Sharif Digital Repository

Rolling element bearing (REB) failure is one of the general damages in rotating machinery. In this manner, the correct prediction of remaining useful life (RUL) of REB is a crucial challenge to move forward the unwavering quality of the machines. One of the main difficulties in implementing data-driven methods for RUL prediction is to choose proper features that represent real damage progression. In this article, by using the outcomes of frequency analysis through the envelope method, the initiated/existed defects on the ball bearings are identified. Also, new features based on developing faults of ball bearings are recommended to estimate RUL. Early-stage faults in ball bearings usually... 

Oil and gas are likely the most important sources for producing heat and energy in both domestic and industrial applications. Hydrocarbon reservoirs that contain these fuels are required to be characterized to exploit the maximum amount of their fluids. Well testing analysis is a valuable tool for the characterization of hydrocarbon reservoirs. Handling and analysis of long-term and noise-contaminated well testing signals using the traditional methods is a challenging task. Therefore, in this study, a novel paradigm that combines wavelet transform (WT) and recurrent neural networks (RNN) is proposed for analyzing the long-term well testing signals. The WT not only reduces the dimension of... 

Accurately determining the nutritional status of plants can prevent many diseases caused by fertilizer disorders. Leaf analysis is one of the most used methods for this purpose. However, in order to get a more accurate result, disorders must be identified before symptoms appear. Therefore, this study aims to identify leaves with excessive nitrogen using one‐dimensional convolutional neural networks (1D‐CNN) on a dataset of spectral data using the Keras library. Seeds of cucumber were planted in several pots and, after growing the plants, they were divided into different classes of control (without excess nitrogen), N30% (excess application of nitrogen fertilizer by 30%), N60% (60% overdose),... 

In recent years, convolutional neural networks and deep neural networks have been used extensively in various fields of analytical chemistry. The use of these models for calibration tasks has been highly effective; however, few reports have been published on their properties and characteristics of analytical figures of merit. Currently, most performance measures for these types of networks only incorporate some function of prediction error. While useful, these measures are incomplete and cannot be used as an objective comparison among different models. In this report, a new method for calculating the sensitivity of any type of neural network is proposed and studied on both simulated and real... 

Shape optimization of a submarine for minimized noise emission has been the objective of this work. Boundary element method and finite element methods have been employed to determine the acoustic field around the object. A combined neural network and genetic algorithm scheme is developed to find the optimum external geometric ratios of the submarine for a minimized emitted acoustic energy in certain reference points. The obtained optimum geometric values stay in normal range for minimum hydrodynamic forces and show a close agreement with the trend of change for models coming to operation. © 2004 IEEE  

The objective of this paper is study of flowfield over the wing of a coplanar and close-coupled wing-canard configuration model using neural network, in order to reduce measurement points and experiment time. The results are based on flowfield pressure and wing surface pressure measurements at difference angles of attack, no sideslip and difference angles of canard. A GRNN (General Regression Neural Network) algorithm is developed to determine the shape and trajectory of vortices over this model. This network uses the total pressure coefficients of flowfield over the wing of this model that are obtained by total pressure probe as its input. The data presented to the network are processed to... 

The aim of this work is to combine different innovative methods to solve the forward kinematics (FK) problem in a parallel manipulator called Stewart platform. It leads to the solution of a set of simultaneous non-linear equations and results in a series of non-unique multiple sets of solutions. Many efforts have been made to solve this challenging problem and usually results in having to find the solution of a 16th order polynomial by means of numerical methods such as Hooks-Jeeves, Steepest descent search and Newton-Raphson method (NR). Accuracy, speed and convergence of these methods are fully dependent to the initial guess vector that is fed to the numerical algorithm. In this paper, a... 

In this paper a new control strategy for adaptive attitude control of multivariable satellite system has been presented. The approach is based on radial basis function neural network (RBFNN). By using four reaction wheels and Modified Rodrigues Parameters (MRPs) for attitude representation, the attitude dynamic of satellite has been considered. The Lyapunov stability theory has been used to achieve a stable closed loop system. Also to enhance the robustness of the controller, the RBF neural network has been employed to estimate the model base terms in control law. The control objective is the plant to track a reference model. Simulation results illustrate the performance of the on-line... 

We present a new method for the robust estimation of a two-dimensional conducting target orientation using measured radar cross-section data. The method is based on a Generalized Regression Neural Network (GRNN) scheme, which belongs to the family of radial basis neural networks  

In this paper, a new approach based on a Generalized Regression Neural Network (GRNN) has been proposed to predict the unsteady forces and moments on a 70° swept wing undergoing sinusoidal pitching motion. Extensive wind tunnel results were being used for training the network and for verification of the values predicted by this approach. The Generalized Regression Neural Network (GRNN) has been trained by the aforementioned experimental data and subsequently was used as a prediction tool to determine the unsteady longitudinal forces and moments of the pitching delta wing for various reduced frequencies. The results are in a good agreement with those determined by the previous experimental... 

