Defect Analysis Of Banded Structure And Property Prediction Of Han Steel Mills High Aseismic Reinforcement

Performance of steel is mainly the comprehensive result of strength and toughness,it depends on the steel,organization and structure of the phase.Alloying elements can change the phase composition and properties in steel with heat treatment and cold treatment to finally form steel of different performance.Influence of Analysis of conventional elements and trace elements of different performance indexes on the specific conditions of use requirements and realize the hot rolled bars performance prediction is a hotspot of current research.In this article,through steel research,laboratory test,pilot test steel studied HRB400 E and formation mechanism of HRB500 E bar banded structure,the results show that some of the dendrite element anomaly partial poly is a major cause of the formation of banded structure,which can be improved by controlling the cooling and rolling speed to a certain extent.This paper studied abnormal element partial together,at the same time the conventional elements of causing banded structure and some trace elements on the mechanical properties of bar were also studied.We accumulated a lot of data of the production of the components and the corresponding mechanical performance since 2013 of Han Steel Mills,establishing performance forecast model which can reduce the performance test consumables and guide Steel composition optimization,promote the development of new products.The regression statistics model and the gen etic algorithm optimized Bp neural network model are mainly adopted to predict mechanical properties of HRB400 E and HRB500 E of Han Steel Mills.These two methods are realized by making alloy composition fit product mechanical property to predict yield strength,tensile strength,Under the most energetically elongation.Through regression statistical model it found that bar mechanics performance can be influenced by not only traditional elements,but some trace elements of small fluctuation can cause the sharp change of mechanical properties.Through the gen etic algorithm optimized Bp neural network,as part of routine elements,trace elements as input parameters,and output mechanical properties parameter,which obviously improves the forecast precision.The above two methods can have promising precision if they are applied in certain types of steel rolling technology and may guide the practical production.Through the existing model it can be used to to design a good alloy composition,optimize composition design conditions and provide theory basis for analyzing abnormal behavior of performance of steel caused by certain trace elements.