Python Data Analysis and Regression Plots of Wear and Hardness Characteristics of Laser Cladded Ti and TiB2 Nanocomposites on Steel Rail

Abstract

A predictive statistical correlation and relationship between the wear rate and the hardness was carried out. A linear and quadratic polynomial regression machine learning details of the factors relationships was studies and stated. An independent variable of hardness property and dependent variable of wear rate property of cladded Ti and TiB2 on carbon steel were proposed. Both linear and quadratic models revealed insignificant lack of fit with their degree of freedom being 3 and 2 respectively. There variables terms are significant, and the models not aliased. The Adjusted R-squared in the model was given as 0.06613 in linear regression and 0.8883 in quadratic regression model summary. Analysis of variance design revealed the responses for the models of their sum of squares and mean of squares with resultant residual of squares values of 0.16318 of the linear regression and 0.0228 of the quadratic regression in a significant reduction postulation. The F-Value derived is significant with 0.75189 value in the linear regression and 7.94963 value in the quadratic regression. The result also correlates with the Python data analysis.The predictive equation for the linear and quadratic polynomial regression were given to enable predictive determination of dependent variable of the wear rate from their dependent values of the micro-hardness property values evaluation. A clear optimization relevance of higher order polynomial regression analysis of the quadratic for maximised analytical results were stated and emphasized.