[1]王昱琛①,杨仕教①,郭钦鹏①,等.基于MPA-SVM的煤矿抛掷爆破爆堆形态预测[J].爆破器材,2023,52(01):58-64.[doi:10.3969/j.issn.1001-8352.2023.01.010]
　WANG Yuchen,YANG Shijiao,GUO Qinpeng,et al.Prediction of Blasting Muckpile Morphology in Throw Blasting of Coal Mine Based on MPA-SVM[J].EXPLOSIVE MATERIALS,2023,52(01):58-64.[doi:10.3969/j.issn.1001-8352.2023.01.010]
点击复制

基于MPA-SVM的煤矿抛掷爆破爆堆形态预测()
分享到：

《爆破器材》[ISSN:1001-8352/CN:32-1163/TJ]

卷:
52
期数:
2023年01
页码:
58-64
栏目:
爆破技术
出版日期:
2023-01-08

文章信息/Info

Title:
Prediction of Blasting Muckpile Morphology in Throw Blasting of Coal Mine Based on MPA-SVM
文章编号:
5732
作者:
王昱琛①; 杨仕教①; 郭钦鹏①; 尹裕①②
①南华大学资源环境与安全工程学院（湖南衡阳,421000）
②中钢集团马鞍山矿山研究总院股份有限公司（安徽马鞍山,243000）
Author(s):
WANG Yuchen① ;  YANG Shijiao① ;  GUO Qinpeng① ;  YIN Yu①②
① School of Recourse Environment and Safety Engineering, University of South China (Hu’nan Hengyang, 421000)
② Maanshan General Institute of Mining Research Co., Ltd., Sinosteel Group Co., Ltd. (Anhui Maanshan, 243000)
关键词:
煤矿抛掷爆破; 爆堆形态预测; MPASVM; Weibull模型
Keywords:
throw blasting of coal mine;  prediction of blast muckpile morphology;  MPA-SVM;  Weibull model
分类号:
TD235
DOI:
10.3969/j.issn.1001-8352.2023.01.010
文献标志码:
A
摘要:
为提高爆堆形态预测精度，提出了一种海洋捕食者算法（MPA）优化支持向量机（SVM）的方法，结合黑岱沟露天煤矿爆破工程数据，选取其中8个参数作为影响爆堆形态的输入参数，松散系数ξ和Weibull函数的2个控制变量α、β为输出参数，建立基于MPA-SVM的爆堆形态预测模型，并与同期使用的5个模型进行比较。结果表明：MPA-SVM的预测效果优于其他5个模型，相对误差未超过5%，3个评价指标分别为R2（0.955,0.978,0.946），RMSE（0.063,0.075,0.116），RMAE（0.046,0.056,0.067），证明了MPA-SVM对爆堆形态预测的适用性，且在小样本数据条件下更具有精度优势。
Abstract:
In order to improve the prediction accuracy of blast muckpile morphology, a support vector machine (SVM) optimization method based on marine predator algorithm (MPA) was proposed. Combined with the blasting engineering data of Heidaigou Open-Pit Mine, eight parameters were selected as the input parameters affecting the morphology of blast muckpile, looseness coefficient ξ and two control variables α?and β?of Weibull function were chosen as the output parameters, and the MPA-SVM prediction model of blast muckpile morphology was established. Prediction results of MPA-SVM model were compared with those of the other five models. It shows that the prediction results of MPA-SVM model is better than those of the other models, its relative error does not exceed 5%, and the three evaluation indexes are R2(0.955,0.978,0.946), RMSE(0.063,0.075,0.116), and RMAE?(0.046,0.056,0.067).It is proved that MPA-SVM model is applicable to prediction of blast muckpile morphology, and it has more accuracy advantages under the condition of small sample data.

