Loading...
Search for:
samavatian--m--h
0.138 seconds
Total 25642 records
Machine learning-assisted investigation of anisotropic elasticity in metallic alloys
, Article Materials Today Communications ; Volume 40 , 2024 ; 23524928 (ISSN) ; Alkhazaleh, H. A ; Samavatian, M ; Samavatian, V ; Sharif University of Technology
2024
Abstract
This study aims to develop a machine learning (ML)-based method for predicting the elasticity tensor of anisotropic metallic alloys with 21 independent constants. In this respect, the elastic moduli of a batch of additive-manufactured specimens with various orientations, determined by Tait–Bryan angles, are utilized as input data, while the anisotropic elastic tensor are targeted for prediction. Our primary findings indicate the efficacy of the radial basis function neural network (RBFNN) as a robust ML model for elasticity tensor prediction. Notably, the model demonstrates slightly higher efficiency in predicting elasticity tensor components aligned with principal axes compared to...
Local elasticity assessment of unidirectional fiber-reinforced polymer composites through impulse excitation and machine learning
, Article Journal of Reinforced Plastics and Composites ; 2024 ; 07316844 (ISSN) ; Alkhazaleh, H. A ; Khan, M. A ; Samavatian, M ; Samavatian, V ; Sharif University of Technology
ResearchGate
2024
Abstract
This study presents a novel methodology that integrates the Impulse Excitation Technique (IET) and machine learning (ML) to predict local elastic properties within isolated regions of unidirectional polymeric composite plates. The proposed model incorporates fiber volume and plate thickness as input parameters and leverages the first resonance frequencies of the local region at different fiber orientations, thus accounting for the composite’s anisotropy. Regression results from the deep neural network (DNN) model demonstrate robust prediction performance across all output targets in both testing and training datasets, with R2 coefficients surpassing 0.9. The model exhibits particularly...
Nonlinear modeling for bearing fault diagnosis in non-stationary operating conditions
, Article Journal of the Brazilian Society of Mechanical Sciences and Engineering ; Volume 46, Issue 5 , 2024 ; 16785878 (ISSN) ; Behzad, M ; Mehdigholi, H ; Sharif University of Technology
2024
Abstract
Bearing failure is one of the most important causes of shutdown in rotating machines. Most bearing diagnostic methods can only be used on machines with steady-state operational conditions. Changes in operating conditions cause changes in the statistical characteristics of the vibrating signals, which causes erroneous alarms related to bearing failure. The statistical index of vibration signals, independent of operating conditions of speed and load, is introduced in this paper to diagnosis bearing fault growth and reduce the rate of incorrect bearing failure alarms. The existing theoretical model is employed to simulate and extract the vibration database under variable operational conditions...
An efficient STT-Ram last level cache architecture for GPUs
, Article Proceedings - Design Automation Conference ; 2-5 June , 2014 , pp. 1-6 ; ISSN: 0738100X ; ISBN: 9781479930173 ; Abbasitabar, H ; Arjomand, M ; Sarbazi-Azad, H ; Sharif University of Technology
2014
Abstract
In this paper, having investigated the behavior of GPGPU applications, we present an effcient L2 cache architecture for GPUs based on STT-RAM technology. With the increase of processing cores count, larger on-chip memories are required. Due to its high density and low power characteristics, STT-RAM technology can be utilized in GPUs where numerous cores leave a limited area for on-chip memory banks. They have however two important issues, high energy and latency of write operations, that have to be addressed. Low data retention time STT-RAMs can reduce the energy and delay of write operations. However, employing STT-RAMs with low retention time in GPUs requires a thorough investigation on...
Correlation-driven machine learning for accelerated reliability assessment of solder joints in electronics
, Article Scientific Reports ; Volume 10, Issue 1 , 2020 ; Fotuhi Firuzabad, M ; Samavatian, M ; Dehghanian, P ; Blaabjerg, F ; Sharif University of Technology
Nature Research
2020
Abstract
The quantity and variety of parameters involved in the failure evolutions in solder joints under a thermo-mechanical process directs the reliability assessment of electronic devices to be frustratingly slow and expensive. To tackle this challenge, we develop a novel machine learning framework for reliability assessment of solder joints in electronic systems; we propose a correlation-driven neural network model that predicts the useful lifetime based on the materials properties, device configuration, and thermal cycling variations. The results indicate a high accuracy of the prediction model in the shortest possible time. A case study will evaluate the role of solder material and the joint...
