Search

Three-dimensional (3D) clustering characteristics of large-stokes number sprays interacting with turbulent swirling co-flows are investigated experimentally. The Astigmatic Interferometric Particle Imaging (AIPI) technique is utilized for simultaneous measurement of the spray droplets position in 3D space and their corresponding diameter. The Stokes number estimated based on the Kolmogorov time scale varies from 34 to 142. The results show that the degree of droplet clustering plateaus at about 0.4 and at large Stokes numbers. It is obtained that the mean length scale of the clusters normalized by the Kolmogorov length scale follows a power-law relation with the Stokes number, and the mean void length scale normalized by the integral length scale plateaus at about 1.5. It is shown that the ratio of the number density of the droplets residing within the clusters to the global number density increases with increasing the Stokes number and is about 8 for the largest stokes number examined in this study. The joint characteristics of cluster's normalized volume and the mean diameter of droplets residing within the clusters show that small-volume clusters accommodate droplets with a relatively broad range of diameter. However, large clusters carry droplets with the most probable diameter. The developed AIPI technique in the present study and the corresponding spray characteristics are of importance for engineering applications that aim to understand the 3D clustering characteristics of large-stokes number droplets sprayed into turbulent swirling co-flows.

State-of-the-art rule-based and classification-based food recommendation systems face significant challenges in becoming practical and useful. This difficulty arises primarily because most machine learning models struggle with problems characterized by an almost infinite number of classes and a limited number of samples within an unbalanced dataset. Conversely, the emergence of Large Language Models (LLMs) as recommendation engines offers a promising avenue. However, a general-purpose Recommendation as Language Processing (RLP) approach lacks the critical components necessary for effective food recommendations. To address this gap, we introduce Food Recommendation as Language Processing (F-RLP), a novel framework that offers a food-specific, tailored infrastructure. F-RLP leverages the capabilities of LLMs to maximize their potential, thereby paving the way for more accurate, personalized food recommendations.

Separate and joint droplets, clusters, and voids characteristics of sprays injected in a turbulent co-flow are investigated experimentally. Simultaneous Mie scattering and Interferometric Laser Imaging for Droplet Sizing along with separate hotwire anemometry are performed. The turbulent co-flow characteristics are adjusted using zero, one, or two perforated plates. The Taylor length scale-based Reynolds number varies from 10 to 38, and the Stokes number estimated based on the Kolmogorov time scale varies from 3 to 25. The results show that the mean length scale of the clusters normalized by the Kolmogorov length scale varies linearly with the Stokes number. However, the mean void length scale is on the order of the integral length scale. It is shown that the number density of the droplets inside the clusters is about 10 times larger than that in the voids. The ratios of the droplets number densities in the clusters and voids to the total number density are independent of the test conditions and equal 8.3 and 0.7, respectively. The joint probability density function of the droplets diameter and clusters area shows that the droplets with the most probable diameter are found in the majority of the clusters. It is argued that intensifying the turbulence broadens the range of turbulent eddy size in the co-flow which allows for accommodating droplets with a broad range of diameters in the clusters. The results are of significance for engineering applications that aim to modify the clustering characteristics of large Stokes number droplets sprayed into turbulent co-flows.

A time-lag and linear regression-based framework is developed and its performance is assessed for predicting temporally resolved spray number of droplets using flame chemiluminescence and a sparse number of droplets data. Separate pressure, interferometric laser imaging for droplet sizing, shadowgraphy, and flame chemiluminescence are performed for the spray characterization. Simultaneous 10 kHz flame chemiluminescence and 0.2 Hz Mie scattering measurements are performed for the purposes of the framework development and the number of droplets prediction. Both methane and/or Jet A-1 are used in the experiments. Three conditions corresponding to perfectly premixed methane and air, Jet A-1 spray, and Jet A-1 spray in premixed methane and air flames are examined. For all test conditions, the fuels and air flow rates are adjusted to produce a fixed power of 10 kW. The results show that the frequency of the spatially averaged flame chemiluminescence as well as the number and the mass of the droplets (for both reacting and non-reacting conditions) oscillations frequencies match; however, these frequencies do not match that of the pressure fluctuations. This suggests that the flame chemiluminescence dynamics is driven by the fuel injection system. For signals with matching frequency content, a data-driven framework is developed for predicting an objective signal (the spray number of droplets) using an input signal (the flame chemiluminescence). The performance of the developed framework is assessed for tested spray conditions and the predicted number of droplets agrees well with those measured. For gas turbine engine combustion research, the developed framework is of importance, as it facilitates understanding the time-resolved spray characteristics for instances that the spray data is available sparsely.

Food is central to life. Food provides us with energy and foundational building blocks for our body and is also a major source of joy and new experiences. A significant part of the overall economy is related to food. Food science, distribution, processing, and consumption have been addressed by different communities using silos of computational approaches. In this paper, we adopt a person-centric multimedia and multimodal perspective on food computing and show how multimedia and food computing are synergistic and complementary. Enjoying food is a truly multimedia experience involving sight, taste, smell, and even sound, that can be captured using a multimedia food logger. The biological response to food can be captured using multimodal data streams using available wearable devices. Central to this approach is the Personal Food Model. Personal Food Model is the digitized representation of the food-related characteristics of an individual. It is designed to be used in food recommendation systems to provide eating-related recommendations that improve the user's quality of life. To model the food-related characteristics of each person, it is essential to capture their food-related enjoyment using a Preferential Personal Food Model and their biological response to food using their Biological Personal Food Model. Inspired by the power of 3-dimensional color models for visual processing, we introduce a 6-dimensional taste-space for capturing culinary characteristics as well as personal preferences. We use event mining approaches to relate food with other life and biological events to build a predictive model that could also be used effectively in emerging food recommendation systems.