QQ群聊

• CN 62-1112/TF
• ISSN 1005-2518
• 创刊于1988年

## 基于小波分解的尾矿坝浸润线预测方法研究

1. 中南大学资源与安全工程学院，湖南 长沙 410083

## Study on Prediction Method of Seepage Line of Tailings Dam Based on Wavelet Decomposition

SUI Xiaodan, LUO Zhouquan,, QIN Yaguang, WANG Yule, PENG Dong

1. School of Resources and Safety Engineering，Central South University，Changsha 410083，Hunan，China

 基金资助: 中南大学中央高校基本科研业务费专项资金（编号：502221716）和“十三五”国家重点研发计划课题“深部大矿段多采区时空协同连续采矿理论与技术”（编号2017YFC0602901）联合资助

Received: 2017-08-18   Revised: 2017-11-30   Online: 2019-03-11

Abstract

The seepage line is the safe lifeline of tailings dam, and its location change directly reflects the seepage characteristics inside the dam body.In order to accurately predict the location change of the seepage line of tailings dam in flood season, analyze the law of change,and predict the future change of seepage line,so as to ensure production, reduce accidents such as dam break, and ensure the safety of people’s life and property. Combined with the characteristics of unstable, nonlinear time series and dynamic change of the submerged line of the tailings dam, using wavelet decomposition and reconstruction, the exponential smoothing analysis method of time series based on wavelet decomposition and the analysis method of BP neural network are proposed. Firstly, the non-stationary time series s is decomposed into five detail signal sequences and one approximate signal sequence by wavelet decomposition. The d1~d5 of the detailed signal sequence is predicted by exponential smoothing method of time series. The BP neural network method is used to predict the approximate signal sequence A5. Finally, wavelet reconstruction is used to predict the approximate signal sequence A5 based on MATLAB programming, and the fitting results are superimposed to predict the soakage line of the tailings dam. In this paper, a metal tailings dam with more precipitation and humid climate is selected as the research object, and the data of the buried depth of the infiltration line in the flood season for 100 days are selected for modeling and analysis. The time series exponential smoothing analysis method based on wavelet decomposition and the BP neural network analysis method are used to predict the development trend of the buried depth of the infiltrating line in the next 10 days, and the predicted results are compared with the actual monitoring data.The results show that the prediction results of wavelet decomposition method are compared with those of traditional single exponential smoothing and neural network prediction methods. The prediction accuracy is as follows: wavelet decomposition (0.9287) > exponential smoothing (0.9038) > neural network (0.8725). The prediction error of wavelet decomposition is the smallest and the accuracy is the highest. In terms of fitting degree, wavelet decomposition (0.8837) > exponential smoothing (0.8573) > neural network (0.8462). The fitting result of wave decomposition prediction method is almost consistent with the whole development trend of original time series, and the degree of coincidence is high. Therefore, This method has good applicability and superiority in predicting the buried depth of tailing dam in flood season.

Keywords： tailings dam ; phreatic line ; wavelet decomposition ; exponential smoothing ; BP neural network ; prediction

SUI Xiaodan, LUO Zhouquan, QIN Yaguang, WANG Yule, PENG Dong. Study on Prediction Method of Seepage Line of Tailings Dam Based on Wavelet Decomposition[J]. Gold Science and Technology, 2019, 27(1): 137-143 doi:10.11872/j.issn.1005-2518.2019.01.137

### 图1

Fig.1   General prediction steps

### 图2

Fig.2   Structure diagram of feed-forward neural network

### 图3

Fig.3   Wavelet decomposition prediction steps

### 2.1 预测精度的检验

$Q1=∑i=1n(yi-yi*)2$
$Q2=∑i=1nyi2$

$Rnew=1-Q1Q2$

Rnew拟合度指标结果越接近于1，说明预测值与实测值越接近，拟合优度越高[20]

### 2.2 基于MATLAB的编程实现

MATLAB（矩阵实验室）是一款高性能的数学软件，这里主要使用的是其神经网络和小波分解工具箱。

### 图4

Fig.4   Real map of a tailings

### 3.2 数据的提取

Table 1  Data of 100 days before the penetration depth of tailings dam（m）

113.851113.558113.789113.76
213.921213.648213.719213.67
313.801313.778313.659313.55
413.891413.848413.629413.51
513.801513.758513.559513.48
613.751613.718613.719613.54
713.671713.658713.759713.60
813.541813.598813.809813.69
913.501913.548913.709913.74
1013.432013.659013.6710013.78

### 图5

Fig.5   Results after wavelet db3 decomposition

### 图6

Fig.6   Comparison of prediction results of different methods with measured results

### 图7

Fig.7   Comparison of fitting results of different forecasting methods with original time series

## 4 结论

（1）结合浸润线埋深非稳定、非线性的时间序列以及浸润线高度动态变化的特点，提出基于小波分解的时间序列指数平滑分析方法和BP神经网络分析方法。

（2）通过对比分析，3种预测方法的预测精度为：小波分解$>$指数平滑$>$神经网络。

（3）结合工程实例，将小波分解预测方法的预测结果与传统单一的指数平滑法和神经网络法预测结果进行比较可知，其精确度和拟合度更高。该方法在汛期尾矿坝浸润线埋深预测中具有较好的适用性和优越性。

（4）基于小波分解的预测方法对短期预测的精度较高，但长期预测的精度低，应在以后的研究应寻求能够在长期预测中有较高精度的预测方法。

## 参考文献 原文顺序 文献年度倒序 文中引用次数倒序 被引期刊影响因子

[J].黄金科学技术，2017251）：99-105.

