MétaCan
Menu
Back to cohort
Record W6986634382

Prédiction des propriétés mécaniques des remblais miniers en pâte cimentés par des approches de l’intelligence artificielle (IA)

2024· other· fr· W6986634382 on OpenAlexfundno aff

Bibliographic record

VenueDepositum (Université du Québec en Abitibi-Témiscamingue) · 2024
Typeother
Languagefr
Field
Topic
Canadian institutionsnot available
FundersNatural Sciences and Engineering Research Council of CanadaGovernment of Canada
KeywordsOccupational trainingPlankDomain (mathematical analysis)
DOInot available

Abstract

fetched live from OpenAlex

RÉSUMÉ À l'ère du numérique, l'industrie minière aura besoin de disposer d'outils puissants pour prédire les propriétés mécaniques des remblais miniers pour aider à sélectionner des recettes optimales des mélanges de remblai en pâte cimenté (RPC) dans le but de satisfaire aux contraintes technico-économiques de l’exploitation. Grâce aux progrès récents dans le déploiement de l'intelligence artificielle (IA), l'apprentissage automatique (Machine Learning – ML) et l’apprentissage profond (Deep Learning - DL) sont devenus des techniques de pointe pouvant être employées dans le secteur minier. C’est dans cette perspective que s'inscrit l'objectif de développer des modèles de Machine Learning et d’un réseau de neurones profond (Deep Neural Network - DNN) pour prédire la résistance à la compression uniaxiale (UCS) des remblais en pâte cimentés. Pour ce faire, de nombreuses données ont été collectées auprès des mines d’Agnico-Eagle pour ajouter à des résultats de laboratoire afin de construire une base de données (BD) riche. Avant l’apprentissage des modèles de ML et le DNN, un prétraitement et des analyses ont été faits sur cette BD en optimisant également les hyperparamètres des modèles. En s’appuyant sur des indicateurs de performance, les modèles de prédiction les plus performants ont été sélectionnés; GBR et DNN qui ont donné des coefficients de détermination (R) égale à 0.970 et 0.969 respectivement. Ces modèles ont été validés au laboratoire par la préparation des nouveaux mélanges de RPC. Dans le but de mettre ces modèles en production et pour qu'ils soient utilisables par l’industrie minière, une application web sera développée afin de faciliter leurs exécutions par les utilisateurs. ABSTRACT In the digital era, the mining industry benefits from powerful tools that can help to select optimal cement paste backfill (CPB) mix recipes to meet the techno-economic constraints of mining. With recent advances in the deployment of Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning (DL) creates state-of-the-art techniques in the mining sector that can be employed in the mining sector. This is the perspective behind the objective of developing Machine Learning and Deep Neural Network (DNN) models to predict the uniaxial compressive strength (UCS) of cemented paste backfill. To achieve this, extensive data were collected from Agnico-Eagle's mines and combined with laboratory results to build a rich database (DB). Before training the ML models and DNN, pre-processing and analyses were carried out on this DB, also optimizing the models' hyperparameters. Based on the performance indicators, the best performing prediction models were selected; GBR and DNN which gave determination coefficients equal to 0.970 and 0.969 respectively. These models were validated in the laboratory by the preparation of new CPB mixtures. In order to put this model into production and make it usable by the mining industry, a web application will be developed to facilitate its execution by users.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.006
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.014
Threshold uncertainty score0.028

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0020.006
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0020.002
Science and technology studies0.0010.001
Scholarly communication0.0020.002
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0040.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.010
GPT teacher head0.229
Teacher spread0.219 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2024
Admission routes1
Has abstractyes

Explore more

Same venueDepositum (Université du Québec en Abitibi-Témiscamingue)French-language works237,207