Statistical Vibration Based Damage Identification Using Artificial Neural Network

2012 ◽  
Author(s):  
Norhisham Bakhary
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Monte Carlo ◽  
Finite Element Model ◽  
Element Model ◽  
Point Estimation ◽  
Structural Vibration ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Artificial Neural Network (ANN) telah digunakan dengan meluas bagi tujuan mengesan kerosakan dalam struktur menggunakan data–data mod dari gegaran. Walau bagaimanapun, ketidakpastian yang wujud dalam model unsur terhingga dan data dari lapangan yang tidak dapat dielakkan boleh menyebabkan kesilapan dalam meramalkan magnitud dan lokasi kerosakan. Dalam kajian ini kaedah statistik digunakan untuk mengambil kira ketidakpastian ini. ANN digunakan untuk meramalkan parameter–parameter kekukuhan dari frekuensi dan mod bentuk bagi sesebuah struktur. Untuk mengambil kira ketidakpastian dalam ramalan, kaedah statistik digunakan di mana kaedah Rossenblueth point estimation diperbandingkan dengan kaedah Monte Carlo diaplikasikan bagi mengambil kira ketidakpastian ini. Keputusan menunjukkan bahawa dengan mengambil kira ketidakpastian dalam membuat ramalan menggunakan ANN, kerosakan boleh diramalkan pada tahap keyakinan yang tinggi. Kata kunci: Artificial neural network; ketidakpastian; kesilapan rawak Artificial Neural Network (ANN) has been widely applied to detect damages in structures based on structural vibration modal parameters. However, uncertainties that inevitably exist in finite element model and measured vibration data might lead to false or unreliable prediction of structural damage. In this study, a statistical approach is proposed to include the effect of uncertainties in the ANN algorithm for damage prediction. ANN is used to predict the stiffness parameters of structures from measured structural vibration frequencies and mode shapes. Uncertainties in the measured data and finite element model of the structure are considered in the prediction. The statistics of the identified parameters are determined using Rossenblueth’s point estimation method and verified by Monte Carlo simulation. The results show that by considering these uncertainties in the ANN model, the damages can be detected with a higher confidence level. Key words: Artificial neural network; uncertainties; random error

Damage Assessment in Structures Using Incomplete Modal Data and Artificial Neural Network

2015 ◽  
Vol 15 (06) ◽  
pp. 1450087 ◽  
Author(s):  
Seyed Sina Kourehli
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Finite Element Model ◽  
Structural Damage ◽  
Element Model ◽  
Mass System ◽  
Modal Data ◽  
Artificial Neural ◽  
Propagation Network

This paper presents a novel approach for structural damage detection and estimation using incomplete noisy modal data and artificial neural network (ANN). A feed-forward back propagation network is proposed for estimating the structural damage location and severity. Incomplete modal data is used in the dynamic analysis of damaged structures by the condensed finite element model and as input parameters to the neural network for damage identification. In all cases, the first two natural modes were used for the training process. The present method is applied to three examples consisting of a simply supported beam, three-story plane frame, and spring-mass system. Also, the effect of the discrepancy in mass and stiffness between the finite element model and the actual tested dynamic system has been investigated. The results demonstrated the accuracy and efficiency of the proposed method using incomplete modal data, which may be noisy or noise-free.

Vibration–Based Damage Detection Of Slab Structure Using Artificial Neural Network

