Study On The Mechanical Properties Prediction Of Hot-rolled High-Strength Low-Alloy Steel

Chemical composition is the basic factors which impacted the final microstructure and properties of hot-rolled material. The reasonable Composition design is the basic for the production of high quality steel products. So discussing the relationship between alloy composition and mechanical properties of steel products,and analyzing the influence of different alloy composition on mechanical properties and eventually to achieve the forecasting performance of hot-rolled sheet are currently a hot research topic. In this paper we forecast the final mechanical properties by using the known chemical composition. Practice has proved that forecasting the ultimate mechanical properties of steel by using of the performance prediction methods is better than by traditional manual sampling inspection in many aspects, such as reducing the time for testing performance, shorten the production cycle, to maintain the stability of performance of hot-rolled plates and to promote the development of new products.In this paper we use stepwise regression and artificial neural network to forecast the ultimate mechanical properties of WISCO high-strength low-alloy hot-rolled steel. The data is measured in the production process which product Q345A, Q345B, S355 and S275 of WISCO. Both of the methods are study the relationship between the alloy composition of these kinds of steels and the mechanical properties such as yield strength, tensile strength, elongation and impact energy and then get stepwise regression equations and artificial neural network models respectively.We statistic and analysis the alloy composition from the stepwise regression equations and get the main alloy composition such as C, Mn, Si, S, P, Nb and Ti, which affect the final mechanical properties. Artificial neural network model has a hidden layer of three layers BP network, the input parameters are the full range of alloy composition, the outputs are the mechanical capacities. These parameters are set based on the assumption that the production of high precision process control and a certain rolling process parameters. It proved that these two kinds of models of higher accuracy and can meet the production and application.This article also studies the main alloy compositions which affect final properties by using these established models, and mainly by changing the composition of various alloys to study the way of each alloy influence on mechanical capacities of the final product, and to provide a credible way for optimizing the mechanical properties.