基于MISSA-CNN-BiLSTM模型的尾矿坝位移预测

doi:10.16265/j.cnki.issn1003-3033.2024.09.1091

摘要/Abstract

摘要：

为应对尾矿坝位移预测所面临的复杂情况和精度要求,提出一种基于多算法耦合的尾矿坝位移动态预测模型。首先,基于时间序列分解模型将累计位移分为趋势项和周期项,利用高斯回归时间序列预测模型预测趋势项位移;然后,运用不同Copula函数研究诱发因素与周期项位移的整体相关性,鉴于周期项位移影响因素多样性与强非线性的特点,采用多策略融合的改进麻雀搜索算法改进麻雀搜索算法(MISSA)-卷积神经网络(CNN)-双向长短期记忆(BiLSTM)模型预测周期项位移;最后,将高斯回归趋势项位移预测值和MISSA-CNN-BiLSTM周期项位移预测值叠加。结果表明:尾矿坝累积位移预测值与实测值基本一致,预测结果相关性系数R为0.996,均方根误差(RMSE)为0.13 mm,建立的 MISSA-CNN-BiLSTM多算法耦合模型预测精度较高,且能较好地预测尾矿坝位移的阶跃型变化。

关键词: 改进麻雀搜索算法(MISSA), 卷积神经网络(CNN), 双向长短期记忆(BiLSTM), 尾矿坝, 位移预测, 深度学习模型

Abstract:

A comprehensive and sophisticated multi-algorithm coupled dynamic prediction model is proposed to address the intricate reality and stringent accuracy requirements of predicting tailings dam displacement. Firstly, by employing a time series decomposition model, the cumulative displacement is disaggregated into its trend and cyclical components. The trend term displacement is then forecasted using a Gaussian regression time series prediction model. Secondly, various Copula functions are employed to investigate the overall correlation between the inducing factors and the cyclical term displacement. Owing to the diverse influencing factors and strong nonlinearities associated with the cyclical term displacement, the MISSA-CNN-BiLSTM model is utilized for prediction. Lastly, the predicted trend term displacement from the Gaussian regression model and the predicted cyclical term displacement from the MISSA-CNN-BiLSTM model are merged. The results demonstrate a high degree of consistency between the predicted cumulative landslide displacements and the measured values, with a correlation coefficient of 0.996 and a root mean square error (RMSE) of 0.13 mm. The multi-algorithm coupled model, based on MISSA-CNN-BiLSTM, exhibits remarkable prediction accuracy and effectively captures step changes in tailings dam displacements.

Key words: multi strategy improved sparrow search algorithm(MISSA), convolutional neural networks(CNN), Bi-directional long short-term memory(BiLSTM), tailing dam, displacement prediction, deep learning model

中图分类号:

X936

刘迪, 杨辉, 卢才武, 阮顺领, 江松. 基于MISSA-CNN-BiLSTM模型的尾矿坝位移预测[J]. 中国安全科学学报, 2024, 34(9): 145-154.

LIU Di, YANG Hui, LU Caiwu, RUAN Shunling, JIANG Song. Prediction of displacement of tailings dams based on MISSA-CNN-BiLSTM model[J]. China Safety Science Journal, 2024, 34(9): 145-154.

