基于PSO-DSRVM的边坡变形预测

袁于思; 冯小鹏; 李勇; 易灿灿

doi:10.16031/j.cnki.issn.1003-8035.202112032

基于PSO-DSRVM的边坡变形预测

1.
中铁武汉电气化局集团第一工程有限公司，湖北武汉　430074

2.
武汉科技大学，湖北武汉　430081

基金项目: 国家自然科学基金项目（51805382）；湖北省安全生产专项资金科技项目（KJZX202007003）

详细信息

作者简介: 袁于思（1974-），男，湖北通山人，本科，主要从事结构健康监测的研究。E-mail：1197693411@qq.com

通讯作者: 易灿灿（1989-），男，湖北松滋人，博士，主要从事结构健康监测的研究。E-mail：zhiliwangmr@163.com

中图分类号: P642.22

收稿日期: 2021-12-27

修回日期: 2022-04-18

刊出日期: 2023-02-25

Prediction of mine slope deformation based on PSO-DSRVM

1.
China Railway Wuhan Electrification Bureau Group Co.Ltd.,Wuhan, Hubei　430074, China

2.
Wuhan University of Science and Technology, Wuhan, Hubei　430081, China

More Information

Corresponding author: YI Cancan, zhiliwangmr@163.com

Received Date: 27 December 2021

Revised Date: 18 April 2022

Publish Date: 25 February 2023

摘要

摘要:
为了建立高精度的边坡位移预测模型，文章采用基于粒子群优化（PSO）的双稀疏相关向量机（DSRVM）建立边坡稳定性和影响因素之间的非线性关系。双稀疏相关向量机是在变分和相关向量机（RVM）框架下提出的一种多核组合优化的方法，相比于RVM和其他多核学习方法，DSRVM不仅有更少的训练时间，并且能够得到更高的预测精度。由于DSRVM的核参数对预测效果的影响较大，文章采用粒子群算法实现多个核参数的优化选取并应用于边坡位移预测。最后将本文提出的基于粒子群优化的双稀疏相关向量机（PSO-DSRVM）预测结果与极限学习机 (ELM)和小波神经网络(WNN)预测结果进行对比，通过均方根误差（RMSE）、复相关系数(R²)和平均相对预测误差（ARPE）进行评价，验证了PSO-DSRVM模型在边坡变形预测上的可行性。
- 边坡稳定 /
- 位移预测 /
- 粒子群优化算法 /
- 双稀疏相关向量机
Abstract:
In order to establish a high-precision prediction model of mine slope displacement, Doubly Sparse Relevance Vector Machine （DSRVM） based on Particle Swarm Optimization （PSO） was used to establish the nonlinear relationship between slope stability and influencing factors in this paper. DSRVM was a multi-core combinatorial optimization method, which was proposed under the framework of variational and Relevance Vector Machines （RVM）. Compared with RVM and other multiple-kernel learning methods, DSRVM not only had less training time, but also can obtained higher prediction accuracy. Aiming at the influence of the parameter’s selection of DSRVM on the final prediction effect, the optimal multiple kernel parameters was determined by PSO algorithm to be used in the mine slope displacement prediction. Compared the computational results of DSRVM with Extreme Learning Machine （ELM） and Wavelet Neural Network （WNN）, the feasibility of PSO-DSRVM in slope deformation prediction was verified by the evaluation indicators such as RMSE, R² and ARPE.
- slope stability /
- displacement prediction /
- particle swarm optimization algorithm /
- doubly sparse relevance vector machine

HTML全文

图 1 边坡位移预测流程图

Figure 1.

下载: 全尺寸图片幻灯片

图 2 极限学习机的预测结果

Figure 2.

下载: 全尺寸图片幻灯片

图 3 小波神经网络的预测结果

Figure 3.

下载: 全尺寸图片幻灯片

图 4 本文提出的PSO-DSRVM模型预测结果

Figure 4.

下载: 全尺寸图片幻灯片

图 5 PSO-DSRVM预测的精度

Figure 5.