Sun Yang Luo Liming Deng Hongwei .

Stability analysis and parameter optimization of stope in deep metal mines

[J] . Gold Science and Technology2017251）：99-105.

Hu J Y Xie J B Li W et al .

Experimental research on the liquefaction potential of tailings silt in Zhuziqing tailings dam

[J].Applied Mechanics and Materials2012（204/208）：708-713.

[D].广州华南理工大学2013.

Fan Tuo .

Research on Automatic Safety Monitoring System Technology of Mine Tailing Reservior

[D].Guangzhou:South China University of Technology2013.

Gokmen T Dorota S , Andrew W et al .

Case study：Finite element method and artificial neural network models for flow through Jeziorsko earthfill dam in poland

[J].Journal of Hydraulic Engineering20051316）：431-440.

[J]. 金属矿山，20063610）：69-74.

Xie Zhenhua Chen Qing .

RBF neural network method for analyzing monitoring data of tailings dam

[J].Metal Mine20063610）：69-74.

Tawhid M A Ali A F .

A Hybrid grey wolf optimizer and genetic algorithm for minimizing potential energy function

[J].Memetic Computing201794）：1-13.

[J].安全与环境学报，200335）：16-20.

Jiang Weidong Li Xibing Qiu Gengsheng .

A new method for Chaos-controlling of seepage line in tailings dam based on analysis of largest Lyapunov exponent

[J].Journal of Safety and Environment200335）：16-20.

[J].黄金科学技术，2017253）：19-27.

Song Guozheng Yan Chunming Cao Jia et al .

Breakthrough and significance of exploration at depth more than 1000 m in Jiaojia metallogenic belt，Jiaodong：A case of Shaling mining area

[J].Gold Science and Technology2017253）：19-27.

[J].黄金科学技术，2017253）：98-107.

Deng Gao Yang Shan .

Stability evaluation of goafs based on combined forecasting and variable precision rough fuzzy set

[J].Gold Science and Technology2017253）：98-107.

Torrence C Compo G P .

A practical guide to wavelet analysis

[J].Bulletin of the American Meteorological Society19987979）：61-78.

[J].系统工程理论与实践，2002229）：101-106.

He Guoguang Ma Shoufeng Li Yu .

Study on the short-term forecasting for traffic flow based on wavelet analysis

[J].Systems Engineering-Theory & Practice2002229）：101-106.

[J]. 水文地质工程地质，199761）：16-18.

Yuming Men Hu Gaoshe Liu Yuhai .

Exponential smoothing method and its application in landslide prediction

[J].Hydrogeology and Engineering Geology199761）：16-18.

[J].华东交通大学学报，2005223）：8-11.

Ma Xiaoke Wang Ciguang .

The application of the cubic exponential smoothing method on volume forecasting of Da-Qin railway

[J].Journal of East China Jiaotong University2005223）：8-11.

[J].大连交通大学学报，2004251）：79-80.

Zhang Denan Zhang Xinyan .

Ascertainment of index level and coefficient smooth

[J].Journal of Dalian Railway Institute2004251）：79-80.

MATLAB神经网络应用设计

[M].北京科学出版社2000.

Wen Xin .

Application Design of MATLAB Neural Network

[M].BeijingScience Press2000.

[J].黄金科学技术，2016243）：64-69.

Shi Xiuzhi Fan Yuqian Shang Xueyi .

Strength prediction of filling body based on PCA and BP neural networks

[J].Gold Science and Technology2016243）：64-69.

[J].数学的实践与认识，2002325）：813-820.

Hao Yonghong Wang Xuemeng .

The dynamic model of gray system and its application to population forcasting

[J].Mathematics in Practice and Theory2002325）：813-820.

[J].模式识别与人工智能，2005185）：528-534.

Chen Guo .

Analysis of influence factors for forecasting precision of artificial neural network model and its optimizing

[J].Pattern Recognition and Artificial Intelligence2005185）：528-534.

[J].南京师范大学学报（自然科学版），2008313）：39-43.

Fan Yunzheng Lin Lu .

Depth-Weighted LSE for linear regression model and the Fit-Test

[J].Journal of Nanjing Normal University（Natural Science Edition）2008313）：39-43.

[J].黄石理工学院学报，2008246）：49-51.

Wang Sanbao Yi Xiuying .

Accuracy test of the grey prediction model GM（1，1） based on mathematica

[J].Journal of Huangshi Institute of Technology2008246）：49-51.

/

 〈 〉