基于SBKF-PNN融合的高填方渠道渗漏监测模型研究
|
摘要
针对目前高填方渠道渗漏检测方法通常单一、数据获取易受环境干扰、渗漏等级难以分类等问题,研究了基于SBKF-PNN融合的高填方渠道渗漏实时监测模型.首先建立基于土质高填方渠段的实验模型,设计了基于ZigBee和GPRS的渗漏信息无线传感网络,将高填方渠道的温度信息、湿度信息、GPS信息和渗流信息进行可移动获取;结合高填方渠道渗漏规律,分析传感器多源数据变化的规律及其关联度,定义了高填方渠道渗漏的等级模式,筛选了温度场、电势场和电磁场等多传感器信息作为渗漏监测量;然后应用贯序式块卡尔曼滤波(Sequential Block Kalman Filter,SBKF)方法对多传感器数据块进行处理,同时采用概率神经网络(Probabilistic Neural Network,PNN)算法进行渠道渗漏的等级分类;最后用大量的实测数据对SBKF-PNN模型进行训练,得到高填方渠道渗漏监测的反演模型,并将该反演模型应用到实际的高填方渗漏监测中.结果表明,基于SBKF-PNN的渗漏监测模型可实现多传感数据块的实时滤波,有效融合多种环境量的突变特征,能较准确地实现高填方渠道渗漏等级分类.
Aiming at the problem that the detection method of high-fill channel leakage is usually single,the data acquisition is easily disturbed by environment and the leakage level is difficult to classify,the real-time Fusion model,based on SBKF-PNN for fill channel leakage monitoring,has been studied.First,an experimental model based on high-fill soil channel was established,and a leakage information wireless sensor network based on ZigBee and GPRS was designed.The information of high-fill channel,such as temperature information,humidity information,GPS information and seepage information,were obtained movably.Combined with the leakage rule of high fill channel,the trend of multi-sensor data changes and their relevance were analyzed.The grade pattern of high fill channel leakage was defined.The multisensor information,such as temperature field,potential field and electromagnetic field,were selected as the leakage information.Then,the Sequential Block Kalman Filter(SBKF) method was applied to process the multisensor data blocks,and the Probabilistic Neural Network(PNN) algorithm was used to classify the channel leakage.Finally,the SBKF-PNN model was trained with a large amount of measured data,and the inversion model of channel leakage monitoring was obtained.The inversion model was applied to actual high-fill leakage monitoring.The results show that leakage monitoring model,which based on SBKF-PNN,can realize real-time filtering of multi-sensing data blocks.The SBKF-PNN model can fuse the characteristics of multiple environmental information and accurately classify the leakage of high-filled channels.
Aiming at the problem that the detection method of high-fill channel leakage is usually single,the data acquisition is easily disturbed by environment and the leakage level is difficult to classify,the real-time Fusion model,based on SBKF-PNN for fill channel leakage monitoring,has been studied.First,an experimental model based on high-fill soil channel was established,and a leakage information wireless sensor network based on ZigBee and GPRS was designed.The information of high-fill channel,such as temperature information,humidity information,GPS information and seepage information,were obtained movably.Combined with the leakage rule of high fill channel,the trend of multi-sensor data changes and their relevance were analyzed.The grade pattern of high fill channel leakage was defined.The multisensor information,such as temperature field,potential field and electromagnetic field,were selected as the leakage information.Then,the Sequential Block Kalman Filter(SBKF) method was applied to process the multisensor data blocks,and the Probabilistic Neural Network(PNN) algorithm was used to classify the channel leakage.Finally,the SBKF-PNN model was trained with a large amount of measured data,and the inversion model of channel leakage monitoring was obtained.The inversion model was applied to actual high-fill leakage monitoring.The results show that leakage monitoring model,which based on SBKF-PNN,can realize real-time filtering of multi-sensing data blocks.The SBKF-PNN model can fuse the characteristics of multiple environmental information and accurately classify the leakage of high-filled channels.
