Optimisation of the design parameters of a domestic refrigerator using CFD and artificial neural networks

2016 ◽  
Vol 67 ◽  
pp. 227-238 ◽  
Author(s):  
Hasan Avcı ◽  
Dilek Kumlutaş ◽  
Özgün Özer ◽  
Mete Özşen
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Design Parameters ◽  
Domestic Refrigerator ◽  
Artificial Neural

A Multidisciplinary Design and Optimization Methodology in Electronics Packaging: Application to BGA Design

2003 ◽  
Author(s):  
Hamid Hadim ◽  
Tohru Suwa
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Objective Function ◽  
Thermal Strain ◽  
Design Parameters ◽  
Design And Optimization ◽  
Trade Offs ◽  
Artificial Neural ◽  
Optimization Methodology ◽  
The Individual

In this manuscript a systematic multidisciplinary electronic packaging design and optimization methodology based on the artificial neural networks technique is presented. This method is applied to a Ball Grid Array (BGA) package design as an example. Multidisciplinary criteria including thermal, structural (thermal strain), electromagnetic leakage, and cost are optimized simultaneously. A simplified routability criterion is also considered as a constraint. The artificial neural networks technique is used for thermal and structural performance predictions. Large calculation time reduction is achieved using the artificial neural networks, which also provide enough information to specify the individual weights for each design discipline within the objective function used for optimization. This methodology is able to provide the designers a clear view of the design trade-offs, which are represented in the objective function using various design parameters. This methodology can be applied to any electronic product design at any packaging level.

An Optimal Design Method for Artificial Neural Networks by Using the Design of Experiments

2007 ◽  
Vol 11 (6) ◽  
pp. 593-599 ◽  
Author(s):  
Eiichi Inohira ◽  
◽  
Hirokazu Yokoi
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Optimal Design ◽  
Design Method ◽  
Feedforward Neural Networks ◽  
Learning Rate ◽  
Design Parameters ◽  
Approximation Accuracy ◽  
Artificial Neural ◽  
Trained Neural Network

This paper presents a method to optimally design artificial neural networks with many design parameters using the Design of Experiment (DOE), whose features are efficient experiments using an orthogonal array and quantitative analysis by analysis of variance. Neural networks can approximate arbitrary nonlinear functions. The accuracy of a trained neural network at a certain number of learning cycles depends on both weights and biases and its structure and learning rate. Design methods such as trial-and-error, brute-force approaches, network construction, and pruning, cannot deal with many design parameters such as the number of elements in a layer and a learning rate. Our design method realizes efficient optimization using DOE, and obtains confidence of optimal design through statistical analysis even though trained neural networks very due to randomness in initial weights. We apply our design method three-layer and five-layer feedforward neural networks in a preliminary study and show that approximation accuracy of multilayer neural networks is increased by picking up many more parameters.

Multiple objective crashworthiness optimization of circular tubes with functionally graded thickness via artificial neural networks and genetic algorithms

2016 ◽  
Vol 231 (11) ◽  
pp. 2005-2016 ◽  
Author(s):  
Adil Baykasoğlu ◽  
Cengiz Baykasoğlu
Keyword(s):  
Neural Networks ◽  
Genetic Algorithms ◽  
Artificial Neural Networks ◽  
Functionally Graded ◽  
Design Parameters ◽  
Multiple Objective ◽  
Circular Tubes ◽  
Crashworthiness Optimization ◽  
Artificial Neural ◽  
Graded Thickness

The objective of this paper is to develop a multiple objective optimization procedure for crashworthiness optimization of circular tubes having functionally graded thickness. The proposed optimization approach is based on finite element analyses for construction of sample design space and verification; artificial neural networks for predicting objective functions values (peak crash force and specific energy absorption) for design parameters; and genetic algorithms for generating design parameters alternatives and determining optimal combination of them. The proposed approach seaminglesly integrates artificial neural networks and genetic algorithms. Artificial neural network acts as an objective function evaluator within the multiple objective genetic algorithms. We have shown that the proposed approach is able to generate Pareto optimal designs which are in a very good agreement with the finite element results.

scholarly journals Optimization of Running Blade Prosthetics Utilizing Crow Search Algorithm Assisted by Artificial Neural Networks

2021 ◽  
Vol 67 (3) ◽  
pp. 88-100
Author(s):  
Rosel Solís Manuel Javier ◽  
José Omar Dávalos Ramírez ◽  
Javier Molina Salazar ◽  
Juan Antonio Ruiz Ochoa ◽  
Antonio Gómez Roa
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Carbon Fibre ◽  
Failure Criterion ◽  
Search Algorithm ◽  
Acrylonitrile Butadiene Styrene ◽  
Sensitivity Study ◽  
Design Parameters ◽  
Artificial Neural ◽  
Important Design

A crow search algorithm (CSA) was applied to perform the optimization of a running blade prosthetics (RBP) made of composite materials like carbon fibre layers and cores of acrylonitrile butadiene styrene (ABS). Optimization aims to increase the RBP displacement limited by the Tsai-Wu failure criterion. Both displacement and the Tsai-Wu criterion are predicted using artificial neural networks (ANN) trained with a database constructed from finite element method (FEM) simulations. Three different cases are optimized varying the carbon fibre layers orientations: –45°/45°, 0°/90°, and a case with the two-fibre layer orientations intercalated. Five geometric parameters and a number of carbon fibre layers are selected as design parameters. A sensitivity analysis is performed using the Garzon equation. The best balance between displacement and failure criterion was found with fibre layers oriented at 0°/90°. The optimal candidate with –45°/45° orientation presents higher displacement; however, the Tsai-Wu criterion was less than 0.5 and not suitable for RBP design. The case with intercalated fibres presented a minimal displacement being the stiffer RBP design. The damage concentrates mostly in the zone that contacts the ground. The sensitivity study found that the number of layers and width were the most important design parameters.

