English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Article history Received: 8 June 2020 Accepted: 22 June 2020 Published Online: 24 July 2020 After production of a steel product in a steel plant, a sample of the product is tested in a laboratory for its mechanical properties like yield strength (YS), ultimate tensile strength (UTS) and percentage elongation. This paper describes a mathematical model based method which can predict the mechanical properties without testing. A neural network based adaptation algorithm was developed to reduce the prediction error. The uniqueness of this adaptation algorithm is that the model trains itself very fast when predicted and measured data are incorporated to the model. Based on the algorithm, an ASP.Net based intranet website has also been developed for calculation of the mechanical properties. In the starting Furnace Module webpage, austenite grain size is calculated using semi-empirical equations of austenite grain size during heating of slab in a reheating furnace. In the Mill Module webpage, different conditions of static, dynamic and metadynamic recrystallization are calculated. In this module, austenite grain size is calculated from the recrystallization conditions using corresponding recrystallization and grain growth equations. The last module is a cooling module. In this module, the phase transformation equations are used to predict the grain size of ferrite phase. In this module, structure-property correlation is used to predict the final mechanical properties. In the Training Module, the neural network based adapation algorithm trains the model and stores the weights and bias in a database for future predictions. Finally, the model was trained and validated with measured property data.

Neural Network Based Adaptation Algorithm for Online Prediction of Mechanical Properties of Steel