Faculty Publications

Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736

Publications by NITK Faculty

Browse

Filters

20252

Settings

Author: search.filters.author.Chandar, K.R.Subject: search.filters.subject.algorithm

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    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.
  • No Thumbnail Available
    Item
    Prediction of uniaxial compressive strength of limestone from ball mill grinding characteristics using supervised machine learning techniques
    (Nature Research, 2025) Swamy, S.V.; Kunar, B.M.; Chandar, K.R.; Alwetaishi, M.; Shashikumar, S.; Reddy, S.
    Uniaxial Compressive Strength (UCS) is a fundamental parameter in rock engineering, governing the stability of foundations, slopes, and underground structures. Traditional UCS determination relies on laboratory tests, but these face challenges such as high-quality core sampling, sample preparation difficulties, high costs, and time constraints. These limitations have driven the adoption of indirect approaches for UCS prediction. This study introduces a novel indirect method for predicting uniaxial compressive strength, harnessing the grinding characteristics of a ball mill as predictive variables through supervised machine learning techniques. The correlation between grinding characteristics and UCS was examined to determine whether a linear relationship exists between them. A hybrid support vector machine-recursive feature elimination (SVM-RFE) algorithm is applied to identify the critical grinding parameters influencing UCS. Four supervised machine learning models viz., Multiple Linear Regression (MLR), k-Nearest Neighbor Regression (k-NNR), Support Vector Regression (SVR), and Random Forest Regression (RFR) were developed for UCS prediction, with hyperparameter optimization performed using RandomisedSearchCV technique. The Random Forest model outperformed others as the best prediction model, achieving a coefficient of determination (R²) of 0.95, followed by SVR (R² = 0.87), k-NNR (R² = 0.82), and MLR (R² = 0.758). Model robustness was further assessed using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Variance Accounted For (VAF). Internal validation by means of K-fold cross validation and external validation with independent datasets confirmed generalization capability, showing an average prediction error of ± 10%. The findings demonstrate that combining grinding characteristics with machine learning offers an accurate, cost-effective alternative to conventional UCS testing, with significant practical applications in rock engineering. © The Author(s) 2025.