Neural network modeling of the implosion process in the tasks of optimization of oil production process

2021 ◽  
Author(s):  
A.R. Mukhutdinov ◽  
Z.R. Vakhidova ◽  
M.G. Efimov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Network Model ◽  
Neural Network Model ◽  
Network Modeling ◽  
Water Hammer ◽  
Design Parameters ◽  
Neural Network Modeling ◽  
Artificial Neural

An increase in the productivity of oil wells is possible with the use of a promising technology based on implosion and a device for its implementation. It is known that the effectiveness of the technology depends on the design parameters of the device. Currently, a promising way to study processes is computer modeling based on modern information technologies. Therefore, solving forecasting problems using modern software based on artificial neural networks (ANNs) is an urgent task of scientific and practical interest. In this regard, the aim of the work is to develop a neural network model and its application to identify the features of the influence of the diameter and length of the implosion chamber of the device on the pressure of a water hammer during implosion. In the software environment, the following have been created and tested: a method for developing a neural network model; a method of conducting a computational experiment with it. The possibility of neural network modeling of the implosion process has been studied. The results of predicting the output parameter, in this case the pressure of the water hammer, on a pre-trained network, with a relative error of 3.5%, using the knowledge base are demonstrated. The results of applying the methodology for solving forecasting problems using software based on artificial neural networks are presented. It was found that the diameter and length of the implosion chamber significantly affect the pressure of the water hammer. The practical significance of the work lies in the ability to determine the required values of the diameter and length of the implosion chamber of the device at a given level of water hammer pressure.

A Multidisciplinary Design and Optimization Methodology for Ball Grid Array Packages Using Artificial Neural Networks

2004 ◽  
Vol 127 (3) ◽  
pp. 306-313 ◽  
Author(s):  
Hamid Hadim ◽  
Tohru Suwa
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Ball Grid Array ◽  
System Level ◽  
Computational Time ◽  
Design Parameters ◽  
Design And Optimization ◽  
Trade Offs ◽  
Artificial Neural ◽  
Optimization Methodology

A systematic multidisciplinary electronics packaging design and optimization methodology, which takes into account the complexity of multiple design trade-offs, operated in conjunction with the artificial neural networks (ANNs) technique is presented. To demonstrate its capability, this method is applied to a plastic ball grid array package design. Multidisciplinary criteria including thermal, structural, electromagnetic leakage, and cost are optimized simultaneously using key design parameters as variables. A simplified routability criterion is also considered as a constraint. ANNs are used for thermal and structural performance predictions which resulted in large reduction in computational time. The present methodology is able to provide the designers a tool for systematic evaluation of the design trade-offs which are represented in the objective function. This methodology can be applied to any electronic product design at any packaging level from the system level to the chip level.

scholarly journals A flexible system for initial ship design parameters estimation using a system of neural networks

2011 ◽  
Vol 8 (2) ◽  
pp. 71-82 ◽  
Author(s):  
Hamada Senousy ◽  
Mahmoud Abou-Elmakarem
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Ship Design ◽  
Parameters Estimation ◽  
Design Parameters ◽  
Reliable Estimate ◽  
Multilayer Perceptrons ◽  
New Approach ◽  
Flexible System ◽  
Artificial Neural

To initialize ship design process, it is very important to be able to develop an initial estimate of ship parameters to satisfy designer required specifications. For new emerging designs, this estimate has to be made based on a limited available set of examples. Moreover, a practical estimate prediction strategy should be flexible enough having no distinction between input (specified constraints) and outputs (parameters required to be estimated), since these vary from one design case to another.  Conventional regression-based techniques, which are usually employed to provide the required estimates, suffer from low accuracy in case of a small number of available examples. In addition to that, they fail to capture the interrelation between different design parameters. To overcome these limitations and others, the present paper proposes a new approach based on a system of artificial neural-networks (ANNs). The new approach not only overcomes regression limitations but is also capable of providing a reliable estimate of initial design offset table based on different ANN outputs.  The paper uses a case study for demonstrating the merits of the proposed approach.Keywords: Ship design; regression; ship series; Artificial Neural Networks (ANNs); Multilayer Perceptrons (MLPs); Normalized Gaussian Modified Lagrangian (NGML) doi: http://dx.doi.org/10.3329/jname.v8i2.6945 Journal of Naval Architecture and Marine Engineering 8(2011) 71-82

Sign in / Sign up

Export Citation Format

Share Document