引文
[1] 崔岗,陈俊生,王丽丽.南水北调高填方段渗流监测设计方案[J].西部探矿工程,2013,25(2):39-41 Cui Gang,Chen Junsheng,Wang Lili.Design scheme of seepage monitoring in high fill segment of South-to-North Water Transfer Project[J].Western Exploration Engineering,2013,25(2):39-41
[2] 屈志刚,申黎平,李明新,等.南水北调中线工程高填方渠道加强措施探讨[J].人民长江,2013,44(16):63～66 Qu Zhigang,Shen Liping,Li Mingxin,et al.Effective reinforcement measures for high-filled canal of Middle Route Project of South-to-North Water Diversion[J].Yangtze River,2013,44(16):63-66
[3] 汪易森.南水北调中线工程几个技术问题的解决与思考[J].水利水电技术,2015,46(6):79-86 Wang Yisen.Consideration and solution of several technical problems of the Middle Route of the South-to-North Water Diversion Project[J].Water Resources and Hydropower Engineering,2015,46(6):79-86
[4] 付新航.南水北调中线干线工程高填方渠段渗流分析[D].郑州:华北水利水电大学,2014 Fu Xinhang.Seepage analysis of South-to-North Water transfer midline high fill canal segment[D].Zhongzhou:North China University of Water Conservancy and Electric Power,2014
[5] 蔡运胜,张宝华.几种电法仪器在地质勘查中的应用[J].地质与勘探,2006,42(5):72-78 Cai Yunsheng,Zhang Baohua.The application of a few electrical method instruments in geologic prospecting[J].Geology and Prospecting,2006,42(5):72-78
[6] 胡雄武,张平松,江晓益.并行电法在快速检测水坝渗漏通道中的应用[J].水利水电技术,2012,43(11):51-54 Hu Xiongwu,Zhang Pingsong,Jiang Xiaoyi.Application of parallel electric survey to quick detection of seepage passage through reservoir dam[J].Water Resource and Hydropower Engineering,2012,43(11):51-54
[7] 马若龙,毋光荣,周锡芳.高密度电法和自然电位法在某水库大坝渗漏探测中的应用[J].大坝与安全,2015,(6):55-58 Ma Ruolong,Wu Guangrong,Zhou Xifang.Application of high-density resistivity method and spontaneous electric field method in leakage detection of a dam[J].Dam and Safety,2015,(6):55-58
[8] Sj?dahl P,Dahlin T,Johansson S.Using the resistivity method for leakage detection in a blind test at the R?ssvatn embankment dam test facility in Norway[J].Bulletin of Engineering Geology & the Environment,2010,69(4):643-658
[9] 蒋力,周柏兵,徐国龙,等.基于分布式光纤技术的渗流监测试验探论[J].大坝与安全,2015,(5):32-36 Jiang Li,Zhou Baibing,Xu Guolong,et al.Research of seepage monitoring test based on distributed optical fiber temperature sensing technology [J].Dam and Safety,2015,(5):32-36
[10] Marc Nikles,Bernhard Vogel,Fabien Briffod,et al.Leakage detection using fiber optics distributed temperature rise monitoring [C].The Proceedings of the 11th SPIEAnnual International Symposium on Smart Structures and Materials.San Diego,California:ASCE Publications,2004:18-25
[11] Rath V,Mottaghy D.Smooth inversion for ground surface temperature histories:estimating the optimum regularization parameter by generalized cross-validation[J].Geophysical Journal International,2007,171(3):1440-1448
[12] 付长静,李国英,陈亮,等.利用温度场计算渗透流速的数学模型[J].水利水运工程学报,2015,(6):88-93 Fu Changjing,Li Guoying,Chen Liang,et al. A mathematical model for calculating penetration velocity using temperature field[J]. Hydro-Science and Engineering,2015,(6):88-93
[13] 唐智德,王绍旭,文春龙.青狮潭水库大坝渗漏观测分析及评价[J].水利规划与设计,2016,(11):107-111 Tang Zhide,Wang Shaoxu,Wen Chunlong.Analysis and evaluation of dam leakage in Qingshitan Reservoir[J].Water Resources Planning and Design,2016,(11):107-111
[14] 张茜,陈建生,董海洲,等.示踪法测定井中渗透流速的广义稀释模型研究[J].长江科学院院报,2016,33(10):126-130 Zhang Xi,Chen Jiansheng,Dong Haizhou,et al.Modified generalized dilution model of determining permeability velocity in wells by tracer method[J].Journal of Yangtze River Scientific Research Institute,2016,33(10):126-130
[15] Kerstin Müller,Vanderborght J,Englert A,et al.Imaging and characterization of solute transport during two tracer tests in a shallow aquifer using electrical resistivity tomography and multilevel groundwater samplers[J].Water Resources Research,2010,46(3):1-23
