变形监测数据预报的动态贝叶斯ELM方法

doi:10.11947/j.AGCS.2019.20180504

测绘学报 ›› 2019, Vol. 48 ›› Issue (7): 919-925.doi: 10.11947/j.AGCS.2019.20180504

变形监测数据预报的动态贝叶斯ELM方法

范千¹, 方绪华¹, 许承权², 杨荣华³

1. 福州大学土木工程学院, 福建福州 350116;
2. 闽江学院海洋学院, 福建福州 350108;
3. 重庆大学土木工程学院, 重庆 400045

收稿日期:2018-11-05 修回日期:2019-03-26 出版日期:2019-07-20 发布日期:2019-07-26
作者简介:范千(1981-),男,博士,副教授,研究方向为变形监测数据处理和GNSS精密定位技术。
基金资助:
国家自然科学基金（41404008）；福州市科技计划（2017-G-73）

Dynamic Bayesian ELM method for deformation monitoring data prediction

FAN Qian¹, FANG Xuhua¹, XU Chengquan², YANG Ronghua³

1. College of Civil Engineering, Fuzhou University, Fuzhou 350116, China;
2. Ocean College, Minjiang University, Fuzhou 350108, China;
3. College of Civil Engineering, Chongqing University, Chongqing 400045, China

Received:2018-11-05 Revised:2019-03-26 Online:2019-07-20 Published:2019-07-26
Supported by:
The National Natural Science Foundation of China (No. 41404008);The Science and Technology Program of Fuzhou (No. 2017-G-73)

摘要/Abstract

摘要： 贝叶斯极限学习机（BELM）具有充分利用数据先验信息，可以自适应估计模型参数的特点。但在样本数量不断增加时，如果每次都对BELM重新训练将会降低计算效率。针对此问题，本文提出一种动态贝叶斯极限学习机（DBELM）方法以应用于变形监测数据实时预报。该方法以BELM训练的模型参数为初值，根据新增样本信息可对初始模型参数进行动态更新，并从理论上推导了相关计算公式。通过对仿真数据和实际变形数据进行详细分析表明：DBELM方法的预报精度要优于BELM、正则化极限学习机（RELM）、极限学习机（ELM）3种方法。特别是在长期持续预报过程中，其预报性能相对于其余3种方法优势明显。这充分表明了所提方法应用于变形监测数据预报领域具有可行性和有效性。

关键词: 变形监测, 实时预报, 极限学习机, 动态贝叶斯极限学习机, 预报性能

Abstract: Bayesian extreme learning machine (BELM) has the characteristics of making full use of the prior information of data and self-adaptive estimation of model parameters. However, when the sample size increases, the computational efficiency will be reduced if BELM training is repeated every time. To solve this problem, a dynamic bayesian extreme learning machine (DBELM) method is proposed for real-time prediction of deformation monitoring data. This method takes BELM training model parameters as initial values. According to the new sample information, the initial model parameters can be updated dynamically, and the relevant calculation formula is deduced theoretically. The detailed analysis of simulation data and actual deformation data show that the prediction accuracy of DBELM method is better than that of BELM, RELM and ELM.Especially in the long term continuous forecast, its forecasting performance has obvious advantages over the other three methods.This fully demonstrates the feasibility and validity of the proposed method in the field of deformation monitoring data prediction.

Key words: deformation monitoring, real-time prediction, extreme learning machine, dynamic Bayesian extreme learning machine, forecasting performance

中图分类号:

P258

范千, 方绪华, 许承权, 杨荣华. 变形监测数据预报的动态贝叶斯ELM方法[J]. 测绘学报, 2019, 48(7): 919-925.

FAN Qian, FANG Xuhua, XU Chengquan, YANG Ronghua. Dynamic Bayesian ELM method for deformation monitoring data prediction[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(7): 919-925.

