A new model based on evolutionary computing for predicting ultimate pure bending of steel circular tubes
MetadataShow full item record
In this study, the feasibility of using evolutionary computing for modelling ultimate pure bending of steel circular tubes was investigated. The behaviour of steel circular tubes under pure bending is complex and highly non-linear, and the literature has a number of solutions, most of which are difficult to use in routine design practice as they do not provide a closed-form solution. This work presents a new approach, based on evolutionary polynomial regression (EPR), for developing a simple and easy-to-use formula for prediction of ultimate pure bending of steel circular tubes. The EPR model was calibrated and verified using a large database that was obtained from the literature and comprises a series of 104 pure bending tests conducted on fabricated and cold-formed tubes. The predicted ultimate pure bending of steel circular tubes using this model can be obtained from a number of inputs including the tube thickness, tube diameter, steel yield strength and modulus of elasticity of steel. A sensitivity analysis was carried out on the developed EPR model to investigate the model generalisation ability (or robustness) and relative importance of model inputs to its output. Predictions from the EPR model were compared with those obtained from artificial neural network (ANN) models previously developed by the authors, as well as most available codes and standards.The results indicate that the EPR model is capable of predicting the ultimate pure bending of steel circular tubes with a high degree of accuracy and outperforms most available codes and standards. The results also indicate that the performance of the EPR model agrees well with that of the previously developed ANN models. It was also shown that the EPR model was able to learn the complex relationship between the ultimate pure bending and most influencing factors, and render this knowledge in the form of a simple and transparent function that can be readily used by practising engineers. The advantages of the proposed EPR technique over the ANN approach were also addressed.
NOTICE: this is the author’s version of a work that was accepted for publication in Journal of Constructional Steel Research. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Journal of Constructional Steel Research, Vol. 94 (2014). DOI: 10.1016/j.jcsr.2013.11.011
Showing items related by title, author, creator and subject.
Lagat, Christopher; Roufail, R.; Evans, Brian; Soe, S. (2014)Research is currently being undertaken in Australia to develop new drilling technologies for deep mineral exploration. The Deep Exploration Technologies Cooperative Research Centre (DET CRC) has carried out a comprehensive ...
Modelling pile capacity and load-settlement behaviour of piles embedded in sand & mixed soils using artificial intelligenceAlkroosh, Iyad Salim Jabor (2011)This thesis presents the development of numerical models which are intended to be used to predict the bearing capacity and the load-settlement behaviour of pile foundations embedded in sand and mixed soils. Two artificial ...
Pham, Thong; Hadi, M. (2014)This study establishes a confinement model for FRP confined normal- and high-strength concrete circular columns. A new column parameter was suggested for estimating the compressive strength of FRP confined concrete. The ...