[16] Boleve A,Janod F,Revil A,et al.Localization and quantification of leakages in dams using time-lapse self-potential measurements associated with salt tracer injection[J].Journal of Hydrology,2011,403(3):242-252
[17] 赵二峰,顾冲时,苏怀智.大坝安全监测效应量的信息融合估计降阶模型[C].全国大坝安全监测技术信息网全网大会暨学术交流研讨会.全国大坝安全监测技术信息网,昆明,2010 Zhao Erfeng,Gu Chongshi,Su Huaizhi.Information fusion estimation reduced order model of dam safety monitoring effect[C].China Hydropower Engineering Society Dam Safety Monitoring Special Committee Annual Conference and Academic Exchange Conference Proceedings,National Dam Safety Monitoring Technical Information Network,Kunming,2010
[18] Khaleghi B,Khamis A,Karray F O,et al.Multisensor data fusion:A review of the state-of-the-art[J].Information Fusion,2013,14(1):28-44
[19] Zhou C B,Liu W,Chen Y F,et al.Inverse modeling of leakage through a rockfill dam foundation during its construction stage using transient flow model,neural network and genetic algorithm[J].Engineering Geology,2015,187:183-195
[20] Shivashankarappa N,Adiga S,Avinash R A,et al.Kalman filter based multiple sensor data fusion in systems with time delayed state[C].Signal Processing and Integrated Networks(SPIN),2016 3rd International Conference on.IEEE,Noida,India,2016:375-382
[21] 刘明堂.基于多源多尺度数据融合的黄河含沙量检测模型研究[D].郑州:郑州大学,2015 Liu Mingtang.The study of detection model for sediment concentration in the Yellow River based on multi-source and multi-scale[D].Zhengzhou:Zhengzhou University,2015
[22] 章涛,吴仁彪,李月敏.单传感器多尺度状态融合估计算法[J].信号处理,2013,29(8):971-976 Zhang Tao,Wu Renbiao,Li Yuemin.Single sensor multiscale state fusion estimation algorithm[J].Signal Processing,2013,29(8):971-976
[23] 刘明堂,田壮壮,齐慧勤,等.基于Kalman-BP协同融合模型的含沙量测量[J].应用基础与工程科学学报,2016,24(5):970-977 Liu Mingtang,Tian Zhuangzhuang,Qi Huiqin,et al.Cooperative fusion model based on Kalman-BP neural network for suspended sediment concentration measurement[J].Journal of Basic Scienceand Engineering,2016,24(5):970-977
[24] 全学海,丁宣浩,蒋英春.基于EMD和概率神经网络的说话人识别[J].桂林电子科技大学学报,2012,30(2):108-112 Quan Xuehai,Ding Xuanhao,Jiang Yingchun.Speaker recognition based on EMD and probabilistic neural networks[J].Journal of Guilin University of Electronic Technology,2012,30(2):108-112
[25] Chen Xianyue,Zhou Jianzhong,Xiao Han.Fault diagnosis based on comprehensive geometric characteristic and probability neural network[J].Applied Mathematics and Computation,2014,230(3):542-554
[26] 张淑清,徐剑涛,姜安琦,等.基于极点对称模态分解和概率神经网络的轴承故障诊断[J].中国机械工程,2017,28(4):425-431 Zhang Shuqing,Xu Jiantao,Jiang Anqi,et al.Fault diagnosis of bearings based on extreme-point symmetric mode decomposition and probabilistic neural network[J].China Mechanical Engineering,2017,28(4):425-431
[2] 屈志刚,申黎平,李明新,等.南水北调中线工程高填方渠道加强措施探讨[J].人民长江,2013,44(16):63～66 Qu Zhigang,Shen Liping,Li Mingxin,et al.Effective reinforcement measures for high-filled canal of Middle Route Project of South-to-North Water Diversion[J].Yangtze River,2013,44(16):63-66
[3] 汪易森.南水北调中线工程几个技术问题的解决与思考[J].水利水电技术,2015,46(6):79-86 Wang Yisen.Consideration and solution of several technical problems of the Middle Route of the South-to-North Water Diversion Project[J].Water Resources and Hydropower Engineering,2015,46(6):79-86
[4] 付新航.南水北调中线干线工程高填方渠段渗流分析[D].郑州:华北水利水电大学,2014 Fu Xinhang.Seepage analysis of South-to-North Water transfer midline high fill canal segment[D].Zhongzhou:North China University of Water Conservancy and Electric Power,2014
[5] 蔡运胜,张宝华.几种电法仪器在地质勘查中的应用[J].地质与勘探,2006,42(5):72-78 Cai Yunsheng,Zhang Baohua.The application of a few electrical method instruments in geologic prospecting[J].Geology and Prospecting,2006,42(5):72-78
[6] 胡雄武,张平松,江晓益.并行电法在快速检测水坝渗漏通道中的应用[J].水利水电技术,2012,43(11):51-54 Hu Xiongwu,Zhang Pingsong,Jiang Xiaoyi.Application of parallel electric survey to quick detection of seepage passage through reservoir dam[J].Water Resource and Hydropower Engineering,2012,43(11):51-54