下载: 全尺寸图片幻灯片

表 1 边坡变形量与影响因素监测数据统计样本

Table 1. Statistical sample of monitoring data of slope deformation and influencing factors

编号	X₁/mm	X₂/mm	X₃/kPa	X₄/mm	X₅/（°）	X₆/（°）	X₇/kPa	X₈/mm
1	19	2	12	0	24	33	188.45	14.4
2	12	7	20	0	10	22	115.91	11.7
3	36	4	68	0	44	51	177.86	16.5
4	16	6	30	0	20	40	190.41	18.5
5	16	3	30	0	20	40	188.73	20.3
6	36	2	68	0	38	49	109.65	41.2
7	12	4	35	10	10	22	120.35	25.6
8	30	12	50	40	39	46	159.19	51.3
9	32	1	55	0	38	42	189.07	24.1
10	27	2	46	0	38	46	180.35	11.4
11	23	6	39	12	31	40	190.45	26.8
12	25	7	40	35	35	43	192.13	27.2
13	13	11	20	30	26	30	194.88	18.5
14	16	3	25	0	20	40	182.21	10.9
15	31	1	50	0	38	47	187.79	21.4
16	26	4	49	15	36	48	190.17	22.8
17	23	5	40	0	31	42	115.28	14.4
18	33	4	55	0	39	49	102.77	32.5
19	24	4	40	0	33	41	182.83	18.5
20	15	13	25	24	43	27	194.32	22.4
21	26	8	46	28	36	45	139.18	53.1
22	18	2	30	0	23	32	182.48	11.2
23	25	0	40	0	32	43	106.49	23.7
24	24	8	40	30	33	41	190.45	23.1
25	22	2	37	0	30	39	189.03	24.5
26	21	3	35	0	24	37	113.58	36.6
27	21	2	35	0	32	36	181.89	10.6
28	32	2	55	0	41	49	184.69	21.2
29	23	6	39	0	31	40	116.69	15.1
30	33	11	55	40	44	46	193.55	29.8
31	15	3	25	0	18	35	181.89	12.5
32	18	3	32	0	23	37	191.03	21.2

下载: 导出CSV

表 2 三种不同模型预测效果的比较

Table 2. Comparison of prediction effects of different models

评价指标	PSO-DSRVM	极限学习机	小波神经网络
RMSE	0.478	1.88	1.14
R²	0.99	0.72	0.77
ARPE	0.017	0.29	0.27

下载: 导出CSV

参考文献(15)

