Effect of web perforations on end-two-flange web crippling behaviour of roll-formed aluminium alloy unlipped channels through experimental test, numerical simulation and deep learning.

Aluminium alloy has recently become popular in New Zealand's construction sector as a sustainable building material. However, in roll-formed aluminium alloy (RFA) channel beams, particularly those having web holes, cripples at points of concentrated or localised loading. In this paper, end-two-flange (ETF) web crippling behaviour of RFA unlipped channels having web holes was investigated using experimental tests, numerical simulations and a deep learning model referred as deep belief network (DBN). A total of 1080 data points was generated to train the DBN, using a finite element (FE) model that was validated using 30 new experimental results reported in this paper. Compared to the test data, the DBN predictions were conservative by around 6%. Using the DBN predictions, a comprehensive parametric study was undertaken which used the validated FE model in order to explore the effects of hole size, hole location, section thickness, and bearing plate on the web crippling strength of RFA channels having web holes. Web crippling strengths obtained from the DBN, tests and finite element analysis (FEA) were utilised to evaluate the performance of current design specifications from the American Iron and Steel Institute (AISI), Australian/New Zealand Standards (AS/NZS 1664; AS/NZS 4600:2018) and Eurocode (CEN 2006, 2007). It is shown that the current design rules are not reliable to predict the web crippling strength of RFA unlipped channels with web holes. As a consequence of the parametric analysis, new web crippling strength and web crippling strength reduction factor formulae for perforated RFA unlipped channels were proposed. A reliability analysis was then conducted, which confirmed that the proposed equations are capable of predicting the ETF web crippling strength of perforated RFA unlipped channels. • ETF web crippling behaviour of roll-formed aluminium alloy channels was studied experimentally and numerically. • Finite element analysis was conducted and validated against test results. • A deep-learning model was developed to predict the web crippling strength. • A parametric study was conducted to investigate the effects of different parameters. • New design equations were proposed for web crippling strength of roll-formed aluminium alloy channels. [ABSTRACT FROM AUTHOR]