基于集成学习CatBoost优化模型的爆堆大块率预测

doi:10.12068/j.issn.1005-3026.2023.12.010

东北大学学报（自然科学版） ›› 2023, Vol. 44 ›› Issue (12): 1743-1750.DOI: 10.12068/j.issn.1005-3026.2023.12.010

基于集成学习CatBoost优化模型的爆堆大块率预测

金长宇，于佳强，王强，陈立军

(东北大学深部金属矿山安全开采教育部重点实验室，辽宁沈阳110819)

发布日期:2024-01-30
通讯作者: 金长宇
作者简介:金长宇(1979-)，男，辽宁鞍山人，东北大学教授，博士生导师.
基金资助:
岩土力学与工程国家重点实验室开放基金资助项目(Z020017);中央高校基本科研业务费专项资金资助项目(N2101041).

Prediction of Blasting Fragment Large Block Percentage Ratio Based on Ensemble Learning CatBoost Model

JIN Chang-yu， YU Jia-qiang， WANG Qiang， CHEN Li-jun

Key Laboratory of Ministry of Education on Safe Mining of Deep Metal Mines， Northeastern University， Shenyang 110819， China.

Published:2024-01-30
Contact: YU Jia-qiang
About author:-
Supported by:
-

摘要/Abstract

摘要： 爆破产生的爆堆大块率问题一直以来都影响着矿山的生产效益.利用机器学习机制中集成学习思想实现大块率预测.以满洲里乌山铜矿实际采集的36组实测数据为例，整理形成10种特征数据.通过给定参数循环训练调优，再用交叉验证网格搜索的方法进行模型二次调优，并对比调优实现后模型与随机森林法、XGBoost模型、LightGBM模型和CatBoost模型进行效果对比.结果表明，经过两轮调优后的CatBoost模型预测效果明显高于其他几种模型，R²准确度可达98.83%，证明了两轮调优后CatBoost模型具有较高的预测水平，验证了该方法在大块率预测研究中的可行性，为爆破参数设计和大块率优化分析提供了可靠的参考.

关键词: 大块率预测；机器学习；集成学习；爆破；CatBoost模型

Abstract: The problem of large block percentage ratio of blasting has always affected the production benefit of mine. The implementation of large block percentage ratio of blasting predication based on ensemble learning in machine learning is studied. Firstly， 36 groups of measured data collected from Wushan Copper Mine in Manzhouli are utilized as an example， and 10 kinds of characteristic data are sorted out. Then， the model is trained and tuned by cyclic training of the given parameters， and then the method of cross-validation grid search is used to perform secondary tuning， and the effect of the model after tuning is compared with random forest method， XGBoost model， LightGBM model， and CatBoost model. The results show that the prediction effect of CatBoost model， after two rounds of tuning， is significantly higher than that of other models. The precision rate 98.83% reaches， which proves that CatBoost model has a high prediction level after two rounds of tuning. At the same time， the feasibility of this method in the study of large block percentage ratio of blasting prediction is proved， which provides a reliable reference for blasting parameter design and large block percentage ratio of blasting optimization analysis.

Key words: large block percentage ratio prediction; machine learning; ensemble learning; blasting; CatBoost model

中图分类号:

TD235.3

金长宇，于佳强，王强，陈立军. 基于集成学习CatBoost优化模型的爆堆大块率预测[J]. 东北大学学报（自然科学版）, 2023, 44(12): 1743-1750.

JIN Chang-yu， YU Jia-qiang， WANG Qiang， CHEN Li-jun. Prediction of Blasting Fragment Large Block Percentage Ratio Based on Ensemble Learning CatBoost Model[J]. Journal of Northeastern University(Natural Science), 2023, 44(12): 1743-1750.

