Prediction the correlations between hardness and tensile properties of aluminium-silicon alloys produced by various modifiers and grain refineries using regression analysis and an artificial neural network model

[Display omitted] •The modification of Al-Si alloys using different modifiers and grain refiners improved the mechanical properties.•The best properties were obtained using a mixture from Al-3Ti-3B with Na2SiF6.•Valuable relations were obtained between hardness and tensile properties of these alloys using regression analysis and an artificial neural network model. The hardness test is considered one of the easiest mechanical tests in terms of preparing its specimens, as it does not need machining, unlike the tensile test that needs special machining and preparation. Six groups of Aluminium-Silicon alloys have been produced with different Si contents at different modifier and grains refiner. The correlations between the hardness with yield strength, ultimate tensile strength, and elongation for these groups were investigated by the regression analysis and an artificial neural network model. The measured Brinell hardness, yield strength, ultimate tensile strength, and elongation for these groups ranged between 48 - 98 HB, 49 - 103 MPa, 90 - 202 MPa, and 3 - 10.4%, respectively. The best results were obtained for a mixture of modifier (Na2SiF6) and refiner (Al-3Ti-3B). The results indicated that the trainable cascade-forward back-propagation algorithm has the best forecast accuracy for determining the percentage of Si in the produced alloys based on material properties or predicting the properties of these alloys based on the Si percentage.