Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
Search Results
Item Estimation of Strength Properties of Some Rocks using Ball Mill Grinding Characteristics(World Researchers Associations, 2025) Sahas, S.V.; Bijay, K.M.; Chandar, K.R.The strength properties of rocks namely uniaxial compressive strength and tensile strength are important in design and stability evaluation of various mining, geotechnical engineering and other rock engineering projects. Accurate determination of these properties relies on high-quality samples, but challenges like sample availability, preparation of sample, cost and time constraints have led to an increasing reliance on computational methods for prediction. In this context, an indirect approach is proposed for predicting rock strength properties, specifically the uniaxial compressive strength (UCS) and tensile strength (TS), using grinding characteristics of ball mill, an unconventional yet indirect approach. A predictive modelling using multivariate regression is carried out to estimate the relationship between UCS, TS and the grinding characteristics of ball mill. The developed models demonstrated high accuracy with R² values of 0.93 for UCS and 0.96 for TS. Performance evaluation metrics showed an RMSE of 6.03 MPa and a VAF of 93.45% for UCS and an RMSE of 0.99 MPa and a VAF of 96.47% for TS. The validation was performed using experimental UCS and TS values of basalt rocks along with ball mill grinding test data. The error analysis revealed that UCS prediction error ranged from 5.1% to 11.61% while TS prediction error varied between 4.26% and 16.39%. © 2025, World Researchers Associations. All rights reserved.Item Predicting Burden Rock Velocity in Limestone Mines using Artificial Neural Network Models(World Researchers Associations, 2025) Channabassamma, N.; Akhil, A.; Rama, S.V.; Sahas, S.V.; Ranjit, K.The prediction of burden rock velocity is crucial in optimizing the efficiency of mining and excavation operations. This study presents a novel approach utilizing Artificial Neural Networks (ANNs) to accurately predict the velocity of burden rocks based on various input parameters such as rock property, geological property and bench properties. A comprehensive dataset was collected from field measurements and laboratory experiments to train the ANN models. The performance of the ANN models such as Multi-layered Perceptron (MLP), Deep Neural Network (DNN), simple MLP and Backpropagation Neural Network (BPNN) was evaluated based on performance metrics R-squared (R)2, Mean Squared Error (MSE) and Mean Absolute Error (MAE). Among the developed ANN models, the BPNN model was found to be the most accurate predictive model for burden rock velocity, as evidenced by metrics R2(0.821), MSE (0.099) and MAE (0.226). The results indicate that the BPNN model effectively captures the complex relationships between the predictors and burden rock velocity. Advanced neural network algorithms such as recurrent neural networks and long short-term memory techniques can be used to improve the accuracy of presented neural network models. © 2025, World Researchers Associations. All rights reserved.Item A Hybrid Random Forest optimized with the Dolphin Swarm Algorithm for predicting P-Wave Velocity of Sedimentary Rocks using Ball Mill Grinding Characteristics(World Researchers Associations, 2025) Sahas, S.V.; Bijay, K.M.; Chandar, K.R.Rock properties play a crucial role in mining, geotechnical engineering and various engineering projects. P-wave velocity helps in determining the quality and stability of rock masses, essential for tunnel excavation, slope stability and mining operations. P-wave velocity also provides critical input for designing foundations for dams, bridges and other rock structures. Accurate determination of P-wave velocity relies on high-quality samples. However, challenges such as preparation, cost and time constraints have prompted a growing reliance on computational methods for its prediction. Previous investigations predominantly leaned on laboratory-based tests and indirect methodologies for predicting rock properties including P-wave velocity. In contrast, this study introduces an innovative technique for predicting wave velocity (Vp) of sedimentary rocks, particularly limestone using ball mill grinding characteristics throughout the grinding procedure, an unconventional yet effective approach. A hybrid random forest model optimized with dolphin swarm algorithm was developed to predict Vp from grinding characteristics. The performance of the model in training and testing phases was assessed based on determination coefficients (R2), root mean-squared error (RMSE) and variance account for (VAF) which are 0.984, 96.204 m/s and 98.25% in training and 0.973, 102.32 m/s and 97.63% in testing phase respectively. © 2025, World Researchers Associations. All rights reserved.Item Experimentation and Statistical Prediction of Dust Emission in Iron Ore Mines using Supervised Machine Learning (Regression) Modelling(World Researchers Associations, 2025) Rajib, P.; Harsha, V.; Senapati, A.; Sahas, S.V.In India, the mine area and the processing plant of materials such as iron ore and coal will cause dust emissions. The fugitive dust emission creates a hazardous working environment for the workers. Dust emissions will cause pulmonary-related diseases to the workers and also to the people living in nearby areas of the mine. Environmental effects such as air pollution occur due to the dispersion of particulate matter over the permissible limit in the processing area. This study evaluates dust emission levels and air quality control measures in an iron ore mine (A), Karnataka, India. Fugitive and workplace dust sampling was conducted following DGMS and MoEF and CC guidelines, with a specific focus on PM10 and PM2.5 particulate matter. Measurements revealed that dust concentrations in several mining areas exceeded the permissible limit of 1200 ?g/m³ as per the National Ambient Air Quality Standards (NAAQS, 2009). To analyze and predict these concentrations, supervised machine learning (regression) modeling including linear, polynomial (order 2) and polynomial (order 2) models, was applied. The results indicated that a third-order polynomial regression model provided the best fit for predicting dust concentrations, demonstrating lower error. The study emphasizes the necessity of more robust dust suppression measures including installing a dry fog dust suppression system, to guarantee safe working conditions and adherence to environmental regulations, even in the face of efforts to reduce dust exposure. © 2025, World Researchers Associations. All rights reserved.