Modeling uniaxial compressive strength of building stones using non-destructive test results as neural networks input parameters

Yurdakul, Murat; Akdaş, Hürriyet

doi:10.1016/j.conbuildmat.2013.05.109

Cited by 52 publications

(11 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These statistical indices allow the proposed model to be adjusted to approximate the reference model defined in Equation (4). When MS E and R 2 approach 0 and 1, respectively, accurate predictions of the track-soil functionũ r ff are obtained [47][48][49]. To reduce the prediction error, tests were performed by transforming the raw input and output target data [50]:…”

Section: Nn Architecturementioning

confidence: 99%

Scoping assessment of free-field vibrations due to railway traffic

Galvín

Mendoza

Connolly

et al. 2018

Soil Dynamics and Earthquake Engineering

View full text Add to dashboard Cite

The number of railway lines both operational and under construction is growing rapidly, leading to an increase in the number of buildings adversely affected by ground-borne vibration (e.g. shaking and indoor noise). Post-construction mitigation measures are expensive, thus driving the need for early stage prediction, during project planning/development phases. To achieve this, scoping models (i.e. desktop studies) are used to assess long stretches of track quickly, in absence of detailed design information. This paper presents a new, highly customisable scoping model, which can analyse the effect of detailed changes to train, track and soil on ground vibration levels. The methodology considers soil stiffness and the combination of both the dynamic and static forces generated due to train passage. It has low computational cost and can predict free-field vibration levels in accordance with the most common international standards. The model uses the direct stiffness method to compute the soil Green's function, and a novel two-and-a-half dimensional (2.5D) finite element strategy for train-track interaction. The soil Green's function is modulated using a neural network (NN) procedure to remove the need for the time consuming computation of track-soil coupling. This modulation factor combined with the new train-track approach results in a large reduction in computational time. The proposed model is validated by comparing track receptance, free-field mobility and soil vibration with both field experiments and a more comprehensive 2.5D combined finite element-boundary element (FEM-BEM) model. A sensitivity analysis is undertaken and it is shown that track type, soil properties and train speed have a dominant effect on ground vibration levels. Finally, the possibility of using average shear wave velocity introduced for seismic site response analysis to predict vibration levels is investigated and shown to be reasonable for certain smooth stratigraphy's.

show abstract

Section: Nn Architecturementioning

confidence: 99%

Scoping assessment of free-field vibrations due to railway traffic

Galvín

Mendoza

Connolly

et al. 2018

Soil Dynamics and Earthquake Engineering

View full text Add to dashboard Cite

show abstract

“…NN is a computational framework that is inspired by biological neural systems. Figure 1 shows the input layer, hidden layer, and the output layer in NN system [29]. It consists of a number of interconnected simple processing units called artificial neurons.…”

Section: Neural Networkmentioning

confidence: 99%

Application of a Neural Network Model for Prediction of Wear Properties of Ultrahigh Molecular Weight Polyethylene Composites

Kurt

Oduncuoğlu

2015

International Journal of Polymer Science

View full text Add to dashboard Cite

In the current study, the effect of applied load, sliding speed, and type and weight percentages of reinforcements on the wear properties of ultrahigh molecular weight polyethylene (UHMWPE) was theoretically studied. The extensive experimental results were taken from literature and modeled with artificial neural network (ANN). The feed forward (FF) back-propagation (BP) neural network (NN) was used to predict the dry sliding wear behavior of UHMWPE composites. Eleven input vectors were used in the construction of the proposed NN. The carbon nanotube (CNT), carbon fiber (CF), graphene oxide (GO), and wollastonite additives are the main input parameters and the volume loss is the output parameter for the developed NN. It was observed that the sliding speed and applied load have a stronger effect on the volume loss of UHMWPE composites in comparison to other input parameters. The proper condition for achieving the desired wear behaviors of UHMWPE by tailoring the weight percentage and reinforcement particle size and composition was presented. The proposed NN model and the derived explicit form of mathematical formulation show good agreement with test results and can be used to predict the volume loss of UHMWPE composites.

show abstract

“…ANN modelling utilizes building blocks or elements called ‘neurons’. The neurons are grouped into input, hidden and output layers with respective biases, weights and transfer functions (Abidoye and Das, 2015; Yurdakul and Akdas, 2013; Mueller and Hemond, 2013). The network manipulates the values of the biases and weights in a sequence of training processes and uses the transfer functions to establish the relationships between the inputs and the outputs.…”

Section: Introductionmentioning

confidence: 99%

Effects of environmental and turbine parameters on energy gains from wind farm system: Artificial neural network simulations

et al. 2019

View full text Add to dashboard Cite

Artificial neural network modelling has been employed to investigate the effects of various environmental and machine factors on the energy gain from wind farm systems. Numerical comparison of artificial neural network and nonlinear regression from XLSTAT showed that ANN possessed better numerical accuracy in predicting multivariate data. Several artificial neural network models are developed and tested with several structures to obtain the best prediction performance in energy gain from different wind farms in Jordan. The best performing artificial neural network model was used to predict the energy gain from wind farm based on changes in annual wind speed, turbine rotor diameter and turbine power. As a result of 20% increase in turbine power, 14.4%–31% energy gains were recorded across different wind farms. The proposed artificial neural network model was also a good predictor for energy cost resulting from specific wind farm design.

show abstract

Modeling uniaxial compressive strength of building stones using non-destructive test results as neural networks input parameters

Cited by 52 publications

References 26 publications

Scoping assessment of free-field vibrations due to railway traffic

Scoping assessment of free-field vibrations due to railway traffic

Application of a Neural Network Model for Prediction of Wear Properties of Ultrahigh Molecular Weight Polyethylene Composites

Effects of environmental and turbine parameters on energy gains from wind farm system: Artificial neural network simulations

Contact Info

Product

Resources

About