Improving the neural network method for finite element model updating using homogenous distribution of design points

Sadr, M. H.; Astaraki, Sohrab; Salehi, Sina

doi:10.1007/s00419-007-0129-x

Cited by 11 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, most of the soil sample data, used to train BPNN and PSO-BPNN, had low contents of the soil heavy metals. This led to larger errors for the test samples that had high heavy metal contents and affected performance and prediction accuracy [32]. Thus, in order to improve the prediction accuracy of the soil heavy metal contents, more soil samples with high values of the contents have to be collected to improve the performance of PSO-BPNN in the future.…”

Section: Discussionmentioning

confidence: 99%

Integrating a Hybrid Back Propagation Neural Network and Particle Swarm Optimization for Estimating Soil Heavy Metal Contents Using Hyperspectral Data

Liu

et al. 2019

Sustainability

View full text Add to dashboard Cite

Soil heavy metals affect human life and the environment, and thus, it is very necessary to monitor their contents. Substantial research has been conducted to estimate and map soil heavy metals in large areas using hyperspectral data and machine learning methods (such as neural network), however, lower estimation accuracy is often obtained. In order to improve the estimation accuracy, in this study, a back propagation neural network (BPNN) was combined with the particle swarm optimization (PSO), which led to an integrated PSO-BPNN method used to estimate the contents of soil heavy metals: Cd, Hg, and As. This study was conducted in Guangdong, China, based on the soil heavy metal contents and hyperspectral data collected from 90 soil samples. The prediction accuracies from BPNN and PSO-BPNN were compared using field observations. The results showed that, 1) the sample averages of Cd, Hg, and As were 0.174 mg/kg, 0.132 mg/kg, and 9.761 mg/kg, respectively, with the corresponding maximum values of 0.570 mg/kg, 0.310 mg/kg, and 68.600 mg/kg being higher than the environment baseline values; 2) the transformed and combined spectral variables had higher correlations with the contents of the soil heavy metals than the original spectral data; 3) PSO-BPNN significantly improved the estimation accuracy of the soil heavy metal contents, with the decrease in the mean relative error (MRE) and relative root mean square error (RRMSE) by 68% to 71%, and 64% to 67%, respectively. This indicated that the PSO-BPNN provided great potential to estimate the soil heavy metal contents; and 4) with the PSO-BPNN, the Cd content could also be mapped using HuanJing-1A Hyperspectral Imager (HSI) data with a RRMSE value of 36%, implying that the PSO-BPNN method could be utilized to map the heavy metal content in soil, using both field spectral data and hyperspectral imagery for the large area.

show abstract

Section: Discussionmentioning

confidence: 99%

Integrating a Hybrid Back Propagation Neural Network and Particle Swarm Optimization for Estimating Soil Heavy Metal Contents Using Hyperspectral Data

Liu

et al. 2019

Sustainability

View full text Add to dashboard Cite

show abstract

“…Their suggested updating approach is evaluated on a basic simulated model in the absence and presence of noise and after updating gets an accurate result. Sadr et al [19] conducted the experimental analysis on the structure and compare it with an analytical model by using the updating method. The authors introduced the neural network method to update the structure.…”

Section: Introductionmentioning

confidence: 99%

Finite element model updating of smart structures with direct updating algorithm

Verma¹,

Kango²,

Bagha³

et al. 2022

Phys. Scr.

View full text Add to dashboard Cite

In this paper, a finite element model updating algorithm is proposed to enhance the accuracy of the simulated finite element model of a smart structure (collocated piezoelectric patches embedded on a cantilever beam). Piezoelectric patches are used to sense and control the excessive vibrations of the structures. Mostly, they are mounted on flexible structures to measure their response at different excitations. The finite element method can be used to model the beam embedded with collocated piezoelectric patches. The complete finite element formulation of the smart structure is briefly described in this paper. There are different types of uncertainties that may be present in the simulated finite element model of a smart structure such as uncertainty in the structural boundary conditions, in the material elastic properties, the dimensions of the structure, piezoelectric elastic and electric properties, and the location of the piezoelectric patches mounted on the structure. In the present analytical study, the above uncertainties present in the smart structure are reduced by using the direct updating algorithm. It is found that the direct updating method through updating the mass and the stiffness matrices of the smart structure successfully enhance the accuracy of the simulated finite element model of the beam embedded with PZT patches. The state-space method is used to predict the response in the frequency domain. The maximum percentage error in the simulated finite element model of the piezoelectric embedded beam structure due to its structural and the electrical property uncertainty is 10.36% and 23.52% respectively and that was completely removed by using the direct updating algorithm. The optimal location of the piezoelectric patches is also taken as uncertainty which is successfully updated by using the proposed direct updating algorithm. The maximum percentage error in the natural frequencies of the smart structure due to location uncertainty is 18.39% which was also completely removed. To validate the outcomes, a frequency response function (FRF) is plotted.

show abstract

“…The non-model-based methods mainly include methods based on the change in natural frequency (Lee and Chung, 2000;Kim et al, 2003), the change of structural vibration mode (Khoo et al, 2004), the change of structural flexibility or stiffness (Yan and Golinval, 2005), the transfer function (Park and Park, 2005), the statistic information (Iwasaki et al, 2004;López-Díez et al, 2005;Garcia-Perez et al, 2013), the power flow (Li et al, 2004), and so on. The model-based methods detect damage based on predetermined structural models, in which the model updating methods are widely used (Sinha and Friswell, 2003;Sadr et al, 2007;Zhao et al, 2009;Lei et al, 2012;Liu and Duan, 2012;Erdogan and Bakir, 2013;Liu et al, 2013). However, most of the above methods are not quick, which means they provide structural information with an obvious time delay caused by their massive and off-line calculation process.…”

Section: Introductionmentioning

confidence: 99%

An in-time damage identification approach based on the Kalman filter and energy equilibrium theory

Huang

Dyke

2015

J. Zhejiang Univ. Sci. A

View full text Add to dashboard Cite

Abstract:In research on damage identification, conventional methods usually face difficulties in converging globally and rapidly. Therefore, a fast in-time damage identification approach based on the Kalman filter and energy equilibrium theory is proposed to obtain the structural stiffness, find the locations of damage, and quantify its intensity. The proposed approach establishes a relationship between the structural stiffness and acceleration response by means of energy equilibrium theory. After importing the structural energy into the Kalman filter algorithm, unknown parameters of the structure can be obtained by comparing the predicted energy and the measured energy in each time step. Numerical verification on a highway sign support truss with and without damage indicates that the updated Young's modulus can converge to the true value rapidly, even under the effects of external noise excitation. In addition, the calculation time taken for each step of the approach is considerably shorter than the sampling period (1/256 s), which means that, this approach can be implemented in-time and on-line.

show abstract

Improving the neural network method for finite element model updating using homogenous distribution of design points

Cited by 11 publications

References 16 publications

Integrating a Hybrid Back Propagation Neural Network and Particle Swarm Optimization for Estimating Soil Heavy Metal Contents Using Hyperspectral Data

Integrating a Hybrid Back Propagation Neural Network and Particle Swarm Optimization for Estimating Soil Heavy Metal Contents Using Hyperspectral Data

Finite element model updating of smart structures with direct updating algorithm

An in-time damage identification approach based on the Kalman filter and energy equilibrium theory

Contact Info

Product

Resources

About