A three-stage learning algorithm for deep multilayer perceptron (DMLP) with effective weight initialisation based on sparse auto-encoder is proposed in this paper, which aims to overcome difficulties in training deep neural networks with limited training data in high-dimensional feature space. At the first stage, unsupervised learning is adopted using sparse auto-encoder to obtain the initial weights of the feature extraction layers of the DMLP. At the second stage, error back-propagation is used to train the DMLP by fixing the weights obtained at the first stage for its feature extraction layers. At the third stage, all the weights of the DMLP obtained at the second stage are refined by error back-propagation. Network structures an

The present study reports the effect of temperature and liquid hourly space velocity (LHSV) on the cumene cracking reaction rate and selectivity by using a laboratory continuous flow unit with fixed bed reactor operating at atmospheric pressure. The prepared HX zeolite was made from Iraqi kaolin with good crystallinity .The activity and selectivity of prepared HX-zeolite was compared with standard HY zeolite and HX zeolite catalysts in the temperature range of 673-823K and LHSV of 0.7-2.5 h-1 . It was found that the cumene conversion increases with increasing temperature and decreasing LHSV at 823K and LHSV of 0.7 h-1 the conversions 65.32, 42.88 and 59.42 mol% for HY, HX and prepared HX catalysts respectively and at LHSV of 2.5 h-1 and th

     The bubble columns are widely used as a two or three phase reactor in industrial chemical process such as absorption, biochemical reactions, coal liquefaction, etc. To design such a column, two main parameters should be taken in consideration, the gas hold-up (), and the liquid phase mass transfer coefficient K_La. The study includes the effect of gas velocity and the addition of alcohols on gas hold-up and mass transfer coefficient in bubble column with draught tube when the length of the column is 1.5m and the ratio of the draught tube diameter to the column diameter equals 0.5 and the air dispersion into the base of the draught tube using a multi hole tuyere is equivalent to a diameter of 0.15 mm and

The petrophysical analysis is very important to understand the factors controlling the reservoir quality and production wells. In the current study, the petrophysical evaluation was accomplished to hydrocarbon assessment based on well log data of four wells of Early Cretaceous carbonate reservoir Yamama Formation in Abu-Amood oil field in the southern part of Iraq. The available well logs such as sonic, density, neutron, gamma ray, SP, and resistivity logs for wells AAm-1, AAm-2, AAm-3, and AAm-5 were used to delineate the reservoir characteristics of the Yamama Formation. Lithologic and mineralogic studies were performed using porosity logs combination cross plots such as density vs. neutron cross plot and M-N mineralogy plot. Thes

The increase globally fossil fuel consumption as it represents the main source of energy around the world, and the sources of heavy oil more than light, different techniques were used to reduce the viscosity and increase mobility of heavy crude oil. this study focusing on the experimental tests  and modeling with Back Feed Forward Artificial Neural Network (BFF-ANN) of the dilution technique to reduce a  heavy oil viscosity that was collected from the south- Iraq oil fields using organic solvents, organic diluents with different weight percentage  (5, 10 and  20 wt.% )  of  (n-heptane, toluene, and a mixture of  different ratio

well log analysis is used to determine the rock properties like porosity, water saturation, and shale volume. Archie parameters in Archie equation, which sometimes considered constants greatly affect the determination of water saturation, also these parameters may be used to indicate whether the rocks are fractured or not so they should be determined. This research involves well logging analysis for Zubair formation in Luhais field which involves the determination of Archie parameters instead of using them as constant.

The log interpretation proved that the formation is hydrocarbon reservoir, as it could be concluded from Rwa (high values) and water saturation values (low values), the lithology of Zubair from cro

Monaural source separation is a challenging issue due to the fact that there is only a single channel available; however, there is an unlimited range of possible solutions. In this paper, a monaural source separation model based hybrid deep learning model, which consists of convolution neural network (CNN), dense neural network (DNN) and recurrent neural network (RNN), will be presented. A trial and error method will be used to optimize the number of layers in the proposed model. Moreover, the effects of the learning rate, optimization algorithms, and the number of epochs on the separation performance will be explored. Our model was evaluated using the MIR-1K dataset for singing voice separation. Moreover, the proposed approach achi

The proliferation of many editing programs based on artificial intelligence techniques has contributed to the emergence of deepfake technology. Deepfakes are committed to fabricating and falsifying facts by making a person do actions or say words that he never did or said. So that developing an algorithm for deepfakes detection is very important to discriminate real from fake media. Convolutional neural networks (CNNs) are among the most complex classifiers, but choosing the nature of the data fed to these networks is extremely important. For this reason, we capture fine texture details of input data frames using 16 Gabor filters indifferent directions and then feed them to a binary CNN classifier instead of using the red-green-blue

The rapid rise in the use of artificially generated faces has significantly increased the risk of identity theft in biometric authentication systems. Modern facial recognition technologies are now vulnerable to sophisticated attacks using printed images, replayed videos, and highly realistic 3D masks. This creates an urgent need for advanced, reliable, and mobile-compatible fake face detection systems. Research indicates that while deep learning models have demonstrated strong performance in detecting artificially generated faces, deploying these models on consumer mobile devices remains challenging due to limitations in computing power, memory, privacy, and processing speed. This paper highlights several key challenges: (1) optimiz

Estimating an individual's age from a photograph of their face is critical in many applications, including intelligence and defense, border security and human-machine interaction, as well as soft biometric recognition. There has been recent progress in this discipline that focuses on the idea of deep learning. These solutions need the creation and training of deep neural networks for the sole purpose of resolving this issue. In addition, pre-trained deep neural networks are utilized in the research process for the purpose of facial recognition and fine-tuning for accurate outcomes. The purpose of this study was to offer a method for estimating human ages from the frontal view of the face in a manner that is as accurate as possible and takes