Neural Networks for the Analysis and Design of Domes

Kaveh, A.; Dehkordi, Morteza Raissi

doi:10.1260/026635103322437463

Cited by 16 publications

(5 citation statements)

References 11 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the case studies herein, two AI models and one optimization algorithm are applied. The selected AI models are radial basis function neural network (RBFNN) and LSSVR, which have been used to solve many complex and nonlinear problems in engineering, because they have advantageous features that make them superior to other AI models 14–16 …”

Section: Introductionmentioning

confidence: 99%

Metaheuristics‐optimized ensemble system for predicting mechanical strength of reinforced concrete materials

Chou

Nguyen

2021

Struct Control Health Monit

View full text Add to dashboard Cite

Summary This paper develops a novel artificial intelligence (AI)‐based approach, called the metaheuristics‐optimized ensemble system (MOES), to assist civil engineers significantly in achieving accurate estimations of the mechanical strength of reinforced concrete (RC) materials. MOES integrates the advantages of hybrid and ensemble models by combining a metaheuristic optimization algorithm and efficient AI models. The metaheuristic algorithm finds the optimal hyperparameters of individual AI techniques and simultaneously adjusts their weights to yield the best optimized‐weight‐ensemble model. Particularly, the developed MOES was established by integrating the forensic‐based investigation optimization algorithm, the radial basis function neural network, and the least squares support vector regression. Four case studies of predicting structural mechanics of RC beams were performed to evaluate the performance of MOES and compare it to those of other single AI models, conventional ensemble models, hybrid models, and empirical methods. The analytical results of cross‐validation reveal that MOES was the most reliable approach, achieving the best values of all performance evaluation indexes. The automated predictive analytics revealed the robustness, efficiency, and stability of MOES. Thus, the proposed approach is a highly promising tool for predicting the structural mechanics of RC beams. The success of MOES in estimating the mechanical strength of RC beams has redefined the way of optimizing an ensemble AI model, which is the primary contribution of this research to the relevant body of knowledge.

show abstract

Section: Introductionmentioning

confidence: 99%

Metaheuristics‐optimized ensemble system for predicting mechanical strength of reinforced concrete materials

Chou

Nguyen

2021

Struct Control Health Monit

View full text Add to dashboard Cite

show abstract

“…Although sigmoidal transfer functions are the most popular, other types of functions can be employed. A significant variety of potential transfer functions have been presented in previous research [ 42 ]. The logistic sigmoid function was determined to be appropriate for the problem examined in this work, as shown in Figure 15 .…”

Section: Machine Learning Modelsmentioning

confidence: 99%

“…The input data proceed in a single path and are routed through artificial neural nodes to the output nodes in this approach. The number of layers depends on the complexity of the function [ 42 ]. There are 10 input data and 2 output data in this study.…”

Section: Machine Learning Modelsmentioning

confidence: 99%

Mechanical Properties, Crack Width, and Propagation of Waste Ceramic Concrete Subjected to Elevated Temperatures: A Comprehensive Study

et al. 2022

View full text Add to dashboard Cite

Waste ceramic concrete (WOC) made from waste ceramic floor tiles has several economic and environmental benefits. Fire is one of the most common disasters in buildings, and WOC is a brittle construction material; therefore, the mechanical properties of WOC structures under high temperatures should be considered. According to previous studies, hybrid fiber can further reduce damage to concrete under high temperatures. Meanwhile, crack width and propagation are among the key characteristics of concrete materials that need to be considered, but few studies have focused on their behavior when subjected to elevated temperatures. The new concrete materials proposed by the authors are WOC and WOC-Hybrid. WOC was prepared with Natural Coarse Aggregates (NCA), Natural Fine Aggregate (NFA), Ordinary Portland Cement (OPC 43 grade), and ceramic waste tiles with 20% replacements for coarse aggregates, 10% replacements for fine aggregates, and 10% replacement for cement. In contrast, WOC-Hybrid was prepared with the addition of hybrid fiber (1% crimped steel fiber and 1% polyvinyl alcohol fiber) in WOC. The specimens were exposed to temperatures of 100–300 °C, and then the specimens were tested for tensile and compressive strength. The present study aims to find a new method to improve concrete resistance to elevated temperatures at the lowest costs by experimental and computational analysis via machine learning models. The application of machine learning models such as artificial neural networks (ANN) and multiple linear regression (MLR) was employed in this study to predict the compressive and tensile strength of concrete. The linear coefficient correlation (R2) and mean square error (MSE) were evaluated to investigate the performance of the models. Based on the experimental analysis, the results show that the effect of hybrid fiber on the crack width and propagation is greater than that on the crack width and propagation of WOC and PC after exposure to high temperatures. However, the enhanced effect of hybrid fiber on the mechanical properties, rack width, and propagation decreases after subjecting it to a high-temperature treatment, owing to the melting and ignition of hybrid fibers at high temperatures. Regarding the computational analysis, it was found that the developed MLR model shows higher efficiency than ANN in predicting the compressive and tensile strength of PC, WOC, and WOC-Hybrid concrete.

show abstract

“…A comparative study using meta-heuristic algorithms for the optimization of single layer domes is presented in (Kaveh & Talatahari, 2010). The optimization of domes is discussed also in (Kaveh & Dehkordi, 2003) where neural networks are trained for the analysis, design and prediction of the displacements of domes using back propagation and radial basis functions networks. In several articles the Enhanced Colliding Bodies Optimization (ECBO) method is used for the topology optimization of different types of domes (Kaveh & Rezaei, 2015.…”

Section: Introductionmentioning

confidence: 99%

Gradient-based prestress and size optimization for the design of cable domes

Pollini

2021

International Journal of Solids and Structures

View full text Add to dashboard Cite

Neural Networks for the Analysis and Design of Domes

Cited by 16 publications

References 11 publications

Metaheuristics‐optimized ensemble system for predicting mechanical strength of reinforced concrete materials

Metaheuristics‐optimized ensemble system for predicting mechanical strength of reinforced concrete materials

Mechanical Properties, Crack Width, and Propagation of Waste Ceramic Concrete Subjected to Elevated Temperatures: A Comprehensive Study

Gradient-based prestress and size optimization for the design of cable domes

Contact Info

Product

Resources

About