Iterative machine learning-aided framework bridges between fatigue and creep damages in solder interconnections
, Article IEEE Transactions on Components, Packaging and Manufacturing Technology ; 2021 ; 21563950 (ISSN) ; Fotuhi Firuzabad, M ; Samavatian, M ; Dehghanian, P ; Blaabjerg, F ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2021
Abstract
Costly and time-consuming approaches for solder joint lifetime estimation in electronic systems along with the limited availability and incoherency of data challenge the reliability considerations to be among the primary design criteria of electronic devices. In this paper, an iterative machine learning framework is designed to predict the useful lifetime of the solder joint using a set of self-healing data that reinforces the machine learning predictive model with thermal loading specifications, material properties, and geometry of the solder joint. The self-healing dataset is iteratively injected through a correlation-driven neural network to fulfill the data diversity. Outcomes show a...
Iterative machine learning-aided framework bridges between fatigue and creep damages in solder interconnections
, Article IEEE Transactions on Components, Packaging and Manufacturing Technology ; Volume 12, Issue 2 , 2022 , Pages 349-358 ; 21563950 (ISSN) ; Fotuhi Firuzabad, M ; Samavatian, M ; Dehghanian, P ; Blaabjerg, F ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2022
Abstract
Costly and time-consuming approaches for solder joint lifetime estimation in electronic systems along with the limited availability and incoherency of data challenge the reliability considerations to be among the primary design criteria of electronic devices. In this article, an iterative machine learning framework is designed to predict the useful lifetime of the solder joint using a set of self-healing data that reinforce the machine learning predictive model with thermal loading specifications, material properties, and geometry of the solder joint. The self-healing dataset is iteratively injected through a correlation-driven neural network (CDNN) to fulfill the data diversity. Outcomes...
Estimation of compressive strength of cement mortars using impulse excitation technique and a genetic algorithm
, Article Advances in Cement Research ; Volume 36, Issue 5 , 2023 , Pages 230-239 ; 09517197 (ISSN) ; Sari, A ; Abdullaev, S. S ; Samavatian, M ; Samavatian, V ; Sharif University of Technology
ICE Publishing
2023
Abstract
Compressive strength, a crucial mechanical property of cement mortars, is typically measured destructively. However, there is a need to evaluate the strength of unique cement-based samples at various ages without causing damage. In this paper, a predictive framework using a genetic algorithm (GA) is proposed for estimating the compressive strength of ordinary cement-based mortars based on their dynamic elastic modulus, measured non-destructively using the impulse excitation technique. By combining the Popovics model (PM) and the Lydon-Balendran model (LBM), the static elastic modulus of samples was calculated using constant coefficients, representing an equivalent compressive strength. A GA...
Bayesian machine learning-aided approach bridges between dynamic elasticity and compressive strength in the cement-based mortars
, Article Materials Today Communications ; Volume 35 , 2023 ; 23524928 (ISSN) ; Samavatian, M ; Samavatian, V ; Sun, H ; Sharif University of Technology
Elsevier Ltd
2023
Abstract
This study tries to establish a powerful machine learning (ML) model for predicting the compressive strength of cement-based mortars by using dynamic elasticity data. The ML model was developed on the basis of Bayesian theorem, leading to a decrease in the overfitting problem compared with other conventional neural networks. Moreover, for the first time, the empirical equations were embedded in the ML model, enhancing the correlation between dynamic elasticity and compressive strength in the cement-based mortars. The results showed that the ML model efficiently predicted the compressive strength with determination coefficient (R2) of 95.2% and root mean square error (RMSE) of 0.0488 for...
Architecting the last-level cache for GPUs using STT-RAM technology
, Article Transactions on Design Automation of Electronic Systems ; Volume 20, Issue 4 , 2015 ; 10844309 (ISSN) ; Arjomand, M ; Bashizade, R ; Sarbazi Azad, H ; Sharif University of Technology
2015
Abstract
Future GPUs should have larger L2 caches based on the current trends in VLSI technology and GPU architectures toward increase of processing core count. Larger L2 caches inevitably have proportionally larger power consumption. In this article, having investigated the behavior of GPGPU applications, we present an efficient L2 cache architecture for GPUs based on STT-RAM technology. Due to its high-density and low-power characteristics, STT-RAM technology can be utilized in GPUs where numerous cores leave a limited area for on-chip memory banks. They have, however, two important issues, high energy and latency of write operations, that have to be addressed. Low retention time STT-RAMs can...
