Finite element analysis on implicitly defined domains: An accurate representation based on arbitrary parametric surfaces

In this paper, we present some novel results and ideas for robust and accurate implicit representation of geometric surfaces in finite element analysis. The novel contributions of this paper are threefold: (1) describe and validate a method to represent arbitrary parametric surfaces implicitly; (2) represent arbitrary solids implicitly, including sharp features using level sets and boolean operations; (3) impose arbitrary Dirichlet and Neumann boundary conditions on the resulting implicitly defined boundaries. The methods proposed do not require local refinement of the finite element mesh in regions of high curvature, ensure the independence of the domain’s volume on the mesh, do not rely on boundary regularization, and are well suited to methods based on fixed grids such as the extended finite element method (XFEM). Numerical examples are presented to demonstrate the robustness and effectiveness of the proposed approach and show that it is possible to achieve optimal convergence rates using a fully implicit representation of object boundaries. This approach is one step in the desired direction of tying numerical simulations to computer aided design (CAD), similarly to the isogeometric analysis paradigm.     

Electric load forecasting by using dynamic neural network

Electrical energy is fundamental for the wellbeing and for the economic development of any country. However, all countries must ensure access to essential resources and ensure the continuity of its supply. Due to the non-storable nature of electrical energy, the amount of consumed active power should always be equal the produced active power just to avoid power system frequency deviation problem. In order to keep the relationship production–consumption relation in compliance with different standards and to secure profitable operations of power system, electric load consumption must be predicted and controlled instantaneously. Several statistical and classical techniques are proposed in the literature but unfortunately all these methods are not accurate in a satisfactory manner. In this paper, a dynamic neural network is used for the prediction of daily power consumption. The suitability and the performance of the proposed approach is illustrated and verified with simulations on load data collected from French Transmission System Operator (RTE) website. The obtained results show that the accuracy and the efficiency are improved comparatively to conventional methods widely used in this field of research.     

Characterization of nanostructured hydroxyapatite prepared by Nd:YAG laser deposition

Pulsed laser deposition, under dry and water vapor conditions, was employed to synthesize nanostructured hydroxyapatite films by pulsed laser deposition (PLD) of chlorapatite target for the purpose of coating metallic bone implants by this material. A pulsed Nd:YAG laser operating at a wavelength of 1064 nm and emitting 9 ns pulses was used for deposition. AFM microscopy, FTIR spectroscopy, optical microscopy, adhesion and microhardness measurements were conducted to characterize the films. The in vitro test for the synthesized hydroxyapatite was performed using simulated body fluid (SBF). The results showed a successful transformation of the chlorapatite to hydroxyapatite films characterized by all the HAp peaks with 60 nm root mean square roughness, (80–327) nm grain size, and a microhardness of 512 HV.     

Effect of organic eutectic on nitrocellulose stability during artificial aging

ABSTRACT

Recently, the eutectic composition of organic binary mixtures MENA+DPA has been found to be an efficient stabilizer of nitrocellulose. However, the behavior of such new stabilizer must be further investigated, especially after a long period of storage of the energetic composition. In this study, the stabilizing effect of the eutectic MENA+DPA on NC stability during artificial aging (at 338.65 K for 120 days), compared to its pure components, has been investigated by conventional stability tests and kinetic modeling. According to Bergmann & Junk (T = 405.15 K) results and over the entire period of aging, the eutectic MENA+DPA exhibits the best stabilizing effect on NC. The vacuum stability test (T = 353.15, 363.15, 373.15, and 383.15 K) results revealed that the amount of the evolved gas of NC containing this eutectic composition is very close to that of DPA, known as the conventional stabilizer. Furthermore, the kinetic modeling on VST data, obtained at four isothermal temperatures, was performed by two different methodologies, viz, fitting and free models to determine the kinetic triplets, which have been subsequently used to predict the storage lifetime for the studied systems. It was found that the NC stabilized by the eutectic MENA+DPA presents the highest value of the activation energy and has the longest storage lifetime during the entire heating period corroborating the stability tests results and affirming the excellent stabilizing effect of this mixture with respect to its pure component even if a long storage period is expected.     

Effectiveness of Deep Learning Models for Brain Tumor Classification and Segmentation

A brain tumor is a mass or growth of abnormal cells in the brain. In children and adults, brain tumor is considered one of the leading causes of death. There are several types of brain tumors, including benign (non-cancerous) and malignant (cancerous) tumors. Diagnosing brain tumors as early as possible is essential, as this can improve the chances of successful treatment and survival. Considering this problem, we bring forth a hybrid intelligent deep learning technique that uses several pre-trained models (Resnet50, Vgg16, Vgg19, U-Net) and their integration for computer-aided detection and localization systems in brain tumors. These pre-trained and integrated deep learning models have been used on the publicly available dataset from The Cancer Genome Atlas. The dataset consists of 120 patients. The pre-trained models have been used to classify tumor or no tumor images, while integrated models are applied to segment the tumor region correctly. We have evaluated their performance in terms of loss, accuracy, intersection over union, Jaccard distance, dice coefficient, and dice coefficient loss. From pre-trained models, the U-Net model achieves higher performance than other models by obtaining 95% accuracy. In contrast, U-Net with ResNet-50 outperforms all other models from integrated pre-trained models and correctly classified and segmented the tumor region.