when physical descriptors are uncertain, or insufficient.
Condensation is ubiquitous in nature and industry. Heterogeneous condensation on surfaces is typified by the continuous cycle of droplet nucleation, growth, and departure. Central to the mechanistic understanding of the thermofluidic processes governing condensation is the rapid and high‐fidelity extraction of interpretable physical descriptors from the highly transient droplet population. However, extracting quantifiable measures out of dynamic objects with conventional imaging technologies poses a challenge to researchers. Here, an intelligent vision‐based framework is demonstrated that unites classical thermofluidic imaging techniques with deep learning to fundamentally address this challenge. The deep learning framework can autonomously harness physical descriptors and quantify thermal performance at extreme spatio‐temporal resolutions of 300 nm and 200 ms, respectively. The data‐centric analysis conclusively shows that contrary to classical understanding, the overall condensation performance is governed by a key tradeoff between heat transfer rate per individual droplet and droplet population density. The vision‐based approach presents a powerful tool for the study of not only phase‐change processes but also any nucleation‐based process within and beyond the thermal science community through the harnessing of big data.
more » « less- Award ID(s):
- 1752147
- NSF-PAR ID:
- 10366488
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Advanced Science
- Volume:
- 8
- Issue:
- 22
- ISSN:
- 2198-3844
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract -
Atmospheric condensation is very important for multiple practical applications such as heat transfer, thermal management, aerospace, and condensate harvesting. Water droplets heterogeneously nucleate on the surfaces when the temperature is below the dew point temperature. The nucleation energy barrier for a condensed droplet varies significantly with the humidity content in the operating environment. The freezing of this condensate is also dependent on the operating conditions and surface properties. This article presents an experimental study of condensation and freezing from humid air with the objective of understanding how the surface morphology and chemistry determines the droplet shape and wetting state. Hexagonal close-packed arrays of titanium (Ti) pillars are patterned using microsphere photolithography (MPL). The Ti nanostructured surface was tested with and without a Teflon© coating to reveal the condensate harvesting, passive freezing, and dropwise condensation applications, respectively. Condensation and freezing tests were conducted in the presence of non-condensable gases (air) with different relative humidity (RH) levels to control the nucleation site density. The experiments showed that droplet growth occurs in the following stages: initial nucleation, direct growth, and coalescence events. By pinning droplets, coalescence is suppressed for the Ti nanopillared surface altering the size distribution of droplets and significantly accelerating the freezing process.more » « less
-
Droplet nucleation and condensation are ubiquitous phenomena in nature and industry. Over the past century, research has shown dropwise condensation heat transfer on nonwetting surfaces to be an order of magnitude higher than filmwise condensation heat transfer on wetting substrates. However, the necessity for nonwetting to achieve dropwise condensation is unclear. This article reports stable dropwise condensation on a smooth, solid, hydrophilic surface (θ a = 38°) having low contact angle hysteresis (<3°). We show that the distribution of nano- to micro- to macroscale droplet sizes (about 100 nm to 1 mm) for coalescing droplets agrees well with the classical distribution on hydrophobic surfaces and elucidate that the wettability-governed dropwise-to-filmwise transition is mediated by the departing droplet Bond number. Our findings demonstrate that achieving stable dropwise condensation is not governed by surface intrinsic wettability, as assumed for the past eight decades, but rather, it is dictated by contact angle hysteresis.more » « less
-
Abstract Boiling is arguably Nature’s most effective thermal management mechanism that cools submersed matter through bubble-induced advective transport. Central to the boiling process is the development of bubbles. Connecting boiling physics with bubble dynamics is an important, yet daunting challenge because of the intrinsically complex and high dimensional of bubble dynamics. Here, we introduce a data-driven learning framework that correlates high-quality imaging on dynamic bubbles with associated boiling curves. The framework leverages cutting-edge deep learning models including convolutional neural networks and object detection algorithms to automatically extract both hierarchical and physics-based features. By training on these features, our model learns physical boiling laws that statistically describe the manner in which bubbles nucleate, coalesce, and depart under boiling conditions, enabling in situ boiling curve prediction with a mean error of 6%. Our framework offers an automated, learning-based, alternative to conventional boiling heat transfer metrology.more » « less
-
more » « less
Abstract Artificial intelligence (AI) is shifting the paradigm of two-phase heat transfer research. Recent innovations in AI and machine learning uniquely offer the potential for collecting new types of physically meaningful features that have not been addressed in the past, for making their insights available to other domains, and for solving for physical quantities based on first principles for phase-change thermofluidic systems. This review outlines core ideas of current AI technologies connected to thermal energy science to illustrate how they can be used to push the limit of our knowledge boundaries about boiling and condensation phenomena. AI technologies for meta-analysis, data extraction, and data stream analysis are described with their potential challenges, opportunities, and alternative approaches. Finally, we offer outlooks and perspectives regarding physics-centered machine learning, sustainable cyberinfrastructures, and multidisciplinary efforts that will help foster the growing trend of AI for phase-change heat and mass transfer.
