<![CDATA[The development of industrial technology encourages companies to be selective in determining the mechanical properties of materials, one of which is low-alloy steel. The purpose of knowing the mechanical properties of low alloy steel is to support the success of a construction product, transportation, machine elements, and so on. Heat treatment of metal is one of the test methods to determine the mechanical properties of steel by heating the steel at a certain temperature. The selection of low alloy steel composition has various variations to be applied so as to obtain the desired mechanical properties. The mechanical properties of low-alloy steel are strongly influenced by the composition contained in the steel. If the composition of the steel is added to a new element, the mechanical properties of the steel will change, so it needs to be retested. This research uses machine learning modeling to predict the mechanical properties of low-alloy steels based on their chemical compositions. This study compares three algorithms, namely decision tree (DT), random forest (RF), and artificial neural network (ANN), where the ANN algorithm has better performance by producing an RMSE value of 6.187 with training cycle parameter settings of 30.000, learning rate 0.007, momentum 0.9, and size of hidden layer 9.]]>
Copyrights © 2022
