[1]施龙①,崔大勇①,李龙②,等.基于SSA-DBN的隧道爆破效果的预测[J].爆破器材,2025,54(04):38-45.[doi:10.3969/j.issn.1001-8352.2025.04.007]
　SHI Long,CUI Dayong,LI Long,et al.Prediction of Tunnel Blasting Outcomes Based on SSA-DBN[J].EXPLOSIVE MATERIALS,2025,54(04):38-45.[doi:10.3969/j.issn.1001-8352.2025.04.007]
点击复制

基于SSA-DBN的隧道爆破效果的预测(/HTML)

《爆破器材》[ISSN:1001-8352/CN:32-1163/TJ]

卷:
54
期数:
2025年04
页码:
38-45
栏目:
爆破技术
出版日期:
2025-07-08

文章信息/Info

Title:
Prediction of Tunnel Blasting Outcomes Based on SSA-DBN
文章编号:
5943
作者:
施龙①; 崔大勇①; 李龙②; 陈迪③; 周长春①
① 中铁建大桥工程局集团第一工程有限公司(辽宁大连, 116000)
② 西安建筑科技大学土木工程学院(陕西西安, 710055)
③ 湖北交投宜楚建设管理有限公司(湖北宜昌, 443200)
Author(s):
SHI Long①;  CUI Dayong①;  LI Long②;  CHEN Di③;  ZHOU Changchun①
① 1st Engineering Co., Ltd., China Railway Construction Bridge Engineering Bureau Group (Liaoning Dalian, 116000)
② School of Civil Engineering, Xi’an University of Architecture and Technology (Shaanxi Xi’an, 710055)
③ Hubei Jiaotou Yichu Construction Management Co., Ltd. (Hubei Yichang, 443200)
关键词:
爆破工程; DBN神经网络; 麻雀搜索算法(SSA); 爆破效果预测
Keywords:
blasting engineering;  DBN neural network;  sparrow search algorithm;  prediction of blasting outcome
分类号:
TD235.4； U455.6
DOI:
10.3969/j.issn.1001-8352.2025.04.007
文献标志码:
A
摘要:
以麒麟观隧道工程为依托，基于麻雀搜索算法（SSA）优化深度置信网络（DBN）的SSA-DBN预测模型，将选取的8种影响爆破效果的参数作为输入指标，以平均绝对误差EMA、均方误差EMS和决定系数R2作为评价指标，对DBN模型、主成分分析(PCA)优化DBN的PCA-DBN模型和SSA-DBN模型的最大线性超、欠挖和破碎块度等输出指标进行对比评价。结果表明：SSA-DBN模型最大线性超、欠挖和破碎块度的R2分别为0.997 3、 0.997 7和0.998 1；EMA分别为0.461 0、 0.338 0和0.360 2；EMS分别为0.297 5、 0.178 2和0.175 3。SSA-DBN模型对预测值与实测值的拟合程度最高，DBN模型次之，PCA-DBN模型最低。输入参数对爆破效果影响的敏感性指标r2主要在0.6~0.7之间。研究结果验证了SSA-DBN模型的准确度和稳定性。
Abstract:
A prediction study on tunnel blasting outcomes was conducted using the Qilinguan Tunnel project as an example. SSA-DBN prediction model based on sparrow search algorithm (SSA) optimized deep belief network (DBN) was used. Using the selected eight parameters that affect the blasting outcomes as input indicators, and the average absolute error EMA, mean square error EMS, and determination coefficient of R2?as evaluation indicators, a comparative evaluation was conducted on the output indicators (maximum linear over excavation, under excavation and fragmentation) of DBN model, principal component analysis (PCA) optimized DBN model (PCA-DBN), and SSA-DBN model. The results show that R2 of the maximum linear over excavation, under excavation, and fragmentation of SSA-DBN model is 0.997 3, 0.997 7, and 0.998 1, respectively. EMAis 0.461 0, 0.338 0, and 0.360 2, respectively. EMSis 0.297 5, 0.178 2, and 0.175 3, respectively. SSA-DBN model has the highest fitting degree between predicted values and actual values, followed by DBN model, and PCA-DBN model has the lowest. The sensitivity index r2of input parameters to blasting outcomes is mainly between 0.6 and 0.7. The accuracy and stability of SSA-DBN model have been verified.

参考文献/References:

［1］马鑫民, 王毅, 翟中华, 等. 岩石巷道爆破效果预测及应用效果实践研究［J］. 煤炭工程, 2022, 54(4): 92-98.