[1]	王鹏飞. 基于GM-RBF组合模型的高路堑边坡稳定性预测研究[J]. 建筑结构,2021,51(20):140 − 145. [WANG Pengfei. Study on stability prediction of high cutting slope based on GM-RBF combination model[J]. Building Structure,2021,51(20):140 − 145. (in Chinese with English abstract) doi: 10.19701/j.jzjg.2021.20.023
[2]	方然可,刘艳辉,苏永超,等. 基于逻辑回归的四川青川县区域滑坡灾害预警模型[J]. 水文地质工程地质,2021,48(1):181 − 187. [FANG Ranke,LIU Yanhui,SU Yongchao,et al. A early warning model of regional landslide in Qingchuan County,Sichuan Province based on logistic regression[J]. Hydrogeology & Engineering Geology,2021,48(1):181 − 187. (in Chinese with English abstract) doi: 10.16030/j.cnki.issn.1000-3665.201911034
[3]	史笑凡,杨春风. 基于支持向量机的边坡垂直位移方向率预测及边坡稳定性研究[J]. 河北工业大学学报,2021,50(1):92 − 98. [SHI Xiaofan,YANG Chunfeng. Support vector machine for vertical displacement direction rate of slope estimate and slope stability research[J]. Journal of Hebei University of Technology,2021,50(1):92 − 98. (in Chinese with English abstract)
[4]	晏红波,杨庆,任超,等. 基于EEMD的BP神经网络边坡预测研究[J]. 水力发电,2017,43(7):37 − 40. [YAN Hongbo,YANG Qing,REN Chao,et al. Research on side slope prediction using BP neural network based on EEMD[J]. Water Power,2017,43(7):37 − 40. (in Chinese with English abstract) doi: 10.3969/j.issn.0559-9342.2017.07.010
[5]	邓超,胡焕校,张天乐,等. 基于改进极限学习机模型的岩质边坡稳定性评价与参数反演[J]. 中国地质灾害与防治学报,2020,31(3):1 − 10. [DENG Chao,HU Huanxiao,ZHANG Tianle,et al. Stability evaluation and parameter inversion of rock slope using modified extreme learning machine model[J]. The Chinese Journal of Geological Hazard and Control,2020,31(3):1 − 10. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2020.03.01
[6]	韩连生,宋光仁. 基于灰色理论的露天矿边坡位移预测预警研究[J]. 化工矿物与加工,2018,47(5):48 − 52. [HAN Liansheng,SONG Guangren. Research of prediction and early warning of slope displacement in open-pit mines based on Gray Theory[J]. Industrial Minerals & Processing,2018,47(5):48 − 52. (in Chinese with English abstract) doi: 10.16283/j.cnki.hgkwyjg.2018.05.012
[7]	李麟玮,吴益平,苗发盛,等. 基于变分模态分解与GWO-MIC-SVR模型的滑坡位移预测研究[J]. 岩石力学与工程学报,2018,37(6):1395 − 1406. [LI Linwei,WU Yiping,MIAO Fasheng,et al. Displacement prediction of landslides based on variational mode decomposition and GWO-MIC-SVR model[J]. Chinese Journal of Rock Mechanics and Engineering,2018,37(6):1395 − 1406. (in Chinese with English abstract) doi: 10.13722/j.cnki.jrme.2017.1508
[8]	LIU Z Q,GUO D,LACASSE S,et al. Algorithms for intelligent prediction of landslide displacements[J]. Journal of Zhejiang University-SCIENCE A,2020,21(6):412 − 429. doi: 10.1631/jzus.A2000005
[9]	罗亦泳,张豪,张立亭. 基于进化相关向量机的边坡安全系数估算[J]. 人民黄河,2016,38(2):103 − 107. [LUO Yiyong,ZHANG Hao,ZHANG Liting. Estimation of slope safety factor based on evolutionary relevance vector machine[J]. Yellow River,2016,38(2):103 − 107. (in Chinese with English abstract) doi: 10.3969/j.issn.1000-1379.2016.02.027
[10]	江婷,沈振中,徐力群,等. 基于支持向量机-小波神经网络的边坡位移时序预测模型[J]. 武汉大学学报(工学版),2017,50(2):174 − 181. [JIANG Ting,SHEN Zhenzhong,XU Liqun,et al. Time series prediction model of slope displacement based on support vector machines-wavelet nerual network[J]. Engineering Journal of Wuhan University,2017,50(2):174 − 181. (in Chinese with English abstract)
[11]	TIPPING M E. Escaping the convex hull with extrapolated vector machines[M]//Advances in Neural Information Processing Systems 14. MIT: The MIT Press, 2002: .
[12]	沈力华,陈吉红,曾志刚,等. 多稀疏回声状态网络预测模型[J]. 控制理论与应用,2018,35(4):421 − 428. [SHEN Lihua,CHEN Jihong,ZENG Zhigang,et al. Prediction model with multiple sparse echo state network[J]. Control Theory & Applications,2018,35(4):421 − 428. (in Chinese with English abstract) doi: 10.7641/CTA.2017.70315
[13]	KALTWANG S,TODOROVIC S,PANTIC M. Doubly sparse relevance vector machine for continuous facial behavior estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(9):1748 − 1761. doi: 10.1109/TPAMI.2015.2501824
[14]	LUO Z Y,BUI X N,NGUYEN H,et al. A novel artificial intelligence technique for analyzing slope stability using PSO-CA model[J]. Engineering with Computers,2021,37(1):533 − 544. doi: 10.1007/s00366-019-00839-5
[15]	李胜,韩永亮. 基于MFOA-SVR露天矿边坡变形量预测研究[J]. 中国安全生产科学技术,2015,11(1):11 − 16. [LI Sheng,HAN Yongliang. Research on forecasting of slope deformation in open-pit mine based on MFOA-SVR[J]. Journal of Safety Science and Technology,2015,11(1):11 − 16. (in Chinese with English abstract)