参考文献

[1] SU Huaizhi, LI Xing, YANG Beibei, et al. Wavelet support vector machine-based prediction model of dam deformation[J]. Mechanical Systems and Signal Processing, 2018, 110:412-427.
[2] CHEN Huangqiong, ZENG Zhigang. Deformation prediction of landslide based on improved back-propagation neural network[J]. Cognitive Computation, 2013, 5(1):56-62.
[3] 王新洲, 范千, 许承权, 等. 基于小波变换和支持向量机的大坝变形预测[J]. 武汉大学学报(信息科学版), 2008, 33(5):469-471, 507. WANG Xinzhou, FAN Qian, XU Chengquan, et al. Dam deformation prediction based on wavelet transform and support vector machine[J]. Geomatics and Information Science of Wuhan University, 2008, 33(5):469-471, 507.
[4] LIU Haifeng, REN Chao, ZHENG Zhongtian, et al. Study of a gray genetic BP neural network model infault monitoring and a diagnosis system fordam safety[J]. International Journal of Geo-Information, 2018, 7(4):1-19.
[5] WANG Xiaoyu, KANG Kan, SHEN Changsong. Study on MPGA-BP of gravity dam deformation prediction[J]. Mathematical Problems in Engineering, 2017, 2017:2586107.
[6] 钱程, 李连基, 周子东. 基于GM-ANN组合模型的大坝变形监测数据分析方法[J]. 中国农村水利水电, 2015(10):157-159, 162. QIAN Cheng, LI Lianji, ZHOU Zidong. Dam deformation monitoring data analysis method based on GM-ANN model[J]. China Rural Water and Hydropower, 2015(10):157-159, 162.
[7] 邢尹, 陈闯, 刘立龙, 等. 改进遗传算法和BP神经网络的大坝变形预测[J]. 计算机工程与设计, 2018, 39(8):2628-2631, 2686. XING Yin, CHEN Chuang, LIU Lilong, et al. Dam deformation prediction based on improved genetic algorithm and BP neural network[J]. Computer Engineering and Design, 2018, 39(8):2628-2631, 2686.
[8] 马琳, 马福恒, 范振东, 等. SVM大坝变形监测模型研究[J]. 中国农村水利水电, 2015(5):144-147. MA Lin, MA Fuheng, FAN Zhendong, et al. Dam deformation monitoring model based on improved MMA-SVM[J]. China Rural Water and Hydropower, 2015(5):144-147.
[9] HUANG Guangbin, ZHU Qinyu, SIEW C K. Extreme learning machine:theory and applications[J]. Neurocomputing, 2006, 70(1/3):489-501.
[10] LIAN Cheng, ZENG Zhigang, YAO Wei, et al. Extreme learning machine for the displacement prediction of landslide under rainfall and reservoir level[J]. Stochastic Environmental Research and Risk Assessment, 2014, 28(8):1957-1972.
[11] KANG Fei, LIU Jia, LI Junjie, et al. Concrete dam deformation prediction model for health monitoring based on extreme learning machine[J]. Structural Control and Health Monitoring, 2017, 24(10):e1997.
[12] 廉城, 曾志刚, 苏义鑫, 等. 基于误差修正EOS-ELM的滑坡位移预测[J]. 华中科技大学学报(信息科学版), 2017, 45(9):52-57. LIAN Cheng, ZENG Zhigang, SU Yixin, et al. Landslide displacement prediction based on error correction and ensemble of online sequential extreme learning machine[J]. Journal of Huazhong University of Science and Technology (Nature Science Edition), 2017, 45(9):52-57.
[13] 张海龙, 范振东. 基于改进PSO-ELM算法的混凝土坝变形非线性监控模型[J]. 水电能源科学, 2018, 36(1):82-84, 199. ZHANG Hailong, FAN Zhendong. Nonlinear monitoring model of concrete dam deformation based on improved PSO-ELM algorithm[J]. Water Resources and Power, 2018, 36(1):82-84, 199.
[14] 邓万宇, 郑庆华, 陈琳, 等. 神经网络极速学习方法研究[J]. 计算机学报, 2010, 33(2):279-287. DENG Wanyu, ZHENG Qinghua, CHEN Lin, et al. Research on extreme learning of neural networks[J]. Chinese Journal of Computers, 2010, 33(2):279-287.
[15] HUANG Guangbin, ZHOU Hongming, DING Xiaojian, et al. Extreme learning machine for regression and multiclass classification[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2012, 42(2):513-529.
[16] 魏冠军, 党亚民, 章传银, 等. 顾及不确定性影响的变形概率预报法[J]. 测绘学报, 2017, 46(4):526-532. DOI:10.11947/j.AGCS.2017.20160531. WEI Guanjun, DANG Yamin, ZHANG Chuanyin, et al. Method of deformation probability prediction considering the influence of uncertainty factors[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(4):526-532. DOI:10.11947/j.AGCS.2017.20160531.
[17] SORIA-OLIVAS E, GOMEZ-SANCHIS J, MARTIN J D, et al. BELM:Bayesian extreme learning machine[J]. IEEE Transactions on Neural Networks, 2011, 22(3):505-509.
[18] NING Kefeng, LIU Min, DONG Mingyu. A new robust ELM method based on a Bayesian framework with heavy-tailed distribution and weighted likelihood function[J]. Neurocomputing, 2015, 149:891-903.
[19] 邓兴升, 王新洲. 动态模糊神经网络在大坝变形预报中的应用[J]. 水电自动化与大坝监测, 2007, 31(2):64-67. DENG Xingsheng, WANG Xinzhou. Application of dynamic fuzzy neural network to dam deformation prediction[J]. Hydropower Automation and Dam Monitoring, 2007, 31(2):64-67.
[20] MA Junshui, THEILER J, PERKINS S. Accurate on-line support vector regression[J]. Neural Computation, 2003, 15(11):2683-2703.
[21] LUO Jiahua, VONG C M, WONG P K. Sparse Bayesian extreme learning machine for multi-classification[J]. IEEE Transactions on Neural Networks and Learning Systems, 2014, 25(4):836-843.
[22] 韩敏, 李德才. 基于替代函数及贝叶斯框架的1范数ELM算法[J]. 自动化学报, 2011, 37(11):1344-1350. HAN Min, LI Decai. An norm 1 regularization term ELM algorithm based on surrogate function and Bayesian framework[J]. Acta Automatica Sinica, 2011, 37(11):1344-1350.
[23] CHEN Yarui, YANG Jucheng, WANG Chao, et al. Variational Bayesian extreme learning machine[J]. Neural Computing and Applications, 2016, 27(1):185-196.
[24] BISHOP C. Pattern recognition and machine learning[M]. New York:Springer, 2006.
[25] CHEN Tao, MARTIN E. Bayesian linear regression and variable selection for spectroscopic calibration[J]. Analytica Chimica Acta, 2009, 631(1):13-21.

