Shear strength parameters prediction of rock materials using hybrid machine learning model

Nondestructive Testing And Evaluation, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 24

Published: March 7, 2025

Language: Английский

---

Predicting Porosity in Grain Compression Experiments Using Random Forest and Metaheuristic Optimization Algorithms

Food Science & Nutrition, Journal Year: 2025, Volume and Issue: 13(4)

Published: March 28, 2025

ABSTRACT Grain stored for long periods is highly susceptible to localized condensation, mold growth, and insect infestations, leading significant storage losses. These issues are particularly acute in large‐capacity bungalow warehouses, where food security concerns even more pronounced. The porosity of grain piles a critical parameter that influences heat moisture transfer within the mass, as well ventilation storage. To investigate distribution pattern bulk pile this study employs machine learning (ML) techniques predict based on compression experiments. Four metaheuristic optimization algorithms—particle swarm (PSO), gray wolf optimizer (GWO), sine cosine algorithm (SCA), tunicate (TSA)—were introduced enhance random forest (RF) algorithm, five ML‐based models (RF, PSO‐RF, GWO‐RF, SCA‐RF, TSA‐RF) predicting were developed. predictive performance was analyzed using error analysis, Taylor diagrams, evaluation metrics, multi‐criteria assessments identify optimal ML prediction model. results indicate four RF‐based hybrid surpasses single RF Among these models, TSA‐RF model demonstrated best performance, achieving R 2 values 0.9923 training set 0.9723 test set. employed conduct hierarchical warehouse. exhibits being higher middle smaller at edges depth increases. developed offers novel efficient method porosity, enabling rapid

Language: Английский

---

Performance of Statistical and Intelligent Methods in Estimating Rock Compressive Strength

Sustainability, Journal Year: 2023, Volume and Issue: 15(7), P. 5642 - 5642

Published: March 23, 2023

This research was conducted to forecast the uniaxial compressive strength (UCS) of rocks via random forest, artificial neural network, Gaussian process regression, support vector machine, K-nearest neighbor, adaptive neuro-fuzzy inference system, simple and multiple linear regression approaches. For this purpose, geo-mechanical petrographic characteristics sedimentary in southern Iran were measured. The effect petrography on assessed. carbonate sandstone samples classified as mudstone grainstone calc-litharenite, respectively. Due shallow depth studied mines low amount quartz minerals samples, rock bursting phenomenon does not occur these mines. To develop UCS predictor models, porosity, point load index, water absorption, P-wave velocity, density considered inputs. Using variance accounted for, mean absolute percentage error, root-mean-square-error, determination coefficient (R2), performance index (PI), efficiency methods evaluated. Analysis model criteria using allowed for development a user-friendly equation, which proved have adequate accuracy. All intelligent (with R2 > 90%) had excellent accuracy estimating UCS. difference average all six with measured value equal +0.28%. By comparing methods, machine radial basis function predicting (R2 = 0.99 PI 1.92) outperformed other investigated.

Language: Английский

---

Bearing capacity prediction of the concrete pile using tunned ANFIS system

Journal of Engineering and Applied Science, Journal Year: 2024, Volume and Issue: 71(1)

Published: Feb. 12, 2024

Abstract The design process for pile foundations necessitates meticulous deliberation of the calculation pertaining to bearing capacity piles. primary objective this work was investigate potential use Coot bird optimization ( $${\text{CBO}}$$ CBO ) techniques in predicting load-bearing concrete-driven Despite availability several suggested models, investigation estimating pile-carrying has been somewhat neglected research. This presents and validates a unique approach that combines model with Multi-layered perceptron $${\text{MLP}}$$ MLP neural network adaptive neuro-fuzzy inference system $${\text{ANFIS}}$$ ANFIS ). findings 472 different driven static load tests were put database. proposed framework's building, validation, testing stages each accomplished utilizing training set (70%), validation (15%), (15%) dataset, respectively. According findings, $${{\text{MLP}}}_{{\text{CBO}}}$$ $${{\text{ANFIS}}}_{{\text{CBO}}}$$ both offer remarkable possibilities accurately pile-bearing given structure. $${R}^{2}$$ R 2 values during stage 0.9874, while validating stage, they 0.9785, 0.987. After considering various kinds performance studies contrasting them existing literature, it concluded provides more appropriate

Language: Английский

---

Mechanical characteristics of waste-printed circuit board-reinforced concrete with silica fume and prediction modelling using ANN

Environmental Science and Pollution Research, Journal Year: 2024, Volume and Issue: 31(19), P. 28474 - 28493

Published: April 1, 2024

Language: Английский

---