参考文献/References:

［1］陈庆凯, 赵文龙, 郭赵杰, 等.基于模糊神经网络爆破效果评价模型［J］.工程爆破, 2019, 25(4): 74-79.
CHEN Q K, ZHAO W L, GUO Z J, et al. Blasting effect evaluation model based on fuzzy neural network ［J］. Engineering Blasting, 2019, 25(4): 74-79.
［2］马力, 李天翔, 来兴平, 等.基于Fourier级数的露天矿抛掷爆破效果GALSSVM预测［J/OL］.煤炭学报. http: //doi.org/10.13225/j.cnki.jccs.2021. 1832.
MA L, LI T X, LAI X P, et al. Genetic algorithm optimization and least squares support vector machine prediction of blasting casting effect in open-pit mine based on fourier series［J/OL］. Journal of China Coal Society. http: //doi.org/10.13225/j.cnki.jccs.2021. 1832.
［3］CHESKIDOV V I, TSYMBALYUK T A, REZNIK A V. Enhancing efficiency of direct dumping by cast blasting of overburden rocks ［J］. Journal of Mining Science, 2020, 56(2): 246-251.
［4］CHEN M, WEI D, YI C P, et al. Failure mechanism of rock mass in bench blasting based on structural dynamics［J］. Bulletin Engineering Geology and the Environment, 2021, 80: 6841-6861.
［5］李胜林, 王宗睿, 孙建成, 等.基于无人机航测的露天矿爆堆形态与延期间隔的研究［J］.爆破, 2021, 38(1): 70-74.
LI S L, WANG Z R, SUN J C, et al. Blasting shape and delay interval of open pit mine based on UVA aerial survey ［J］. Blasting, 2021, 38(1): 70-74.
［6］齐留洋, 郭建新, 顾春雷.基于Weibull模型的爆堆形态神经网络预测［J］.有色金属(矿山部分), 2019, 71(3): 92-95.
QI L Y, GUO J X, GU C L. Prediction of blastheap morphology based on Weibull model ［J］. Nonferrous Metals(Mining Section), 2019, 71(3): 92-95.
［7］韩亮, 毕文广, 李红江.基于Weibull模型的爆堆形态主要影响因素的灰色关联分析［J］.矿业研究与开发, 2011, 31(5): 112-115, 120.
HAN L, BI W G, LI H J. Grey correlation analysis on main factors influencing the shape of blasted stockpile based on Weibull model ［J］. Mining Research and Development, 2011, 31(5): 112-115，120.
［8］李志航, 周永利, 付相超.基于Weibull模型的BP神经网络在黑岱沟露天煤矿爆堆预测中的应用研究［J］.露天采矿技术, 2012(5): 27-29.
［9］刘希亮, 赵学胜, 陆锋, 等.基于GA-SVM的露天矿抛掷爆破抛掷率预测［J］.煤炭学报, 2012, 37(12): 1999-2005.
LIU X L, ZHAO X S, LU F, et al. A GA-SVM based model for throwing rate prediction in the open-pit cast blasting［J］. Journal of China Coal Society, 2012, 37(12): 1999-2005.
［10］温廷新, 陈晓宇, 邵良杉, 等.参数优化GA-ELM模型在露天煤矿抛掷爆破的预测［J］.煤炭学报, 2017, 42(3): 630-638.
WEN T X, CHEN X Y, SHAO L S, et al. Prediction on parameters optimized GAELM model for cast blasting in open-pit mine［J］. Journal of China Coal Society, 2017, 42(3): 630-638.
［11］FARAMARZI A, HEIDARINEJAD M, MIRJALILI S, et al. Marine predators algorithm: a nature-inspired metaheuristic［J］. Expert Systems with Applications, 2020, 152: 113377.
［12］CORTES C, VAPNIK V N. Support-vector networks ［J］. Machine Learning, 1995, 20(3): 273-297.

备注/Memo

备注/Memo:
收稿日期：2021-06-01
基金项目：湖南省研究生科研创新项目（CX20200916 ，QL20210216, QL20220211）
第一作者：王昱琛（1998-），女，硕士，主要从事矿山工程爆破的研究。E-mail：1299916821@qq.com
通信作者：杨仕教（1965-），男，教授，博导，主要从事采矿工程、岩土工程研究。E-mail：649292197@qq.com
更新日期/Last Update: 2023-01-07