Prediction of Moment Redistribution in Statically Indeterminate Reinforced Concrete Structures Using Artificial Neural Network and Support Vector Regression

Li, Ling; Zheng, Wenzhong; Wang, Ying

doi:10.3390/app9010028

Cited by 18 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ANN has been the most popular (thirty-five studies) in the construction/building materials field, while ANFIS and SVM were used by five, and FL and MGGP by four studies. In terms of the sub-disciplines where the AI tools were used, the list comprises a wide range of topics, such as predicting compressive strength of concrete (Pham et al , 2019; Dutta et al , 2018; Tanyildizi, 2018; Liu and Zheng, 2019; Ghanizadeh and Rahrovan, 2019; Al-Gburi et al , 2018; Sadowski et al , 2018; Gazder et al , 2017; Reddy, 2018; Behnood and Golafshani, 2018; Nguyen et al , 2019), measuring the adhesion/cohesion force between asphalt molecules at nanoscale level (Hassan et al , 2018), estimating the shear strength from the relative density, particle size, distribution (gradation), material hardness, gradation and fineness modulus (Yang et al , 2019; Li et al , 2019), confining (normal) stress, forecasting atmospheric corrosion of metallic materials (Zhu et al , 2019) and modelling the surface roughness in the turning of hardened AISI H11 steel (Saini et al , 2012).…”

Section: Discussionmentioning

confidence: 99%

Systematic review of application of artificial intelligence tools in architectural, engineering and construction

Momade

Durdyev

Estrella

et al. 2021

FEBE

View full text Add to dashboard Cite

PurposeThis study reviews the extent of application of artificial intelligence (AI) tools in the construction industry.Design/methodology/approachA thorough literature review (based on 165 articles) was conducted using Elsevier's Scopus due to its simplicity and as it encapsulates an extensive variety of databases to identify the literature related to the scope of the present study.FindingsThe following items were extracted: type of AI tools used, the major purpose of application, the geographical location where the study was conducted and the distribution of studies in terms of the journals they are published by. Based on the review results, the disciplines the AI tools have been used for were classified into eight major areas, such as geotechnical engineering, project management, energy, hydrology, environment and transportation, while construction materials and structural engineering. ANN has been a widely used tool, while the researchers have also used other AI tools, which shows efforts of exploring other tools for better modelling abilities. There is also clear evidence of that studies are now growing from applying a single AI tool to applying hybrid ones to create a comparison and showcase which tool provides a better result in an apple-to-apple scenario.Practical implicationsThe findings can be used, not only by the researchers interested in the application of AI tools in construction, but also by the industry practitioners, who are keen to further understand and explore the applications of AI tools in the field.Originality/valueThere are no studies to date which serves as the center point to learn about the different AI tools available and their level of application in different fields of AEC. The study sheds light on various studies, which have used AI in hybrid/evolutionary systems to develop effective and accurate predictive models, to offer researchers and model developers more tools to choose from.

show abstract

Section: Discussionmentioning

confidence: 99%

Systematic review of application of artificial intelligence tools in architectural, engineering and construction

Momade

Durdyev

Estrella

et al. 2021

FEBE

View full text Add to dashboard Cite

show abstract

“…Moreover, the comparative analysis of the results of calculations of reinforced-concrete statically determinable beams obtained according to the techniques of construction norms and, as a result, of applying the trained ANN with verification according to the data of samples of reinforced-concrete structures has shown that the results obtained with the use of the ANN have higher accuracy [2]. When evaluating the results of tests of reinforced-concrete frames, the accuracy is even higher in comparison with that provided in the design standards: the error when using the ANN is 6.8% versus 30-64% provided in various national design standards [3].…”

Section: Ann In the Problems Of Calculation Of Structural Unitsmentioning

confidence: 98%

Application of Artificial Neural Networks for Assessing the Reinforcement of Reinforced-Concrete Floor Slabs

