Development of multilayer perceptron artificial neural network (MLP-ANN) and least square support vector machine (LSSVM) models to predict Nusselt number and pressure drop of TiO2/water nanofluid flows through non-straight pathways

Abstract

In this study, Multilayer Perceptron Artificial Neural Network (MLP-ANN) model and Least Square Support Vector Machine (LSSVM), were developed to predict the thermal performance and pressure loss of nanofluid flow through coils as non-straight pathways. There different coils with various curvature ratios and coil pitches were constructed and used. Stable TiO₂ (50?nm)/water nanofluid in different concentrations from 0.0 to 2.0% were prepared using appropriate method. As it is expected, considerable enhancement of heat transfer was achieved by application of nanofluids instead of water in system. Volume concentration of nanofluid, Prandtl number (ranging from 4.82 to 9.11) and Helical number (106.80 to 1282.87) were introduced to the developed models to obtain Nusselt number (9.89 to 53.30) and pressure drop (291.35 to 18784?kPa) as the output data of the models. According to the output results of developed models, MLP-ANN model was able to predict both Nusselt number and pressure drop of nanofluid flow more precisely in comparison to LSSVM model. The developed MLP model of this study exceeded LSSVM model to high correlation coefficient value of 0.97.     

Improved Incompressible Smoothed Particle Hydrodynamics method for simulating flow around bluff bodies

In this article, we present numerical solutions for flow over an airfoil and a square obstacle using Incompressible Smoothed Particle Hydrodynamics (ISPH) method with an improved solid boundary treatment approach, referred to as the Multiple Boundary Tangents (MBT) method. It was shown that the MBT boundary treatment technique is very effective for tackling boundaries of complex shapes. Also, we have proposed the usage of the repulsive component of the Lennard-Jones Potential (LJP) in the advection equation to repair particle fractures occurring in the SPH method due to the tendency of SPH particles to follow the stream line trajectory. This approach is named as the artificial particle displacement method. Numerical results suggest that the improved ISPH method which is consisting of the MBT method, artificial particle displacement and the corrective SPH discretization scheme enables one to obtain very stable and robust SPH simulations. The square obstacle and NACA airfoil geometry with the angle of attacks between 0° and 15° were simulated in a laminar flow field with relatively high Reynolds numbers. We illustrated that the improved ISPH method is able to capture the complex physics of bluff-body flows naturally such as the flow separation, wake formation at the trailing edge, and the vortex shedding. The SPH results are validated with a mesh-dependent Finite Element Method (FEM) and excellent agreements among the results were observed.     

Parallel variable neighborhood search for solving fuzzy multi-objective dynamic facility layout problem

In spite of the classic approaches of solution of dynamic facility layout problem, which only material handling and rearrangement costs are considered as objective function, these problems are the multi-objective problems. In this paper, a mixed integer linear programming formulation is presented for multi-objective dynamic facility layout problem concerning flexible bay structure. In addition, three current objectives in dynamic facility layout problems including minimizing material handling and rearrangement costs, maximizing adjacency rate, and minimizing shape ratio difference have been considered. Also, for solving this problem, two methods including the GAMS software and proposed parallel variable neighborhood search (PVNS) algorithm are used. So, it is worth mentioning that four test problems are solved by them, and the results show that the proposed PVNS algorithm is more efficient than the GAMS software.     

Market segmentation of organ donors in Egypt: a bio-inspired computational intelligence approach

First performed in 1954, organ transplantation is a universally practiced clinical procedure. This study uses ant colony optimization (ACO), radial basis function neural network (RBFNN), Kohonen’s self-organizing maps (SOM), and support vector machines (SVMs) to examine the effect of various cognitive, psychographic, and attitudinal factors on organ donation. ACO, RBFNN, SOM, and SVMs are compared to a standard statistical method (linear discriminant analysis [LDA]). The variable sets considered are altruistic values, perceived risks/benefits, knowledge, attitudes toward organ donation, and intention to donate organs. The paper shows how it is possible to identify various dimensions of organ donation behavior by uncovering complex patterns in the dataset and also shows the classification and clustering abilities of machine-learning systems.     

The impacts of capabilities and constraints of generating units on simultaneous scheduling of energy and primary reserve

Frequency control, as an ancillary service, is usually provided by generation reserves. Modern generating units have special technical capabilities; e.g., their governor operation mode can be selected to be either active or passive, their ramp rate can be selected to be either normal or fast, etc. On the other hand, generating units have some technical constraints; e.g. some generating units cannot participate in primary frequency control at their capacity limits. In this paper, operational capabilities and constraints of generating units are incorporated in a “simultaneous scheduling of energy and primary reserve” problem. Furthermore, a heuristic iterative method (based on genetic algorithm) is proposed to obtain the optimal scheduling. The impacts of capabilities and constraints on scheduling are investigated through simulation studies. Simulation results depicts that taking these capabilities and constraints of generating units into account, not only reduces the total operational cost of generating units, but also will end up with a feasible solution for the system, even in cases where the previously proposed methods fail.     