2012 ◽  
Author(s):  
Norhisham Bakhary
Keyword(s):  
Neural Network ◽  
Finite Element Analysis ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Damage Detection ◽  
Ann Model ◽  
Element Analysis ◽  
Severity Prediction ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Kertas kerja ini memaparkan kajian berkenaan keberkesanan Artificial Neural Network (ANN) dalam mengenal pasti kerosakan di dalam struktur. Data dari getaran seperti frekuensi semula jadi dan mod bentuk digunakan sebagai data masukan bagi ANN untuk meramalkan lokasi dan tahap kerosakan bagi struktur lantai. Analisis unsur terhingga (Finite Element Analysis) telah digunakan untuk memperoleh ciri–ciri dinamik bagi struktur–struktur rosak dan tidak rosak untuk ‘melatih’ model ‘neural network’. Senario kerosakan yang berbeza disimulasikan dengan mengurangkan kekukuhan elemen pada lokasi yang berbeza sepanjang struktur tersebut. Berbagai kombinasi data masukan bagi mod yang berbeza telah digunakan untuk memperolehi model ANN yang terbaik. Hasil kajian ini menunjukkan ANN mampu memberikan keputusan yang baik dalam meramal kerosakan pada struktur lantai tersebut. Kata kunci: Ramalan kerosakan struktur, Artificial Neural Network This paper investigates the effectiveness of artificial neural network (ANN) in identifying damages in structures. Global (natural frequencies) and local (mode shapes) vibration–based data has been used as the input to ANN for location and severity prediction of damages in a slab–like structure. A finite element analysis has been used to obtain the dynamic characteristics of intact and damaged structure to train the neural network model. Different damage scenarios have been introduced by reducing the local stiffness of the selected elements at different locations along the structure. Several combinations of input variables in different modes have been used in order to obtain a reliable ANN model. The trained ANN model is validated using laboratory test data. The results show that ANN is capable of providing acceptable result on damage prediction of tested slab structure. Key words: Structural damage detection, artificial neural network

scholarly journals Artificial Neural Network (ANN) and Finite Element (FEM) Models for GFRP-Reinforced Concrete Columns under Axial Compression

Materials ◽  
2021 ◽  
Vol 14 (23) ◽  
pp. 7172
Author(s):  
Haytham F. Isleem ◽  
Bassam A. Tayeh ◽  
Wesam Salah Alaloul ◽  
Muhammad Ali Musarat ◽  
Ali Raza
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Reinforced Concrete ◽  
Concrete Structures ◽  
Fiber Reinforced Polymer ◽  
Concrete Core ◽  
Reinforced Polymer ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

In reinforced concrete structures, the fiber-reinforced polymer (FRP) as reinforcing rebars have been widely used. The use of GFRP (glass fiber-reinforced polymer) bars to solve the steel reinforcement corrosion problem in various concrete structures is now well documented in many research studies. Hollow concrete-core columns (HCCs) are used to make a lightweight structure and reduce its cost. However, the use of FRP bars in HCCs has not yet gained an adequate level of confidence due to the lack of laboratory tests and standard design guidelines. Therefore, the present paper numerically and empirically explores the axial compressive behavior of GFRP-reinforced hollow concrete-core columns (HCCs). A total of 60 HCCs were simulated in the current version of Finite Element Analysis (FEA) ABAQUS. The reference finite element model (FEM) was built for a wide range of test variables of HCCs based on 17 specimens experimentally tested by the same group of researchers. All columns of 250 mm outer diameter, 0, 40, 45, 65, 90, 120 mm circular inner-hole diameter, and a height of 1000 mm were built and simulated. The effects of other parameters cover unconfined concrete strength from 21.2 to 44 MPa, the internal confinement (center to center spiral spacing = 50, 100, and 150 mm), and the amount of longitudinal GFRP bars (ρv = 1.78–4.02%). The complex column response was defined by the concrete damaged plastic model (CDPM) and the behavior of the GFRP reinforcement was modeled as a linear-elastic behavior up to failure. The proposed FEM showed an excellent agreement with the tested load-strain responses. Based on the database obtained from the ABAQUS and the laboratory test, different empirical formulas and artificial neural network (ANN) models were further proposed for predicting the softening and hardening behavior of GFRP-RC HCCs.

scholarly journals Generating Component Designs for an Improved NVH Performance by Using an Artificial Neural Network as an Optimization Metamodel