Liashkevich¹

2019

Problemy Sovremennogo Betona I Zhelezobetona

View full text Add to dashboard Cite

In this article the problem of assessment of working documentation quality in terms of trustworthiness of the calculation of area of main reinforcement of reinforced-concrete structures is reviewed. In spite of development of automated designing systems, no application solutions for fully automated check of quality of working documentation for reinforced-concrete structures as regards sufficiency and necessity of reinforcement of them have been proposed until now. Moreover, this rather routine procedure can be fully automated to exclude the subjective nature of its results. Artificial neural networks (ANN) constitute the most promising mathematical model for this purpose. There are known examples demonstrating the possibility of applying the ANN for various types of calculations and analysis of experimental data for reinforced-concrete structures. In particular, the ANN allows predicting the actual deformation parameters of reinforced-concrete structures with significantly greater accuracy than any of the current national design standards. The article presents the results of calculations of reinforcement and sag for various input parameters using the example of reinforced-concrete slab structure. Using the simplest ANN with one hidden layer over the entire training sample, the predicted values with sufficient accuracy for practical use were obtained. It has been established that ANN makes it possible to predict effectively not only values of the required reinforcement for slab structures, but also their deformation. Within the framework of BIM-technologies used currently in building design, the use of ANN to assess the quality of ready-made design documentation in terms of reinforcement will reduce considerably the cost and time of relevant examinations with significantly higher trustworthiness of their results.В статье рассмотрена задача оценки качества рабочей документации в части достоверности расчета площади рабочей арматуры железобетонных конструкций. Несмотря на развитие систем автоматизированного проектирования, до настоящего времени не предложено прикладных решений для полностью автоматизированной проверки качества рабочей документации железобетонных конструкций на предмет достаточности и необходимости их армирования. При этом эта весьма относительно рутинная процедура может быть полностью автоматизирована для исключения субъективного характера ее результатов. Наиболее перспективной математической моделью для этой цели являются искусственные нейронные сети (ИНС). Известны примеры, демонстрирующие возможность прикладного применения ИНС для различного рода расчетов и анализа экспериментальных данных для железобетонных конструкций. В частности, ИНС позволяет с существенно большей точностью прогнозировать фактические параметры деформирования железобетонных конструкций, чем любые из действующих национальных норм проектирования. В статье на примере железобетонной плитной конструкции приведены результаты расчетов армирования и прогиба при различных значениях входных параметров. С помощью простейшей ИНС с одним скрытым слоем по всей обучающей выборке получены прогнозные значения с достаточной для практического их использования точностью. Установлено, что ИНС позволяет достаточно эффективно прогнозировать не только значения требуемого армирования для плитных конструкций, но и их деформации. В рамках используемых в настоящее время в строительном проектировании BIM-технологий применение ИНС для оценки качества готовой проектной документации в части армирования позволит значительно сократить стоимость и сроки соответствующих экспертиз при существенно более высокой достоверности их результатов.

show abstract

“…. SVM is a learning method to define a hyperplane for data classification and regression [18,[40][41][42]. In the regression case, the goal of the SVM is to define the hyperplane close to as many of the data points as possible [40].…”

Section: Support Vector Machine (Svm)mentioning

confidence: 99%

A Development of Seismic Fragility Curve of Box-Type Squat Shear Walls Using Data-Driven Surrogate Model Techniques

Lee,

Kwag,

2024

International Journal of Energy Research

View full text Add to dashboard Cite

Recent experience with the strongest earthquakes greater than magnitude 5.4 in Korea leads to public interest in the safety and resilience of critical infrastructures. There are many nuclear power plants near the epicenters of the earthquakes. Nuclear power plants are essential infrastructures that provide stable and enough energy for human life, and simultaneously, control systems for the safety and security of nuclear power plants are critical due to the risk of nuclear accidents on public health and the environment. The nuclear area uses probabilistic risk assessment to estimate the risk of structures, systems, and components in nuclear power plants, and the evaluation of the fragility curve is a key process for probabilistic risk assessment. The challenges of a seismic fragility analysis lie in estimating the influence of various uncertainties in material, geometry, and earthquake and improving the existing fragility analysis methods, which require time-consuming nonlinear time history analysis. Thus, this paper conducts a multivariate seismic fragility analysis using surrogate models for reinforced concrete squat shear walls and proposes a simplified closed form equation for a chosen surrogate seismic demand model. The surrogate models are trained and validated by several approaches: response surface method, support vector machine, Gaussian process regression, and neural network. In addition, a correlation analysis is used to evaluate the relative importance of the variables to the seismic demand to simplify the surrogate model further. Finally, simplified surrogate models based on the importance of the variables are proposed as closed form of polynomials, and the performance of these models on the fragility analysis is evaluated.

show abstract

Prediction of Moment Redistribution in Statically Indeterminate Reinforced Concrete Structures Using Artificial Neural Network and Support Vector Regression

Cited by 18 publications

References 38 publications

Systematic review of application of artificial intelligence tools in architectural, engineering and construction

Systematic review of application of artificial intelligence tools in architectural, engineering and construction

Application of Artificial Neural Networks for Assessing the Reinforcement of Reinforced-Concrete Floor Slabs

A Development of Seismic Fragility Curve of Box-Type Squat Shear Walls Using Data-Driven Surrogate Model Techniques

Contact Info

Product

Resources

About