MA X M, WANG Y, ZHAI Z H, et al. Forecast model of rock roadway blasting effect and its application result［J］. Coal Engineering, 2022, 54(4): 92-98.
［2］宋战平, 张艺多, 郭德赛, 等. 邻近既有建（构）筑物隧道爆破方案评价及优化方法［J］. 土木与环境工程学报(中英文), 2023, 45(1): 14-24.
SONG Z P, ZHANG Y D, GUO D S, et al. Evaluation and optimization method of tunnel blasting scheme for adjacent existing structures ［J］. Journal of Civil and Environmental Engineering, 2023, 45(1): 14-24.
［3］单良. 基于BP神经网络的露天矿爆破参数优化研究［D］. 包头: 内蒙古科技大学, 2019.
SHAN L. Study on blasting parameters optimization of open-pit mine based on BP neural network ［D］. Baotou: Inner Mongolia University of Science & Technology, 2019.
［4］李洪超, 陈勇, 刘殿书, 等. 岩石RHT模型主要参数敏感性及确定方法研究［J］. 北京理工大学学报, 2018, 38(8): 779-785.
LI H C, CHEN Y, LIU D S, et al. Sensitivity analysis determination and optimization of rock RHT parameters［J］. Transaction of Beijing Institute of Technology, 2018, 38(8): 779-785.
［5］XU T, SONG Z P, GUO D S, et al. A cloud model-based risk assessment methodology for tunneling-induced damage to existing tunnel ［J］. Advances in Civil Engineering, 2020, 2020(1): 8898362.
［6］雷明锋, 张运波, 秦桂芳, 等. 山岭隧道爆破效果神经网络评价模型及爆破参数优化决策方法研究［J］. 现代隧道技术, 2023, 60(2): 54-61.
LEI M F, ZHANG Y B, QIN G F, et al. A study on neural network evaluation model of blasting effect in mountain tunnel and decision-making method for blasting parameter optimization ［J］. Modern tunnelling technology, 2023, 60(2): 54-61.
［7］王赟, 薛大伟, 汤万钧. 基于深度神经网络的露天矿岩石爆破效果预测［J］. 工程爆破, 2018, 24(6): 18-22.
WANG Y, XUE D W, TANG W J. Rock blasting effect forecast based on deep neural network in open pit mines［J］. Engineering Blasting, 2018, 24(6): 18-22.
［8］宋战平, 郭德赛, 徐甜, 等.基于非线性模糊层次分析法的TBM施工风险评价模型研究［J］. 岩土力学，2021, 42(5): 1424-1433.
SONG Z P, GUO D S, XU T, et al. Risk assessment model in TBM construction based on nonlinear fuzzy analytic hierarchy process［J］. Rock and Soil Mechanics, 2021, 42(5): 1424-1433.
［9］张万志. 岩质隧道炮孔图像识别算法及光面爆破参数优化研究［D］. 济南: 山东大学, 2019.
ZHANG W Z. Research on blasthole image recognition algorithms and optimization of smooth blasting parameters of rock tunnel ［D］. Ji-nan: Shandong University, 2019.
［10］戴俊, 杜晓丽. 岩石巷道楔形掏槽爆破参数研究［J］. 矿业研究与开发, 2011, 31(2): 90-93, 104.
DAI J, DU X L. Research on blasting parameters of wedge-shaped cutting for rock tunnel driving ［J］. Mining Research and Development, 2011, 31(2): 90-93, 104.
［11］刘至家. 神经网络规模自适应技术研究［D］.南京: 南京大学, 2019.
LIU Z J. Research on selfadaption technology of the scale of neural networks ［D］. Nanjing: Nanjing University, 2019.
［12］施建俊, 李庆亚, 张琪, 等. 基于Matlab和BP神经网络的爆破振动预测系统［J］. 爆炸与冲击, 2017, 37(6): 1087-1092.
SHI J J, LI Q Y, ZHANG Q, et al. Forecast system for blasting vibration velocity peak based on Matlab and BP neural network ［J］. Explosion and Shock Waves, 2017, 37(6): 1087-1092.
［13］SONG Z P, CHENG Y, ZHANG Z K, et al. Tunnelling performance prediction of cantilever boring machine in sedimentary hard-rock tunnel using deep belief network［J］. Journal of Mountain Science, 2023, 20(7): 2029-2040.
［14］王功明. 深度信念网络结构优化设计方法与应用［D］. 北京: 北京工业大学, 2019.
WANG G M. Optimization design for deep belief network and its application ［D］. Beijing: Beijing University of Technology, 2019.
［15］HUO R K, LIU H D, LI N, et al. An expert system for prediction fissure rock water in neotectonic movement area based on fuzzy neural network ［J］. Systems Engineering-Theory & Practice, 2003, 23(11): 135-139.
［16］TIAN X X, SONG Z P, WANG J B. Study on the propagation law of tunnel blasting vibration in stratum and blasting vibration reduction technology ［J］. Soil Dynamics and Earthquake Engineering, 2019, 126: 105813.
［17］李杰, 韩正之. 神经网络的学习误差函数及泛化能力［J］. 控制与决策, 2000,15(1): 95-97.
LI J, HAN Z Z. The learning error function of neural network and its generalization ［J］. Control and Decision, 2000,15(1): 95-97.
［18］杨寒雨, 赵晓永, 王磊. 数据归一化方法综述［J］. 计算机工程与应用, 2023, 59(3): 13-22.
YANG H Y, ZHAO X Y, WANG L. Review of data normalization methods ［J］. Computer Engineering and Applications, 2023, 59(3): 13-22.
［19］GUO D S, SONG Z P, XU T, et al. Coupling analysis of tunnel construction risk in complex geology and construction factors ［J］. Journal of Construction Engineering and Management, 2022, 148(9): 4022097.
［20］宋森森, 霍润科. 铁路隧道空孔直线掏槽爆破方案优化研究及应用［J］. 爆破, 2023, 40(4): 66-72, 81.
SONG S S, HUO R K. Optimization research and application of burn cut blasting with empty hole for railway tunnel ［J］. Blasting, 2023, 40(4): 66-72, 81.
［21］赵蔷. 主成分分析方法综述［J］. 软件工程, 2016, 19(6): 1-3.
ZHAO Q. A review of principal component analysis ［J］. Software Engineering, 2016, 19(6): 1-3.

备注/Memo

备注/Memo:
收稿日期：2024-04-19
基金项目：陕西省创新能力支撑计划（2020TD-005）
第一作者：施龙（1988—），男，高级工程师，主要从事地下工程与隧道工程方面的研究工作。E-mail: 565835852@qq.com
通信作者：李龙（2000—），男，硕士研究生，研究方向为隧道与地下工程。E-mail: 1283180643@qq.com
更新日期/Last Update: 2025-07-08