图/表 19

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

表1

表2

图12

表3

图13

表4

图14

表5

参考文献 20

[1]	LIU Zhongqiang, GUO Dong, LACASSE S, et al. Algorithms for intelligent prediction of landslide displacements[J]. Journal of Zhejiang University-Science A: Applied Physics & Engineering, 2020, 21(6): 412-429.
[2]	ZHAI Qian, RAHARDJO H, SATYANAGA A, et al. Effect of the uncertainty in soil-water characteristic curve on theestimated shear strength of unsaturated soil[J]. Journal of Zhejiang University-Science A:Applied Physics & Engineering, 2020, 21(4): 317-330.
[3]	冯文凯, 易小宇, 孟睿, 等. 三峡库区木鱼包滑坡不同库水升降速率变形响应离心模型试验研究[J]. 水利学报, 2021, 52(5): 578-588.
	FENG Wenkai, Yl Xiaoyu, MENG Rui, et al. Study on deformation response of Muyubao landslide in Three Gorges region under differentwater fluctuation rates by centrifugal model test[J]. Journal of Hydraulic Engineering, 2021, 52(5): 578-588.
[4]	刘迪, 卢才武, 连民杰, 等. 基于贝叶斯决策的尾矿坝干滩长度预警研究[J]. 中国安全科学学报, 2022, 32(增1): 120-126.
	LIU Di, LU Caiwu, LIAN Minjie, et al. Dry beach length pre-warning of tailings dam based on Bayesian decision[J]. China Safety Science Journal, 2022, 32(S1): 120-126. doi: 10.16265/j.cnki.issn1003-3033.2022.S1.0650
[5]	王晨辉, 赵贻玖, 郭伟, 等. 滑坡位移EEMD-SVR预测模型[J]. 测绘学报, 2022, 51(10): 2196-2204. doi: 10.11947/j.AGCS.2022.20220291
	WANG Chenhui, ZHAO Yijiu, GUO Wei, et al. Displacement prediction model of landslide based on ensemble empiricalmode decomposition and support vector regression[J]. Journal of Geodesy and Geoinformation Science, 2022, 51(10): 2196-2204.
[6]	高雅萍, 陈曦, 涂锐. 顾及降雨影响的动态优化时滞时序GM(1,2)模型在滑坡位移预测中的应用[J]. 测绘学报, 2022, 51(10): 2183-2195. doi: 10.11947/j.AGCS.2022.20220290
	GAO Yaping, CHEN Xi, TU Rui. Application of dynamic optimization time delay GM(1,2) model in landslidedisplacement prediction considering the influence of rainfall[J]. Journal of Geodesy and Geoinformation Science, 2022, 51(10): 2183-2195.
[7]	张振坤, 张冬梅, 李江, 等. 基于多头自注意力机制的LSTM-MH-SA滑坡位移预测模型研究[J]. 岩土力学, 2022, 43(增2): 477-486,507.
	ZHANG Zhenkun, ZHANG Dongmei, LI Jiang, et al. LSTM-MH-SA landslide displacement prediction model based on multi-head sef-attention mechanism[J]. Rock and Soil Mechanies, 2022, 43(S2): 477-486,507.
[8]	唐菲菲, 唐天俊, 朱洪洲, 等. 结合注意力机制和Bi-LSTM的降雨型滑坡位移预测[J]. 测绘通报, 2022(9):74-79. doi: 10.13474/j.cnki.11-2246.2022.0267
	TANG Feifei, TANG Tianjun, ZHU Hongzhou, et al. Rainfalllandslide deformation prediction based on attention mechanismand Bi-LSTM[J]. Bulletin of Surveying and Mapping, 2022(9): 74-79. doi: 10.13474/j.cnki.11-2246.2022.0267
[9]	李雅丽, 王淑琴, 陈倩茹, 等. 若干新型群智能优化算法的对比研究[J]. 计算机工程与应用, 2020, 56(22): 1-12. doi: 10.3778/j.issn.1002-8331.2006-0291
	LI Yali, WANG Shuqin, CHEN Qianru, et al. Comparative study of several new swarm intelligence optimization algorithms[J]. Computer Engineering and Applications, 2020, 56(22):1-12. doi: 10.3778/j.issn.1002-8331.2006-0291
[10]	TANG Andi, ZHOU Huan, HAN Tong, et al. A chaos sparrow search algorithm with logarithmic spiral and adaptive step for engineering problems[J]. Computer Modeling in Engineering & Sciences 2022, 130(1):331-364.
[11]	OUYANG Chengtia, QIU Yaxian, ZHU Dongli. Adaptive spiral flying sparrow search algorithm[J]. Scientific Programming, 2021, 7:DOI:10.1155/2021/6505253.
[12]	毛清华, 张强. 融合柯西变异和反向学习的改进麻雀算法[J]. 计算机科学与探索, 2021, 15(6): 1155-1164. doi: 10.3778/j.issn.1673-9418.2010032
	MAO Qinghua, ZHANG Qiang. Improved sparrow algorithm combining cauchy mutation and opposition-basedlearning[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(6): 1155-1164.
[13]	张琳, 汪廷华, 周慧颖. 一种多策略改进的麻雀搜索算法[J]. 计算机工程与应用, 2022, 58(11): 133-140. doi: 10.3778/j.issn.1002-8331.2112-0427
	ZHANG Lin, WANG Tinghua, ZHOU Huiying. Multi-strategy improved sparrow search algorithm[J]. Computer Engineering and Applications, 2022, 58(11): 133-140. doi: 10.3778/j.issn.1002-8331.2112-0427
[14]	戴健非, 杨鹏, 诸利一, 等. 集成PCA和LSTM神经网络的浸润线预测方法[J]. 中国安全科学学报, 2020, 30(3): 94-101. doi: 10.16265/j.cnki.issn1003-3033.2020.03.015
	DAI Jianfei, YANG Peng, ZHU Liyi, et al. A PCA-LSTM neural network-integrated method for phreatic line prediction[J]. China Safety Science Journal, 2020, 30(3): 94-101. doi: 10.16265/j.cnki.issn1003-3033.2020.03.015
[15]	MA Junwei, XIA Ding, GUO Haixiang, et al. Metaheuristic-based support vector regression for landslide displacement prediction: a comparative study[J]. Landslides, 2022, 19(10): 2489-2511.
[16]	武立功, 肖利兴, 刘晓峰, 等. 尾砂粒径对尾矿坝漫顶溃坝的影响[J]. 中国安全科学学报, 2020, 30(4): 160-165. doi: 10.16265/j.cnki.issn1003-3033.2020.04.025
	WU Ligong, XIAO Lixing, LlU Xiaofeng, et al. Influence of particle size on overtopping dam break of tailings dam[J]. China Safety Science Journal, 2020, 30(4): 160-165. doi: 10.16265/j.cnki.issn1003-3033.2020.04.025
[17]	ISLAM S, WILLIAMS D J, LLANO M, et al. Settling, consolidation and shear strength behaviour of coal tailings slurry[J]. International Journal of Mining Science and Technology, 2020, 30(6): 108-116.
[18]	WANG Yixuan, CHAI Junrui, CAO Jing, et al. Effects of seepage on a three-layered slope and its stability analysis under rainfall conditions[J]. Natural Hazards, 2020, 102: 1269-1278.
[19]	刘迪, 卢才武, 连民杰, 等. 基于粒径效应影响的尾矿毛细特性试验[J]. 中国有色金属学报, 2020, 30(11): 2746-2757.
	LIU Di, LU Caiwu, LIAN Minjie, et al. Experiment on tailings capillary characteristies based on particle size effect[J]. Transactions of Nonferrous Metals Society of China, 2020, 30(11): 2746-2757.
[20]	罗小峰, 侯运炳, 闫浩东, 等. 干湿循环作用下全尾砂固结体的损伤机理[J]. 中国有色金属学报, 2021, 31(12): 3730-3739.
	LUO Xiaofeng, HOU Yunbing, YAN Haodong, et al. Damage mechanism of total tailings consolidation underaction of dry-wet cycle[J]. Transactions of Nonferrous Metals Society of China, 2021, 31(12): 3730-3739.