Discovery of novel quaternary bulk metallic glasses using a developed correlation-based neural network approach
, Article Computational Materials Science ; Volume 186 , 2021 ; 09270256 (ISSN) ; Gholamipour, R ; Samavatian, V ; Sharif University of Technology
Elsevier B.V
2021
Abstract
The immense space of composition-processing parameters leads to numerous trial-and-error experimental works for engineering of novel bulk metallic glasses (BMGs). To tackle this challenging problem, it is required to consider specific guidelines which are able to restrict the productive alloying compositions. In this work, a correlation-based neural network (CBNN) approach was developed, based on a dataset of 7950 alloying compositions, to design potential new MGs through prediction of casting ability, reduced glass transition (Trg) and critical thickness (Dmax). This approach involves individual and mutual characteristics of contributory factors to improve the prediction accuracy. To...
Characterization of nanoscale structural heterogeneity in metallic glasses: A machine learning study
, Article Journal of Non-Crystalline Solids ; Volume 578 , 2022 ; 00223093 (ISSN) ; Gholamipour, R ; Bokov, D.O ; Suksatan, W ; Samavatian, V ; Mahmoodan, M ; Sharif University of Technology
Elsevier B.V
2022
Abstract
Atomic force microscopy (AFM) is an efficient tool for studying the structural heterogeneity in metallic glasses (MGs). However, time-consuming analysis and limitations in the scanning process are downsides of this experiment. To tackle these problems, a machine learning (ML) model was developed to predict the distribution of energy dissipation on the MG surface with the increase in number of AFM scanning. The results indicated that it was possible to accurately predict the energy of scanning points, leading to a timesaving and reliable study. Moreover, characterization of structural heterogeneity shows that the viscoelastic response of each nanoscale region under sequences of AFM scans...
Machine learning-enabled characterization of concrete mechanical strength through correlation of flexural and torsional resonance frequencies
, Article Physica Scripta ; Volume 99, Issue 7 , 2024 ; 00318949 (ISSN) ; Samavatian, M ; Samavatian, V ; Sharif University of Technology
2024
Abstract
In this study, an assessment of concrete compressive strength was conducted using an impulse excitation data-driven machine learning (ML) framework. The model was constructed upon a deep neural network and aided by the backpropagation method, ensuring a precise training process. In contrast to prior research, which mainly focused on mixture components, a meaningful relationship between physical parameters—resonant frequencies and elastic moduli—and compressive strength was established by our ML model. Remarkable performance was demonstrated, with a root mean square error value of 2.8MPa and a determination factor of 0.97. Through Pearson analysis, correlations between input features and...
Development of a high-gain step-up dc/dc power converter with magnetic coupling for low-voltage renewable energy
, Article IEEE Access ; Volume 11 , 2023 , Pages 90038-90051 ; 21693536 (ISSN) ; Samavatian, V ; Samavatian, M ; Gono, T ; Jasinski, M ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2023
Abstract
There exists an extensive range of applications for elevated gain DC/DC converters, as numerous low-voltage resources are exploited for power supply. Therefore, this study introduces a groundbreaking magnetically coupled DC/DC converter specifically designed for resources with low voltage, including micro PV or fuel cell systems. By enduring low current and voltage stresses, the power devices in this converter ensure remarkable efficiency while maintaining proven voltage ratio capability. The operational principles of the converter are thoroughly discussed and supported by the implementation of a 200W-400V prototype. In order to confirm the effectiveness of the converter, a range of...
Reliability modeling of multistate degraded power electronic converters with simultaneous exposure to dependent competing failure processes
, Article IEEE Access ; Volume 9 , 2021 , Pages 67096-67108 ; 21693536 (ISSN) ; Fotuhi Firuzabad, M ; Dehghanian, P ; Blaabjerg, F ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2021
Abstract
Depending on the application in which power electronic converters (PECs) are deployed, failure processes may endanger the desirable performance of PECs. This paper offers holistic insights on reliability modeling of PECs considering dependencies in two simultaneous failure processes, namely gradual wearing-out degradation and vibration sudden degradation. While sudden and gradual degradation processes may individually affect the useful lifetime of PECs, their mutual interdependencies could significantly accelerate the aging mechanisms. A new analytical model for reliability assessment of PECs is proposed that can capture such mutual interdependence of simultaneous failure processes. The...
