Multivariable predictive models for the estimation of power consumption (kW) of a Semi-autogenous mill applying Machine Learning algorithms - read research results

Aug 5, 2026

Upload by Sage

This research develops machine learning models to estimate power consumption (kW) of a semi-autogenous mill used in the mining industry. Multiple Linear Regression, Decision Tree Regression, Random Forest Regression, and Artificial Neural Networks are trained using different operational variables. The study follows an applied methodology with a descriptive, cross-sectional experimental design. Results show clear differences in predictive efficiency: RLM and ANN achieve higher determination coefficients (R²) and outperform tree-based models under MAE, MSE, and RMSE evaluations.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

AccessFree to download

Introduction

Energy consumption and sensor-driven data in mining

Importance of grinding and SAG mills

Rationale for ML-based power estimation

Materials and Methods

Data and operational variables

Machine learning models applied

Experimental design and evaluation metrics

Results and Discussion

Predictive performance across models

Determination coefficients (R²)

Error metrics: MAE, MSE, RMSE

Conclusions

Preferred models for SAG power estimation

EESj Journal of Energy & Environmental Sciences Vol. 8, N° 1, 2024 Copyright © 2024 , CINCADER. ISSN 2523-0905 A publication of CINCADER Centre of Research and Training for Regional Development Online [at www.journals.cincader.org](at www.journals.cincader.org) DOI: [https://doi.org/10.32829/eesj.v8i1.207](https://doi.org/10.32829/eesj.v8i1.207) Multivariable predictive models for the estimation of power consumption (kW) of a Semi-autogenous mill applying Machine Learning algorithms [Modelos predictivos multivariables para la estimaciónde consumo de potencia (kW) de un molino Semiautógeno aplicando algoritmos de Machine Learning] Miguel Vera R. a, , Juan Vega G. a, , Franklin Bailon V.*a, , a Facultad de Ingeniería, Universidad Nacional de Trujillo. Av. Juan Pablo II s/n –Ciudad Universitaria, Trujillo, Perú . [fbailon@unitru.edu.pe](fbailon@unitru.edu.pe) Received: 12 January 2024; Accepted: 13 February 2024; Published: 10 March 2024 Resumen Esta investigación tuvo como objetivo desarrollar modelos de aprendizaje automático (ML) para estimar el consumo de potencia (Kw) en un molino Semi-autógeno en la industria minera. Empleando algoritmos de Machine Learning considerando diversas variables operativas para los diferentes modelos como se incluyen el Regresión Lineal Múltiple (RLM), Regresión Árbol de Decisiones (RAD), Regresión Bosque Aleatorio (RBA) y Regresión Redes Neuronales Artificiales (RRNA) . La metodología adoptada fue de tipo aplicado, con un diseño experimental de enfoque descriptivo y transversal. Los resultados de la aplicación de estos modelos revelaron diferenciassignificativas en términos de eficiencia predictiva. El RLM y la RRNA destacaron con coeficientes de determinación (R²) de 0.922 y 0.939, respectivamente, indicando una capacidad sustancial para explicar la variabilidad en el consumo de potencia. En contraste, los modelos basados en árboles (RAD y RBA) mostraron desempeño inferior, con R² de 0.762 y 0. 471. Al analizarmétricas clave como el Error Absoluto Medio (MAE), el Error Cuadrático Medio (MSE) y la Raíz del Error Cuadrático Medio (RMSE), se confirmó que tanto el RLM como la RRNA superaron alos modelos basados en árboles. Estos resultados respaldan la elección de RLM y RRNA como modelos preferidos para la estimación del consumo de potencia en un molino Semi-autógeno. Palabras clave: Machine Learning, molino Semi-autógeno, potencia (kW) . Abstract This research aimed to develop machine learning (ML) models to estimate power consumption (Kw) in a Semi-autogenous mill in the mining industry. Using Machine Learning algorithms considering various operating variables for the different models such as Multiple Linear Regression (RLM), Decision Tree Regression (RAD), Random Forest Regression (RBA) and Regression Artificial Neural Networks (ANN) . The methodology adopted was applied, with an experimental design with a descriptive and transversal approach. The results of the application of these models revealed significant differences in terms of predictive efficiency. The RLM and RRNA stood out with coefficients of determination (R²) of 0.922 and 0.939, respectively, indicating a substantial capacity to explain the variability in power consumption. In contrast, the tree-based models (RAD and RBA) showed inferior performance, with R² of 0.762 and 0.471. When analyzing key metrics such as Mean Absolute Error (MAE), Mean Square Error (MSE) and Root Root Mean Square Error (RMSE), it was confirmed that both RLM and RRNA outperformed the tree-based models. These results support the choice of RLM and RRNA as preferred models for estimating power consumption in a Semi-autogenous mill. Keywords: Machine Learning, Semi-autogenous mill, power (kW) . Please cite this article as: Vera M. , Vega J. , Bailon F. , Multivariable predictive models for the estimation of power consumption (kW) of a Semi-autogenous mill applying Machine Learning algorithms , Journal of Ene

FAQs

What is the main objective of the study?

To develop machine learning models that estimate power consumption (kW) for a semi-autogenous mill in the mining industry.

Which machine learning algorithms are compared?

The study compares Multiple Linear Regression, Decision Tree Regression, Random Forest Regression, and Artificial Neural Networks, using operational variables as inputs.

How do the models perform in terms of accuracy?

RLM and ANN show stronger predictive efficiency, with higher R² values than tree-based models, and they also outperform them on MAE, MSE, and RMSE.

Aug 5, 2026

Upload by Sage