A Neural Network System for Patch Load Prediction (original) (raw)
Related papers
Neural network evaluation of steel beam patch load capacity
Advances in Engineering Software, 2003
This work presents a neural network modelling to forecast steel beam patch load resistance. In preceding studies, the results of a neural network system composed of four neural networks, have been compared and calibrated with experimental data and existing design formulae, showing a good agreement. Despite these results, the adopted system did not properly consider the differences in behaviour of slender, intermediate and compact beams. This paper introduces a new strategy based on a single neural network, which is trained with a different normalisation parameter. The neural network presented a maximum error value lower than 30%, while existing formulas presented errors greater than 40%. q
Assessment of load carrying capacity of castellated steel beams by neural networks
Journal of Constructional Steel Research, 2011
In this paper, load carrying capacity of simply supported castellated steel beams, susceptible to webpost buckling, is studied. The accuracy of the nonlinear finite element (FE) method to evaluate the load carrying capacity and failure mode of the beams is discussed. In view of the high computational burden of the nonlinear finite element analysis, a parametric study is achieved based on FE and an empirical equation is proposed to estimate the web-posts' buckling critical load of the castellated steel beams. Also as other alternatives to achieve this task, the traditional back-propagation (BP) neural network and adaptive neuro-fuzzy inference system (ANFIS) are employed. In this case, the accuracy of the proposed empirical equation, BP network and ANFIS are examined by comparing their provided results with those of conventional FE analysis. The numerical results indicate that the best accuracy associates with the ANFIS and the neural network models provide better accuracy than the proposed equations.
In this paper, a sensitivity analysis of artificial neural networks (NNs) is presented and employed for estimating the patch load resistance of plate girders subjected to patch loading. To evaluate the accuracy of the proposed NN model, the results are compared with the previously proposed empirical models, so that we can estimate the resistance of plate girders subjected to patch loading. The empirical models are calibrated, for improving the formulae, with experimental data set which was collected from the corresponding literature. NNs models are later trained and validated through using the existing experimental data. In this process several NNs architectures are taken into account. A set of good NNs models are selected and then analyzed regarding their robustness when confronted with the test data set and regarding their ability to reproduce the effect of uncertainty on the data. A sensitivity analysis is conducted herein in order to investigate the effect of variability in material and geometrical properties of plate girders. Thereafter, several estimates measuring the efficiency and the quality of the NN model and the calibrated models are obtained and discussed.
SN applied sciences, 2020
This paper aims at establishing a framework for the development of artificial neural networks (ANNs) capable of realistically predicting the load-carrying capacity of reinforced concrete (RC) members. Multilayer back propagation neural networks are developed through the use of MATLAB and enriched databases which contain information describing the variation of load-carrying capacity in relation to key design parameters associated with the RC specimens (i.e. beams) considered. This work forms the basis for the development of a knowledge-based structural analysis tool capable of predicting RC structural response. A detailed discussion is provided on the different aspects of the proposed framework which include (1) the formation and analysis of the relevant (experimental) data, (2) the architecture of the ANNs, (3) the training/calibration process they undergo and finally, (4) ways of extending their applicability enabling them to predict the behaviour of RC structural forms with design parameters not represented in the available experimental database. Non-linear finite element analysis is used for validating the predictions of the ANN models developed. The comparative study reveals that the ANN models developed through the proposed framework are capable of effectively predicting the load-carrying capacity s of the RC structural forms considered quickly, accurately and without requiring significant computational resources. Keywords Artificial neural network • Database • Sampling method • Ultimate limit state • Reinforced concrete • Training process • Finite element analysis • Failure • Latin hypercube sampling List of symbols v Shear span b Width of the beam specimen cross-section d Effective depth of the beam specimen cross-section A s Area of longitudinal reinforcement acting in tension A sw Area of transverse reinforcement v ∕d Shear span to depth ratio f c Uniaxial compressive strength of concrete f yl Yield stress of longitudinal reinforcement bars f yw Yield stress of transverse reinforcement bars s Spacing between shear links l Ratio of tensile reinforcement (l = A s ∕b ⋅ d) w Ratio of transverse reinforcement (l = A sw ∕b ⋅ s) V u Shear strength Abbreviations CFP Compressive force path ANN Artificial neural network ULS Ultimate limit state LHS Latin hypercube sampling * Afaq Ahmad,
Artificial neural networks for the prediction of shear capacity of steel plate strengthened RC beams
Construction and Building Materials, 2004
