Modeling of the Sintered Density in Cu-Al Alloy Using Machine Learning Approaches

In powder metallurgymaterials, sintered density in Cu-Alalloyplays a critical role in detecting mechanical properties. Experimentalmeasurement of this property is costly and time-consuming. In thisstudy, adaptive boosting decision tree, support vector regression, k-nearest neighbors, extreme gradient boosting, and fourmultilayer perceptron (MLP) models tuned by resilient backpropagation,Levenberg-Marquardt (LM), scaled conjugate gradient, and Bayesianregularization were employed for predicting powder densification throughsintering. Yield strength, Young's modulus, volume variationcaused by the phase transformation, hardness, liquid volume, liquidustemperature, the solubility ratio among the liquid phase and the solidphase, sintered temperature, solidus temperature, sintered atmosphere,holding time, compaction pressure, particle size, and specific shapefactor were regarded as the input parameters of the suggested models.The cross plot, error distribution curve, and cumulative frequencydiagram as graphical tools and average percent relative error (APRE),average absolute percent relative error (AAPRE), root mean squareerror (RMSE), standard deviation (SD), and coefficient of correlation(R) as the statistical evaluations were utilizedto estimate the models' accuracy. All of the developed modelswere compared with preexisting approaches, and the results exhibitedthat the developed models in the present work are more precise andvalid than the existing ones. The designed MLP-LM model was foundto be the most precise approach with AAPRE = 1.292%, APRE = -0.032%,SD = 0.020, RMSE = 0.016, and R = 0.989. Lately,outlier detection was applied performing the leverage technique todetect the suspected data points. The outlier detection discoveredthat few points are located out of the applicability domain of theproposed MLP-LM model.