نوع مقاله : مقاله پژوهشی
نویسندگان
1 کارشناسی ارشد مهندسی استخراج معدن- دانشگاه صنعتی امیرکبیر
2 عضو هیأت علمی دانشکده مهندسی معدن-دانشگاه صنعتی امیرکبیر
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Circular saws are the most widely employed tools in stone cutting plants for slicing slabs into precise dimensions required for construction and decorative applications. The cutting rate of these saws is influenced by a combination of factors, including the physical and mechanical properties of the stone, such as strength, hardness, and abrasivity, as well as the technical specifications of the saw, including blade diameter and design features, and the operational parameters, such as rotational speed, feed rate, and machine power. Accurate prediction of the cutting rate is therefore essential, as it not only enables higher productivity and reduced operational costs but also supports optimal scheduling and resource allocation in stone processing operations. In this study, the Random Tree (RT) algorithm, a machine learning method capable of capturing complex nonlinear relationships, was employed to predict the cutting rate of circular saws. Four key variables were selected for model development: Uniaxial Compressive Strength (UCS) and Brazilian Tensile Strength (BTS), representing the mechanical strength of the stone; the Cerchar Abrasivity Index (CAI), representing rock abrasiveness; and the Rotation Speed of the Saw (RSS), reflecting operational conditions. The developed RT model demonstrated excellent predictive performance, achieving coefficients of determination (R²) of 0.975 for the training dataset and 0.923 for the testing dataset. In comparison, a conventional multiple linear regression model achieved a significantly lower R² of 0.77. The Root Mean Square Error (RMSE) for the RT model was 0.6167 for training and 0.9221 for testing data, further highlighting its accuracy. These findings indicate that the RT algorithm is highly effective in modeling the complex interactions between stone properties and operational parameters, providing a reliable tool for predicting cutting performance. Implementation of such predictive models in stone cutting plants can lead to improved productivity, reduced waste, optimized energy consumption, and more efficient production scheduling, thereby supporting both operational and economic objectives.
کلیدواژهها [English]
)