Heat treatment effects on tribological characteristics for AISI A8 tool steel and development of wear mechanism maps using K means clustering and neural networks

Abstract

Viking steel classified under AISI A8 cold working tool steel is widely used for heavy duty blanking and forming operations. The excellent combination of wear resistance and toughness make this material superior when compared to other tool steels in such applications. This cold working tool steel is easy to machine and gives excellent mechanical properties upon proper heat treatment. The heat treatment carried out with air and oil as quenching media at various conditions of temperature and time are discussed. Optical microscopy and hardness tests are performed on the heat-treated specimens. The samples with best results from each group are selected for further analysis. The tribological properties of these specimens are analysed with the help of wear tests conducted on pin on disc tribometer under different load and velocity levels. The data obtained from these experiments are used to find out the coefficient of friction and wear rate. The experimental results are used to develop wear mechanism maps with the help of K means clustering analysis and probabilistic neural networks. A novel approach based on clustering analysis is proposed as the analysis of wear data and wear mechanism maps are developed using with neural networks for tool steel. The data analysis is correlated with the scanning electron microscopic observations and the results are discussed.

Keywords:

• Cold working tool steel
• K means clustering analysis
• probabilistic neural networks
• tribological properties
• wear mechanism maps