变形监测数据预报的动态贝叶斯ELM方法

Dynamic Bayesian ELM method for deformation monitoring data prediction

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 11

编辑推荐

Metrics

本文评价

[1]	林珲, 吴贤宇, 潘家祎, 邹海波. 中国城市洪涝实时预报研究:现状与挑战[J]. 测绘学报, 2022, 51(7): 1306-1316.
[2]	黎奇, 白征东, 陈波波, 过静珺, 辛浩浩, 程宇航, 黎琼, 吴斐. GNSS/INS多传感器组合高速铁路轨道测量系统[J]. 测绘学报, 2020, 49(5): 569-579.
[3]	郭迎钢, 李宗春, 何华, 张冠宇, 冯其强, 杨浩. 变形监测网稳定点选取的平方型M_split相似变换法[J]. 测绘学报, 2020, 49(11): 1419-1429.
[4]	王鹤, 曾永年. 城市扩展极限学习机模型[J]. 测绘学报, 2018, 47(12): 1680-1690.
[5]	朱建军, 李志伟, 胡俊. InSAR变形监测方法与研究进展[J]. 测绘学报, 2017, 46(10): 1717-1733.
[6]	肖玉钢, 姜卫平, 陈华, 袁鹏, 席瑞杰. 北斗卫星导航系统的毫米级精度变形监测算法与实现[J]. 测绘学报, 2016, 45(1): 16-21.
[7]	章浙涛朱建军匡翠林周璀. 小波包多阈值去噪法及其在形变分析中的应用[J]. 测绘学报, 2014, 43(1): 13-20.
[8]	朱武,张勤,丁晓利. 多参考点的PS-InSAR变形监测数据处理[J]. 测绘学报, 2012, 41(6): 886-890.
[9]	罗飞雪,,戴吾蛟,唐成盼,黄大伟,伍锡锈. 参考经验模态分解-独立分量分析及其在GPS多路径误差处理中的应用[J]. 测绘学报, 2012, 41(3): 0-416.
[10]	朱建军,谢建. 附不等式约束平差的一种简单迭代算法[J]. 测绘学报, 2011, 40(2): 209-211.
[11]	尹宏杰,,朱建军,李志伟,丁晓利,汪长城. 基于SBAS的矿区形变研究[J]. 测绘学报, 2011, 40(1): 52-58.