Designs ◽  
2021 ◽  
Vol 5 (2) ◽  
pp. 36
Author(s):  
Timo von Wysocki ◽  
Frank Rieger ◽  
Dimitrios Ernst Tsokaktsidis ◽  
Frank Gauterin
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Frequency Response Function ◽  
Element Model ◽  
Training Data ◽  
Component Design ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Noise Vibration ◽  
Geometry Parameter

In modern vehicle development, suspension components have to meet many boundary conditions. In noise, vibration, and harshness (NVH) development these are for example eigenfrequencies and frequency response function (FRF) amplitudes. Component geometry parameters, for example kinematic hard points, often affect multiple of these targets in a non intuitive way. In this article, we present a practical approach to find optimized parameters for a component design, which fulfills an FRF target curve. By morphing an initial component finite element model we create training data for an artificial neural network (ANN) which predicts FRFs from geometry parameter input. Then the ANN serves as a metamodel for an evolutionary algorithm optimizer which identifies fitting geometry parameter sets, meeting an FRF target curve. The methodology enables a component design which considers an FRF as a component target. In multiple simulation examples we demonstrate the capability of identifying component designs modifying specific eigenfrequency or amplitude features of the FRFs.

scholarly journals An artificial neural network based method to uncover the value-of-travel-time distribution

2020 ◽  
Author(s):  
Sander van Cranenburgh ◽  
Marco Kouwenhoven
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Monte Carlo ◽  
Travel Time ◽  
Discrete Choice Models ◽  
Choice Data ◽  
Monte Carlo Data ◽  
Travel Time Distribution ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Abstract This study proposes a novel Artificial Neural Network (ANN) based method to derive the Value-of-Travel-Time (VTT) distribution. The strength of this method is that it is possible to uncover the VTT distribution (and its moments) without making assumptions about the shape of the distribution or the error terms, while being able to incorporate covariates and taking the panel nature of stated choice data into account. To assess how well the proposed ANN-based method works in terms of being able to recover the VTT distribution, we first conduct a series of Monte Carlo experiments. After having demonstrated that the method works on Monte Carlo data, we apply the method to data from the 2009 Norwegian VTT study. Finally, we extensively cross-validate our method by comparing it with a series of state-of-the-art discrete choice models and nonparametric methods. Based on the promising results we have obtained, we believe that there is a place for ANN-based methods in future VTT studies.

scholarly journals A Comparison between Finite Element Model (FEM) Simulation and an Integrated Artificial Neural Network (ANN)-Particle Swarm Optimization (PSO) Approach to Forecast Performances of Micro Electro Discharge Machining (Micro-EDM) Drilling

Micromachines ◽  
2021 ◽  
Vol 12 (6) ◽  
pp. 667
Author(s):  
Mariangela Quarto ◽  
Gianluca D’Urso ◽  
Claudio Giardini ◽  
Giancarlo Maccarini ◽  
Mattia Carminati
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Particle Swarm Optimization ◽  
Process Parameters ◽  
Element Model ◽  
Micro Edm ◽  
Swarm Optimization ◽  
Artificial Neural Network Ann ◽  
Edm Drilling

Artificial Neural Network (ANN), together with a Particle Swarm Optimization (PSO) and Finite Element Model (FEM), was used to forecast the process performances for the Micro Electrical Discharge Machining (micro-EDM) drilling process. The integrated ANN-PSO methodology has a double direction functionality, responding to different industrial needs. It allows to optimize the process parameters as a function of the required performances and, at the same time, it allows to forecast the process performances fixing the process parameters. The functionality is strictly related to the input and/or output fixed in the model. The FEM model was based on the capacity of modeling the removal process through the mesh element deletion, simulating electrical discharges through a proper heat-flux. This paper compares these prevision models, relating the expected results with the experimental data. In general, the results show that the integrated ANN-PSO methodology is more accurate in the performance previsions. Furthermore, the ANN-PSO model is faster and easier to apply, but it requires a large amount of historical data for the ANN training. On the contrary, the FEM is more complex to set up, since many physical and thermal characteristics of the materials are necessary, and a great deal of time is required for a single simulation.

Sign in / Sign up

Export Citation Format

Share Document