Torsional Vibration Online Monitoring Design and Manufacturing
, M.Sc. Thesis Sharif University of Technology ; Behzad, Mahdi (Supervisor)
Abstract
All rotating machines (such as internal combustion engines, reciprocating compressors, propulsion systems, etc.) experience torsional vibration during Start up, continuous working conditions and blackout. Torsional vibrationis due to pressure rhythms, engine dynamic torques, etc. All components of a rotating system must convey dynamic torque in addition to the static torque.Excessive Torsional vibrations cause wear and damage such as gears, gear teeth damage and failure to the main shaft.Online monitoring of torsional vibrations of the machine can be displayed machine condition at any moment and consider the risk in early diagnosis. Shaft torsional vibration monitoring can detect cracks in...
A Novel STT-RAM Architecture for Last Level Shared Caches in GPUs
, M.Sc. Thesis Sharif University of Technology ; Sarbazi-Azad, Hamid (Supervisor)
Abstract
Due to the high processing capacity of GPGPUs and their requirement to a large and high speed shared memory between thread processors clusters, exploiting Spin-Transfer Torque (STT) RAM as a replacement with SRAM can result in significant reduction in power consumption and linear enhancement of memory capacity in GPGPUs. In the GPGPU (as a many-core) with ability of parallel thread executing, advantages of STT-RAM technology, such as low read latency and high density, could be so effective. However, the usage of STT-RAM will be grantee applications run time reduction and growth threads throughput, when write operations manages and schedules to have least overhead on read operations. The...
A Statistical Model to Eliminate the Effects of Speed and Load on the Vibration Signal Generated by Defective Bearing Under Variable Operational Conditions
, Ph.D. Dissertation Sharif University of Technology ; Behzad, Mahdi (Supervisor) ; Mehdigholi, Hamid (Co-Supervisor) ; Rohani Bastami, Abbas (Co-Supervisor)
Abstract
Bearing failure is the most important causes of shutdown in rotating machines. Most existing fault diagnosis methods are only applicable under constant operating conditions, whereas modern machines often operate under variable loads and speeds. Changes in these operating conditions alter the statistical characteristics of vibration signals, which can lead to false alarms in bearing fault detection. The aim of this research is to propose a statistical indicator for vibration signals that is independent of variations in operating conditions and can more accurately track the progression of bearing faults while reducing false alarm rates.First, the physical mechanisms of bearing degradation were...
Optimal MIMO waveform design with controlled characteristics
, Article Proceedings International Radar Symposium, 24 June 2015 through 26 June 2015 ; Volume 2015-August , 2015 , Pages 1141-1146 ; 21555753 (ISSN) ; 9783954048533 (ISBN); 9783954048533 (ISBN); 9783954048533 (ISBN) ; Radmard, M ; Nayebi, M. M ; Bastani, M. H ; Rohling, H ; Rohling, H ; Rohling, H ; Sharif University of Technology
IEEE Computer Society
2015
Abstract
In a MIMO (Multiple Input Multiple Output) radar system, proper design of transmit signal and receive filter is an advantageous tool to improve the detection performance. Thus, in this paper, we consider the problem of transmit code and receive filter design, in order to maximize the Signal to Noise Ratio (SNR) at the receiver while enforcing a similarity constraint between the transmit space-time code (STC) and a reference STC which has a desirable ambiguity function. We will show that by solving such constrained optimization problem through successive iterations, our proposed method leads to satisfactory results
A fast and novel method of pattern synthesis for non-uniform phased array antennas
, Article Proceedings International Radar Symposium, 24 June 2015 through 26 June 2015 ; Volume 2015-August , 2015 , Pages 924-929 ; 21555753 (ISSN) ; 9783954048533 (ISBN); 9783954048533 (ISBN); 9783954048533 (ISBN) ; Sebt, M. A ; Nayebi, M. M ; Behroozi, H ; Rohling, H ; Rohling, H ; Rohling, H ; Sharif University of Technology
IEEE Computer Society
2015
Abstract
Weighting elements to achieve radiation patterns with desired characteristics is a classical work in phased array antennas. These characteristics can be low sidelobe level, narrow beamwidth, high directivity, pattern nulling in special angle and etc. For each of these characteristics, different methods have been introduced. Most of methods have been presented for uniform arrays, however there are lots of methods to obtain a desired pattern for antennas with non-uniform element distances. The problem with these methods is complexity or not very good results. In this paper, fast and easy methods based on Least Square Error that leads to good results are presented. In addition, weighting of...