Predicting the indirect tensile strength of self-compacting concrete using artificial neural networks

An experimental program was undertaken to evaluate the compressive strength of self-compacting concrete using commercial mathematic program. Sample variation was monitored using an experimental cylinder of concrete measuring 150 mm in diameter and 300 mm in height. This research examined various mixture designs in the laboratory tests with the goal of creating mixtures with desirable flow specification that did not require additional vibration yet provided adequate compressive strength. After 28 days, compressive strength of cylinder concrete determination, a model of Artificial Neural Networks (ANNs) was designed for this research and the results were obtained in this model of ANN. Both experimental tests and mix design program data was analyzed with statistical packet software. The result of statistical analysis has been done in 98.54 percent of confidence interval. It has been seen that the ANN can be used as reliable modelling method for similar experiment.

Download Full-text

Application of Artificial Neural Networks to Estimate Tensile Strength of Austenitic Stainless Steel During Metal Inert Gas Welding Process

Artificial Neural Network Applications in Business and Engineering - Advances in Computational Intelligence and Robotics ◽

10.4018/978-1-7998-3238-6.ch009 ◽

2021 ◽

pp. 197-221

Author(s):

Sudipto Chaki ◽

Dipankar Bose

Keyword(s):

Neural Networks ◽

Tensile Strength ◽

Stainless Steel ◽

Artificial Neural Networks ◽

Austenitic Stainless Steel ◽

Welding Current ◽

Inert Gas ◽

Mig Welding ◽

Artificial Neural ◽

Full Factorial Experimental Design

In the present work, artificial neural networks (ANN) have been used to model the complex relationship between input-output parameters of metal inert gas (MIG) welding processes. Four ANN training algorithms such as back propagation neural network (BPNN) with gradient descent momentum (GDM), BPNN with Levenberg Marquardt (LM) algorithm, BPNN with Bayesian regularization (BR), and radial basis function networks (RBFN) method have been used for prediction modelling. An experimentation based on full factorial experimental design has been conducted on MIG welding of austenitic stainless steel of grade-304 where welding current, welding speed, and voltage have been considered as input parameters, and tensile strength has been considered as measurable output parameter. The dataset so constituted is used for ANN modelling. Altogether, 40 different ANN architectures have been trained and tested using the above-mentioned algorithms, and 3-11-1 ANN architecture trained using BPNN with BR has been considered to show best prediction capability with mean % absolute error of 0.354%.

Download Full-text

Application of Bayesian Artificial Neural Networks for Modeling the Dependence of Nickel-Based Superalloys' Ultimate Tensile Strength on Their Chemical Composition

2019 International Conference on Control, Artificial Intelligence, Robotics & Optimization (ICCAIRO) ◽

10.1109/iccairo47923.2019.00018 ◽

2019 ◽

Author(s):

Dmitry Tarasov ◽

Oleg Milder ◽

Andrey Tyagunov

Keyword(s):

Neural Networks ◽

Tensile Strength ◽

Artificial Neural Networks ◽

Chemical Composition ◽

Ultimate Tensile Strength ◽

Artificial Neural

Download Full-text

Models to estimate Brazilian indirect tensile strength of limestone in saturated state

Rudarsko-geološko-naftni zbornik ◽

10.17794/rgn.2016.2.5 ◽

2016 ◽

Vol 31 (2) ◽

pp. 59-67

Author(s):

Zlatko Briševac ◽

Trpimir Kujundžić

Keyword(s):

Neural Networks ◽

Tensile Strength ◽

Mechanical Characteristics ◽

Indirect Tensile Strength ◽

Engineering Practice ◽

Wide Spread ◽

Efficacy Assessment ◽

Indirect Tensile ◽

Brazilian Indirect Tensile Strength ◽

Physical And Mechanical Characteristics

There are a number of methods for estimating physical and mechanical characteristics. Principally, the most widely used method is regression, but recently, more sophisticated methods such as neural networks have frequently been applied as well. This paper presents the models of a simple and a multiple regression and neural networks –types Radial Basis Function and Multiple Layer Perceptron, which can be used for the estimate of the Brazilian indirect tensile strength in saturated conditions. The paper includes the issues of collecting data for the analysis and modelling and an overview of the performed analysis with an efficacy assessment of the estimate for each model. After the assessment, the model which provided the best estimate was selected, including the model which could have the most wide-spread application in the engineering practice.

Download Full-text