参考文献

[1]程贵海，张勤彬，唐春海，等.数码电子雷管在高边坡爆破开挖中的应用［J］.化工矿物与加工，2018，47(8):52-55.(Cheng Gui-hai，Zhang Qin-bin，Tang Chun-hai，et al.Application of digital electronic detonator in blasting excavation of high slope［J］.Industrial Minerals & Processing，2018，47(8):52-55.)
[2]Monjezi M，Rezaei M，Varjani Y A.Prediction of rock fragmentation due to blasting in Gol-E-Gohar iron mine using fuzzy logic［J］.International Journal of Rock Mechanics and Mining Sciences，2009，46(8):1273-1280.
[3]Hudaverdi P H S W，Qiong W，Hudaverdi T，et al.Mean particle size prediction in rock blast fragmentation using neural networks［J］.Engineering Geology，2010，114(3/4):298-311.
[4]史秀志，王洋，黄丹，等.基于LS-SVR岩石爆破块度预测［J］.爆破，2016，33(3):36-40.(Shi Xiu-zhi，Wang yang，Huang Dan，et al.Prediction of rock blasting fragmentation based on LS-SVR［J］.Blasting，2016，33(3):36-40.)
[5]刘阳，谭凯旋，郭钦鹏，等.运用随机森林和GA-BP神经网络预测岩石爆破块度［J］.矿业研究与开发，2021，41(1):135-140.(Liu Yang，Tan Kai-xuan，Guo Qin-peng，et al.Prediction for blasting fragmentation of rocks using random forest and GA-BP neural network［J］.Mining Research and Development，2021，41(1):135-140.)
[6]叶海旺，胡俊杰，雷涛，等.LOO-XGboost模型预测岩石爆破块度［J］.爆破，2022，39(1):16-21.(Ye Hai-wang，Hu Jun-jie，Lei Tao，et al.LOO-XGboost model predicts rock blasting lumpiness［J］.Blasting，2022，39(1):16-21.)
[7]Monjezi M，Amiri H，Farrokhi A，et al.Prediction of rock fragmentation due to blasting in sarcheshmeh copper mine using artificial neural networks［J］.Geotechnical and Geological Engineering，2010，28(4):423-430.
[8]Esmaeili M，Osanloo M，Rashidinejad F，et al.Multiple regression，ANN and ANFIS models for prediction of backbreak in the open pit blasting［J］.Engineering with Computers，2014，30(4):549-558.
[9]Gao W，Karbasi M，Hasanipanah M，et al.Developing GPR model for forecasting the rock fragmentation in surface mines［J］.Engineering with Computers， 2018，34(2):339-345.
[10]王仁超，朱品光.基于随机森林回归方法的爆破块度预测模型研究［J］.水力发电学报，2020，39(1):89-101.(Wang Ren-chao，Zhu Pin-guang.Study on blasting fragmentation prediction model based on random forest regression method［J］.Journal of Hydroelectric Engineering，2020，39(1):89-101.)
[11]Zhou J，Li C Q，Arslan C A，et al.Performance evaluation of hybrid FFA-ANFIS and GA-ANFIS models to predict particle size distribution of a muck-pile after blasting［J］.Engineering with Computers，2021，37(1):265-274.
[12]Kulatilake P H S W，Hudaverdi T，Wu Q.New prediction models for mean particle size in rock blast fragmentation［J］.Geotechnical and Geological Engineering，2012，30(3):665-684.
[13]Ghaeini N，Mousakhani M，Amnieh H B，et al.Prediction of blasting-induced fragmentation in Meydook copper mine using empirical，statistical，and mutual information models［J］.Arabian Journal of Geosciences，2017，10(18):409.
[14]Breiman L.Random forests［J］.Machine Learning，2001，45(1):5-32.
[15]Chen T Q，Guestrin C.XGBoost:a scalable tree boosting system［C］//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Date Mining.San Francisco，2016:785-794.
[16]Ke G L，Meng Q，Finley T，et al.LightGBM:a highly efficient gradient boosting decision tree［J］.Advances in Neural Information Processing Systems，2017，30(2):188-195.
[17]Prokhorenkova L，Gusev G，Vorobev A，et al.CatBoost:unbiased boosting with categorical features［J］.Advances in Neural Information Processing Systems，2018，31:121-128.

基于集成学习CatBoost优化模型的爆堆大块率预测

Prediction of Blasting Fragment Large Block Percentage Ratio Based on Ensemble Learning CatBoost Model

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价