This paper presents the development of multilayer feedforward artificial neural network models for predicting the ultimate shear capacity of RC beams strengthened with web bonded steel plates. Two models are constructed using the data obtained from FEM model previously developed and validated by the authors. It is found that the neural network models predict the shear capacities of beams quite accurately. The model with dimensionless parameters is found to be slightly less accurate than the ordinary model. Moreover, the neural network models predict the shear capacities of beams more accurately than the formula proposed by the authors in a previous study. Limited parametric studies show that the network models capture the underlying shear behavior of RC beams with web-bonded steel plates quite accurately. ᮊ
2017
Artificial neural networks (ANN) were used in this study to predict ultimate load of simply supported concrete beams reinforced with FRP bars under four point loading. A proposed neural model was used to predict the ultimate load of these beams. A total number of (199) beams (samples) were collected as data set and it was decided to use eight input variables, representing the dimensions of beams and properties of concrete and FRP bars, while the output variable was only the ultimate load of these beams. It was found that the use of 11 and 10 nodes in the two hidden layers was very efficient for predicting the ultimate load. The obtained results were compared with available experimental results and with the ACI 440.1R specifications. The proposed neural model gave very good predictions and more accurate results than the ACI 440.1R approach. The overall average error, in the value of the predicted ultimate load, was 3.6% and 21.7% for the proposed neural model and the ACI 440.1R appro...
Applied Sciences, 2021
The objective of this study is to compare conventional models used for estimating the load carrying capacity of reinforced concrete (RC) members, i.e., Current Design Codes (CDCs), with the method based on different assumptions, i.e., the Compressive Force Path (CFP) method and a non-conventional problem solver, i.e., an Artificial Neural Network (ANN). For this purpose, four different databases with the details of the critical parameters of (i) RC beams in simply supported conditions without transverse steel or stirrups (BWOS) and RC beams in simply supported conditions with transverse steel or stirrups (BWS), (ii) RC columns with cantilever-supported conditions (CWA), (iii) RC T-beams in simply supported conditions without transverse steel or stirrups (TBWOS) and RC T-beams in simply supported conditions with transverse steel or stirrups (TBWS) and (iv) RC flat slabs in simply supported conditions under a punching load (SCS) are developed based on the data from available experimen...
Artificial Neural Networks in Structural Engineering: Concept and Applications
Journal of King Abdulaziz University-Engineering Sciences, 1999
ABSTRACTAE Artificial neural networks are algorithms for cognitive tasks, such as learning and optimization. They have the ability to learn and generalize from examples without knowledge of rules. Research into artificial neural networks and their application to structural engineering problems is gaining interest and is growing rapidly. The use of artificial neural networks in structural engineering has evolved as a new computing paradigm, even though still very limited.
Preliminary design of reinforced concrete beams using neural networks
2000
This paper presents a backpropagation neural network model for the preliminary design of rectangular concrete beams. The model, which is developed based on the strength design procedure of the American Concrete Institute (ACI), minimizes the beam total cost including the costs of concrete, steel, and shuttering. The backpropagation neural network was successful in accurately capturing the nonlinear characteristics of the strength design procedure. The network adequately learned a set of 375 examples during the training phase. A case study, where a set of 960 new cases were considered, was used to validate the network and to demonstrate the system's generalization and fault-tolerance properties. The network showed good generalization properties since it was able to predict the correct beam depth and steel area with a fair accuracy.
Application of artificial neural networks to evaluation of ultimate strength of steel panels
Engineering Structures, 2006
Structural design of ships and offshore structures has been moving towards limit state design or reliability-based design. Improving the accuracy and efficiency of predicting ultimate strength of structural components, such as unstiffened panels and stiffened panels, has significant impact on our daily structural design. Empirical formulations have been widely used because of its simplicity and reasonable accuracy. In the past empirical formulations were generally developed by using regression analysis. The model uncertainties of good empirical formulations are around 10-15% in terms of coefficients of variation. In this paper artificial neural networks (ANN) methodology is applied to predict ultimate strength of unstiffened plates under uni-axial compression. The proposed ANN models are trained and cross-validated using the existing experimental data. Different ways to construct ANN models are also explored. It is found out that ANN models can produce more accurate prediction of ultimate strength of panels than the existing empirical formulae. The ANN model with five (original) input variables has slightly better accuracy than the model with three input variables. This demonstrates the capacity of ANN method to successfully establish a functional relationship between input and output parameters.