The Hopfield network is studied from the standpoint of taking it as a computational model in optimizing the fuel management of pressurized water reactors (PWRs). In this paper, the flattening of the neutron flux is considered as the objective function. By this consideration, the power peaking inside the reactor core is also minimized. Regarding the local minimum problem of Hopfield network, the simulated annealing method is applied to improve the Hopfield solution. The method is applied to Bushehr Nuclear Power Plant (PWR design) and the result is compared with the core configuration purposed by the designer  

In this paper, two new approaches, using the combination of Hopfield Neural Network and Genetic Algorithm for solving the Robot Motion Planning Problem both in Crisp and Fuzzified environments are presented. Based on the hypothesis of Genetic Algorithms, the Genomes and Chromosomes of the algorithm are modified so that they can he used to solve the Motion Planning Problem. Because of some problem restrictions and limits hinder us to use the generic Genetic Algorithm; some modifications are applied to the main algorithm to able us to solve the problem. Although the proposed algorithms are both rely on Genetic Algorithm, the heart of both is based on Hopfield Neural Network Robot Path Planner... 

In this paper, a new approach based on Artificial Neural Networks to solve the robot motion planning problem is presented. For this purpose, a Hopfield Neural Network is used in a certain constraint satisfaction problem of the robot motion planning in conjunction with fuzzy modeling of the real robot's environment so that the energy of a state can be interpreted as the extent to which a hypothesis fit the underlying neural formulation model. Thus, low energy values indicate a good level of constraint satisfaction of the problem. Finally, since the obtained answer by the Hopfield Neural Network is not optimal, some algorithms are designed to optimize and generate the final answer  

A quantitative structure-property relationship study based on multiple linear regression (MLR) and artificial neural network (ANN) techniques was carried out to investigate the retention behavior of some terpenes on the polar stationary phase (Carbowax 20 M). A collection of 53 noncyclic and monocyclic terpenes was chosen as data set that was randomly divided into two groups, a training set and a prediction set consist of 41 and 12 molecules, respectively. A total of six descriptors appearing in the MLR model consist of one electronic, two geometric, two topological and one physicochemical descriptors. Except for the geometric parameters the remaining descriptors have a pronounced effect on... 

In this paper, a new formulation for discriminative training of HMMs is introduced as a solution to several speech recognition problems. This formulation uses a properly trained MLP in a simple interconnection with HMMs called "Cascade HMM/ANN Hybrid". The training algorithm has simple realization in comparison with other discriminative training for HMMs such as MDI and MMI. Also a rigid mathematical proof of its convergence has been presented. No significant increase in computational requirements is needed in recognition phase and the recognition task can still be performed in real-time. This structure has been employed in some isolated and continuous speaker-independent speech recognition... 

Coming up with a model which matches biological observations more closely has always been one of the main challenges in the field of artificial neural networks. Lately, an ionic model of reservoir networks containing spiking neurons (ILS-based reservoir network) has been proposed which seems to replicate some of the biological processes we have observed up until now. This paper presents a local learning rule for the ILS-based reservoir inspired by the biological fact that each incoming stimulus causes the formation of new dendritic spines, producing new synapses. This property may result in a higher degree of neuroplasticity, leading to a higher learning capacity. To evaluate the proposed... 

Precise modelling of the interaction of ions with materials is important for many applications including material characterization, ion implantation in devices, thermonuclear fusion, hadron therapy, secondary particle production (e.g. neutron), etc. In this study, a new approach using the Geant4 toolkit in combination with the Bayesian regularization (BR) learning algorithm of the feed-forward neural network (FFNN) is developed to estimate the range of ions in materials accurately and quickly. The different incident ions at different energies are interacted with the target materials. The Geant4 is utilized to model the interactions and to calculate the range of the ions. Afterward, the... 

The classification performance of all-optical Convolutional Neural Networks (CNNs) is greatly influenced by components’ misalignment and translation of input images in the practical applications. In this paper, we propose a free-space all-optical CNN (named Trans-ONN) which accurately classifies translated images in the horizontal, vertical, or diagonal directions. Trans-ONN takes advantages of an optical motion pooling layer which provides the translation invariance property by implementing different optical masks in the Fourier plane for classifying translated test images. Moreover, to enhance the translation invariance property, global average pooling (GAP) is utilized in the Trans-ONN... 

The present study tries to investigate the impact of different speed limits value in the severity of accidents. For this reason, head-on accidents on rural roads have been investigated. Speed limitations in different types of rural roads including inter-state rural arterials, major rural arterials, minor rural arterials, rural main collectors, rural minor collectors, rural local roads or streets in two cases, two-lane and four-lane roadways will be examined individually. The artificial neural network is trained based on the severity of the injuries of drivers of both vehicles involved in head-on accidents. The descriptive variables used in addition to the speed limit include age and gender...