Evolving comprehensible neural network trees using genetic algorithms

This paper presents a comprehensible neural network tree (CNNTREE). CNNTREE is a proposed general modular neural network structure, where each node in this tree is a comprehensible expert neural network (CENN). One advantage of using CNNTREE is that it is a “gray box”; because it can be interpreted easily for symbolic systems; where each node in the CNNTREE is equivalent for symbolic operator in the symbolic system. Another advantage of CNNTREE is that it can be trained as any normal multi layer feed forward neural network. An evolutionary algorithm is given for designing the CNNTREE. Back propagation is also checked as local learning algorithm that fits for real time learning constraints. The tree generalization and training performance are examined using experiments with a digit recognition problem. The article is published in the original. Elsayed Mostafa. Received the B.Sc. degree in electrical (Communication) Eng., Cairo University at 1967. Dipl.-Ing, and Dr-Ing. from Stuttgart University at 1977, 1981 respectively. He is a member of ECS and EEES. He is currently a professor of electronic circuits, Faculty of Engineering, University of Helwan. Amr Kamel. Graduated from Computer Department, Faculty of Engineering of Ain Shams University, Egypt in 1999, and studying M.Sc. degree in computer engineering from the Faculty of Engineering of Helwan University. His special fields of interest include neural networks and genetic algorithms. Alaa Hamdy. Was born in Giza in Egypt, on August 17, 1966. He graduated from the Telecommunications and Electronics Department, Faculty of Engineering and Technology of Helwan University, Cairo, Egypt in 1989. He received the M.Sc. degree in computer engineering from the same university in 1996 and the Ph.D. degree from the Faculty of Electrical Engineering, Poznan University of Technology, Poland in 2004. Currently he is working as a lecturer in the Faculty of Engineering of Helwan University. His special fields of interest, include image processing, pattern analysis, and machine vision.     

Convenient approach of nanohydroxyapatite polymeric matrix composites

Polyvinyl alcohol (PVA) and polyvinylpyrrolidone (PVP) were used to prepare polymeric matrix composites of nanohydroxyapatite (HAV and HAP), respectively, by using a biomimetic approach under 100 °C. A comparison with synthetic nanohydroxyapatite prepared by precipitation (SHA) was evaluated. Characterization of the prepared powders by differential thermal analysis (DTA) and thermogravimetric analysis (TGA) was achieved, X-ray diffraction (XRD), scanning electron microscope (SEM) with elemental analysis by (EDS) and inductively coupled plasma-optical emission spectrometer were used. DTA and TG showed the weight loss in the 400–1200 °C was almost attributed to carbonate decomposition which appeared apparently in the polymer associated samples due to the decomposition of polymer matrix. X-ray diffraction confirmed mainly the formation of hydroxyapatite. Specific surface area (BET) and scanning electron microscope and scanning electron microscopy (SEM) confirmed the range of nanosized of the prepared HA samples. The polymer matrix prefers orientation of the particles to rod-like shape. Elemental analysis showed mainly, Ca, P, C and O ions besides Na and Cl. Inductively coupled plasma showed the composition and the Ca/P atomic ratio of all samples ranging between 1.72 and 1.85.     

Coriander (<Emphasis Type="Italic">Coriandrum sativum</Emphasis> L.) seed oil improves plasma lipid profile in rats fed a diet containing cholesterol

Coriander (Coriandrum sativum L.) seed oil (COR) is a promising oil with high levels of bioactive compounds. Very little information, however, is available on the effect of administration of COR on different aspects of plasma lipid profile in experimental animals. In view of the important implications, the effect of administration of COR and oil blend [a mixture of soybean oil, coriander oil and sunflower oil (4:2:4, w/w/w; Blend)] on the profile of plasma lipids was investigated in 24 male albino rats placed on a cholesterol-rich (1%) basal diet as compared to rats on a cholesterol-free basal diet. Coriander seed oil and Blend were analyzed for composition of fatty acid, sterol and tocopherol. The levels of bioactive compounds (sterols and tocopherols) were higher in COR than in the Blend. In addition, the antiradical potential of COR and Blend was measured and the results showed that COR had stronger radical scavenging activity than Blend. In the biological experiment, rats were divided into four diet groups. The negative control group (control) consumed the basal diet (BD) only, which contained wheat starch, casein and cellulose, as well as mineral and vitamin mixtures. To the BD were added 1 g/100 g cholesterol (Chol/group), or both (Chol/COR group) and (Chol/Blend group). The groups did not differ before the experiment, which lasted 60 days. Plasma total lipids (TL), triacylglycerols (TAG), total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-C) and high-density lipoprotein-cholesterol (HDL-C) were measured at day 15, 30, 45 and 60 during the experiment period. Generally, COR and Blend-supplemented diets decreased the levels of TL, TC, TAG and LDL-C in plasma. In addition, significant increase in the levels of HDL-C was observed for Chol/COR and Chol/Blend groups. The results demonstrated that COR, and to a relatively lesser degree Blend, have hypocholesterolemic properties in rats fed a cholesterol-rich diet.     

Predicting the impact of subsurface heterogeneous hydraulic conductivity on the stochastic behavior of well draw down in a confined aquifer using artificial neural networks

Groundwater flow and behavior have to be investigated based on heterogeneous subsurface formation since the homogeneity assumption of this formation is not valid. Over the past twenty years, stochastic approach and Monte Carlo technique have been utilized very efficiently to understand the groundwater flow behavior. However, these techniques require lots of computational and numerical efforts according to the various researchers’ comments. Therefore, utilizing new techniques with much less computational efforts such as Artificial Neural Network (ANN) in the prediction of the stochastic behavior for the groundwater based on heterogeneous subsurface formation is highly appreciated. The current paper introduces the ANN technique to investigate and predict the stochastic behavior of a well draw down in a confined aquifer based on subsurface heterogeneous hydraulic conductivity. Several ANN models are developed in this research to predict the unsteady two dimensional well draw down and its stochastic characteristics in a confined aquifer. The results of this study showed that ANN method with less computational efforts was very efficiently capable of simulating and predicting the stochastic behavior of the well draw down resulted from the continuous constant pumping in the middle of a confined aquifer with subsurface heterogeneous hydraulic conductivity.