Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Alternative relationships to enhance the applicability of nonlinear filtration models in porous media

Banerjee Ashes, Jagupilla Sarath Chandra K., Pasupuleti Srinivas, Annavarapu Chandra Sekhara Rao

Acta Geophysica

2023

Vol. 71, no. 4

1787--1799

Nonlinear filtration in porous packing has remained a research challenge till this day. There have been numerous attempts to model the flow characteristics under such conditions. However, as demonstrated in the present study, these models are applicable for only some specific conditions. The present study attempts to develop an empirical model which can be widely applicable. The Forchheimer-type models have been the most widely used in the literature for prediction of flow in porous media. The study identifies that the Ergun equation (the most popular form of the Forchheimer equation) with its original coefficients is unable to predict the flow properties over a wide range of data. Similar observation can be made for all other identical models. However, by optimising the coefficient values (A = 3705.79 and B = 6.17), the equation's performance can be significantly improved. The current study aims to create a working model that can be used to predict flow in porous media under a variety of packing, fluid, and flow conditions using multivariate polynomial regression and machine learning tools. It was observed that media size has far greater influence on the coefficients than any other parameter. Empirical models were created to predict Forchheimer coefficients, which represent R2 values greater than 0.9 for training, validation, and test data. These models were further tested on a separate dataset with velocity and hydraulic gradient data compiled from the literature. The models were found to have very reliable performance with R2 values above 0.90.

Multi-class nucleus detection and classification using deep convolutional neural network with enhanced high dimensional dissimilarity translation model on cervical cells

Karri Meghana, Annavarapu Chandra Sekhara Rao, Mallik Saurav, Zhao Zhongming, Acharya U. Rajendra

Biocybernetics and Biomedical Engineering

2022

Vol. 42, no. 3

797--814

Advanced cervical screening via liquid-based cytology (LBC)/Pap smear is a highly efficient precancerous cell detection tool based on cell image analysis, in which cells are classified as normal/abnormal. This paper outlines the drawbacks by introducing a new framework for the accurate classification of cervical cells. The proposed methodology comprises three phases: segmentation, localization of nucleus, and classification. In the segmentation phase, we develop a hybrid system that incorporates two binary image patches obtained by a 19-layered convolutional neural network (ConvNet) model with an enhanced deep high dimensional dissimilarity translation (HDDT) based conspicuous segmentation. To get the relevant information from binary patched images, a technique called optimum semantic similarity selective search (OSS-SS) is proposed that returns the localized RGB patched image. A pre-trained ResNet-50 model is retrained using transfer learning on localized patched images in the classification phase. Following that, the selected features from the average pool and fully connected layers are down-sampled using the t-distribution stochastic neighbor embedding (t-SNE) approach. Finally, these combined features are fed into a multi-class weighted kernel extreme learning machine (WKELM) classifier via a sparse multicanonical correlation (SMCCA) method. Three datasets (SIPaKMed, CRIC, and Harlev) are used to evaluate the segmentation and classification task. The proposed approach obtained an accuracy of 99.12 %, specificity of 99.45 %, sensitivity of 99.25 % with an execution time 99.6248 on SIPaKMed. The experimental analysis indicate that our model is more effective than existing techniques.