Towards Intelligent Mining for Backfill: A genetic programming-based method for strength forecasting of cemented paste backfill

Chongchong Qi, Xiaolin Tang, Xiangjian Dong, Qiusong Chen, Andy Fourie, Enyan Liu

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

As cemented paste backfill (CPB) plays an increasingly important role in minerals engineering, forecasting its mechanical properties becomes a necessity for efficient CPB design. Machine learning (ML) techniques have previously demonstrated remarkable successes in such task by providing black-box predictions. To express the non-linear relationship in an explicit and precise way, we employed genetic programming (GP) for the uniaxial compressive strength (UCS) prediction of CPB. The influence of sampling method, training set size and maximum tree depth on the GP performance was investigated. A detailed analysis was conducted on a representative GP model and the relative variable importance was investigated using the relative variable frequency, partial dependence plots and relative importance scores. The statistical parameters show that a satisfactory performance was obtained by the GP modelling (R 2 > 0.80 on the testing set). Results of this study indicate that cement-tailings ratio, solids content and curing time were the most three important variables for the UCS prediction. The predictive performance of GP modelling was comparable to well-recognised ML techniques, and the trained GP model can be generalised to entirely new tailings with satisfactory performance. This study indicates that the GP-based method is capable of providing explicit and precise forecasting of UCS, which can serve as a reliable tool for quick, inexpensive and effective assessment of UCS in the absence of adequate experimental data.

Original languageEnglish
Pages (from-to)69-79
Number of pages11
JournalMinerals Engineering
Volume133
DOIs
Publication statusPublished - 15 Mar 2019

Fingerprint

Genetic programming
backfill
Ointments
compressive strength
Compressive strength
Tailings
tailings
Learning systems
prediction
method
modeling
Minerals
Curing
mechanical property
Cements
cement
Sampling
engineering
Mechanical properties
sampling

Cite this

@article{880fe7ea9dce4f44ab833e42d23ddaa0,
title = "Towards Intelligent Mining for Backfill: A genetic programming-based method for strength forecasting of cemented paste backfill",
abstract = "As cemented paste backfill (CPB) plays an increasingly important role in minerals engineering, forecasting its mechanical properties becomes a necessity for efficient CPB design. Machine learning (ML) techniques have previously demonstrated remarkable successes in such task by providing black-box predictions. To express the non-linear relationship in an explicit and precise way, we employed genetic programming (GP) for the uniaxial compressive strength (UCS) prediction of CPB. The influence of sampling method, training set size and maximum tree depth on the GP performance was investigated. A detailed analysis was conducted on a representative GP model and the relative variable importance was investigated using the relative variable frequency, partial dependence plots and relative importance scores. The statistical parameters show that a satisfactory performance was obtained by the GP modelling (R 2 > 0.80 on the testing set). Results of this study indicate that cement-tailings ratio, solids content and curing time were the most three important variables for the UCS prediction. The predictive performance of GP modelling was comparable to well-recognised ML techniques, and the trained GP model can be generalised to entirely new tailings with satisfactory performance. This study indicates that the GP-based method is capable of providing explicit and precise forecasting of UCS, which can serve as a reliable tool for quick, inexpensive and effective assessment of UCS in the absence of adequate experimental data.",
keywords = "Cemented paste backfill, Forecasting, Genetic programming, Intelligent Mining for Backfill, Uniaxial compressive strength",
author = "Chongchong Qi and Xiaolin Tang and Xiangjian Dong and Qiusong Chen and Andy Fourie and Enyan Liu",
year = "2019",
month = "3",
day = "15",
doi = "10.1016/j.mineng.2019.01.004",
language = "English",
volume = "133",
pages = "69--79",
journal = "Minerals Engineering",
issn = "0892-6875",
publisher = "Pergamon",

}

Towards Intelligent Mining for Backfill : A genetic programming-based method for strength forecasting of cemented paste backfill. / Qi, Chongchong; Tang, Xiaolin; Dong, Xiangjian; Chen, Qiusong; Fourie, Andy; Liu, Enyan.

In: Minerals Engineering, Vol. 133, 15.03.2019, p. 69-79.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Towards Intelligent Mining for Backfill

T2 - A genetic programming-based method for strength forecasting of cemented paste backfill

AU - Qi, Chongchong

AU - Tang, Xiaolin

AU - Dong, Xiangjian

AU - Chen, Qiusong

AU - Fourie, Andy

AU - Liu, Enyan

PY - 2019/3/15

Y1 - 2019/3/15

N2 - As cemented paste backfill (CPB) plays an increasingly important role in minerals engineering, forecasting its mechanical properties becomes a necessity for efficient CPB design. Machine learning (ML) techniques have previously demonstrated remarkable successes in such task by providing black-box predictions. To express the non-linear relationship in an explicit and precise way, we employed genetic programming (GP) for the uniaxial compressive strength (UCS) prediction of CPB. The influence of sampling method, training set size and maximum tree depth on the GP performance was investigated. A detailed analysis was conducted on a representative GP model and the relative variable importance was investigated using the relative variable frequency, partial dependence plots and relative importance scores. The statistical parameters show that a satisfactory performance was obtained by the GP modelling (R 2 > 0.80 on the testing set). Results of this study indicate that cement-tailings ratio, solids content and curing time were the most three important variables for the UCS prediction. The predictive performance of GP modelling was comparable to well-recognised ML techniques, and the trained GP model can be generalised to entirely new tailings with satisfactory performance. This study indicates that the GP-based method is capable of providing explicit and precise forecasting of UCS, which can serve as a reliable tool for quick, inexpensive and effective assessment of UCS in the absence of adequate experimental data.

AB - As cemented paste backfill (CPB) plays an increasingly important role in minerals engineering, forecasting its mechanical properties becomes a necessity for efficient CPB design. Machine learning (ML) techniques have previously demonstrated remarkable successes in such task by providing black-box predictions. To express the non-linear relationship in an explicit and precise way, we employed genetic programming (GP) for the uniaxial compressive strength (UCS) prediction of CPB. The influence of sampling method, training set size and maximum tree depth on the GP performance was investigated. A detailed analysis was conducted on a representative GP model and the relative variable importance was investigated using the relative variable frequency, partial dependence plots and relative importance scores. The statistical parameters show that a satisfactory performance was obtained by the GP modelling (R 2 > 0.80 on the testing set). Results of this study indicate that cement-tailings ratio, solids content and curing time were the most three important variables for the UCS prediction. The predictive performance of GP modelling was comparable to well-recognised ML techniques, and the trained GP model can be generalised to entirely new tailings with satisfactory performance. This study indicates that the GP-based method is capable of providing explicit and precise forecasting of UCS, which can serve as a reliable tool for quick, inexpensive and effective assessment of UCS in the absence of adequate experimental data.

KW - Cemented paste backfill

KW - Forecasting

KW - Genetic programming

KW - Intelligent Mining for Backfill

KW - Uniaxial compressive strength

UR - http://www.scopus.com/inward/record.url?scp=85059823771&partnerID=8YFLogxK

U2 - 10.1016/j.mineng.2019.01.004

DO - 10.1016/j.mineng.2019.01.004

M3 - Article

VL - 133

SP - 69

EP - 79

JO - Minerals Engineering

JF - Minerals Engineering

SN - 0892-6875

ER -