RETRACTED: Analysis of influential factors for predicting the shear strength of a V-shaped angle shear connector in composite beams using an adaptive neuro-fuzzy technique (Retracted article. See vol. 31, pg. 267, 2020)

作者：Mansouri, I.; Shariati, M.; Safa, M.^*; Ibrahim, Z.; Tahir, M. M.; Petkovic, D.

来源：Journal of Intelligent Manufacturing, 2019, 30(3): 1247-1257.

DOI：10.1007/s10845-017-1306-6

• ANFIS
• Forecasting
• Shear strength
• Shear connector
• Composite
• V-shaped angle
• Push-out test
• Monotonic load

The V-shaped angle shear connector is recognized as to expand certain mechanical properties to the shear connectors, contains adequate ductility, elevate resistance, power degradation resistance under cyclic charging, and high shear transmission, more economical than other shear connectors, for instance, the L-shaped and C-shaped shear connectors. The performance of this shear connector had been investigated by previous researchers (Shariati et al. in Mater Struct 49(9):1-18, 2015), but the strength prediction was not clearly explained. In this investigation, the shear strength prediction of this connector was analyzed based on several factors. The ultimate purpose was to investigate the variations of different factors that were affecting the shear strength of this connector. To achieve this aim, the data (concrete compression strength, thickness, length, height, slope of inclination, and shear strength) were collected from the parametric studies using finite element analysis results for this purpose were input using the ANFIS method (neuro-fuzzy inference system). The finite element analysis results were verified by experimental test results. All variables from the predominant factors that were affected the shear strength of the shear connector (V-shaped angle) were also selected by using the ANFIS process. The results exhibited that the proposed shear connector (V-shaped angle) contained the potentiality to be used practically after several improvements. One option might be the improvement of the testing process for different predictive models with more input variables that will improve the predictive power of the created models.

• 出版日期2019-3