Copula 函数	当月库水位	上月库水位	当月降雨量	上月降雨量	上月位移增量	当月位移增量
Gaussian	0.02	0.10	0.04	0.04	0.02	0.03
Student's t	0.04	0.07	0.21	0.08	0.04	0.08
Gumbel	0.01	0.02	0.03	0.20	0.18	0.01
Clayton	0.09	0.03	0.10	0.12	0.05	0.12
Frank	0.02	0.01	0.04	0.03	0.01	0.03

日期	实际值	MISSA- BiLSTM		MISSA-CNN- BiLSTM		MISSA-SVR
日期	实际值	预测值	误差值	预测值	误差值	预测值	误差值
2020- 07	2.40	2.23	0.17	2.21	0.19	2.77	0.38
2020- 08	3.37	3.05	0.32	3.30	0.07	3.37	0.00
2020- 09	3.85	3.65	0.20	3.81	0.04	4.05	0.20
2020- 10	3.74	3.62	0.12	3.75	0.01	3.67	0.07
2020- 11	3.51	3.42	0.09	3.59	0.08	3.03	0.48
2020- 12	3.16	3.15	0.01	3.15	0.01	2.31	0.85
2021- 01	2.74	2.64	0.10	2.69	0.05	2.00	0.74
2021- 02	2.32	2.32	0.00	2.27	0.05	1.38	0.94
2021- 03	1.93	1.72	0.21	1.88	0.05	1.29	0.64
2021- 04	1.72	1.38	0.34	1.58	0.14	1.07	0.65
2021- 05	1.34	1.31	0.03	1.36	0.02	0.98	0.36
2021- 06	1.32	1.36	0.04	1.50	0.18	1.94	0.62
平均值	—	—	0.14	—	0.07	—	0.49
RMSE	—	—	0.18	—	0.09	—	0.57
R	—	—	0.984	—	0.995	—	0.874

日期	实际值	预测值	误差值
2020-07	35.43	35.29	0.14
2020-08	36.93	36.72	0.21
2020-09	37.92	37.89	0.03
2020-10	38.29	38.42	0.14
2020-11	38.53	38.79	0.25
2020-12	38.66	38.76	0.10
2021-01	38.71	38.71	0.00
2021-02	38.77	38.85	0.08
2021-03	38.85	38.92	0.07
2021-04	39.12	39.00	0.13
2021-05	39.23	39.27	0.04
2021-06	39.70	39.80	0.10