[7] 马若龙,毋光荣,周锡芳.高密度电法和自然电位法在某水库大坝渗漏探测中的应用[J].大坝与安全,2015,(6):55-58 Ma Ruolong,Wu Guangrong,Zhou Xifang.Application of high-density resistivity method and spontaneous electric field method in leakage detection of a dam[J].Dam and Safety,2015,(6):55-58
[8] Sj?dahl P,Dahlin T,Johansson S.Using the resistivity method for leakage detection in a blind test at the R?ssvatn embankment dam test facility in Norway[J].Bulletin of Engineering Geology & the Environment,2010,69(4):643-658
[9] 蒋力,周柏兵,徐国龙,等.基于分布式光纤技术的渗流监测试验探论[J].大坝与安全,2015,(5):32-36 Jiang Li,Zhou Baibing,Xu Guolong,et al.Research of seepage monitoring test based on distributed optical fiber temperature sensing technology [J].Dam and Safety,2015,(5):32-36
[10] Marc Nikles,Bernhard Vogel,Fabien Briffod,et al.Leakage detection using fiber optics distributed temperature rise monitoring [C].The Proceedings of the 11th SPIEAnnual International Symposium on Smart Structures and Materials.San Diego,California:ASCE Publications,2004:18-25
[11] Rath V,Mottaghy D.Smooth inversion for ground surface temperature histories:estimating the optimum regularization parameter by generalized cross-validation[J].Geophysical Journal International,2007,171(3):1440-1448
[12] 付长静,李国英,陈亮,等.利用温度场计算渗透流速的数学模型[J].水利水运工程学报,2015,(6):88-93 Fu Changjing,Li Guoying,Chen Liang,et al. A mathematical model for calculating penetration velocity using temperature field[J]. Hydro-Science and Engineering,2015,(6):88-93
[13] 唐智德,王绍旭,文春龙.青狮潭水库大坝渗漏观测分析及评价[J].水利规划与设计,2016,(11):107-111 Tang Zhide,Wang Shaoxu,Wen Chunlong.Analysis and evaluation of dam leakage in Qingshitan Reservoir[J].Water Resources Planning and Design,2016,(11):107-111
[14] 张茜,陈建生,董海洲,等.示踪法测定井中渗透流速的广义稀释模型研究[J].长江科学院院报,2016,33(10):126-130 Zhang Xi,Chen Jiansheng,Dong Haizhou,et al.Modified generalized dilution model of determining permeability velocity in wells by tracer method[J].Journal of Yangtze River Scientific Research Institute,2016,33(10):126-130
[15] Kerstin Müller,Vanderborght J,Englert A,et al.Imaging and characterization of solute transport during two tracer tests in a shallow aquifer using electrical resistivity tomography and multilevel groundwater samplers[J].Water Resources Research,2010,46(3):1-23
[16] Boleve A,Janod F,Revil A,et al.Localization and quantification of leakages in dams using time-lapse self-potential measurements associated with salt tracer injection[J].Journal of Hydrology,2011,403(3):242-252
[17] 赵二峰,顾冲时,苏怀智.大坝安全监测效应量的信息融合估计降阶模型[C].全国大坝安全监测技术信息网全网大会暨学术交流研讨会.全国大坝安全监测技术信息网,昆明,2010 Zhao Erfeng,Gu Chongshi,Su Huaizhi.Information fusion estimation reduced order model of dam safety monitoring effect[C].China Hydropower Engineering Society Dam Safety Monitoring Special Committee Annual Conference and Academic Exchange Conference Proceedings,National Dam Safety Monitoring Technical Information Network,Kunming,2010
[18] Khaleghi B,Khamis A,Karray F O,et al.Multisensor data fusion:A review of the state-of-the-art[J].Information Fusion,2013,14(1):28-44
[19] Zhou C B,Liu W,Chen Y F,et al.Inverse modeling of leakage through a rockfill dam foundation during its construction stage using transient flow model,neural network and genetic algorithm[J].Engineering Geology,2015,187:183-195
[20] Shivashankarappa N,Adiga S,Avinash R A,et al.Kalman filter based multiple sensor data fusion in systems with time delayed state[C].Signal Processing and Integrated Networks(SPIN),2016 3rd International Conference on.IEEE,Noida,India,2016:375-382
[21] 刘明堂.基于多源多尺度数据融合的黄河含沙量检测模型研究[D].郑州:郑州大学,2015 Liu Mingtang.The study of detection model for sediment concentration in the Yellow River based on multi-source and multi-scale[D].Zhengzhou:Zhengzhou University,2015
[22] 章涛,吴仁彪,李月敏.单传感器多尺度状态融合估计算法[J].信号处理,2013,29(8):971-976 Zhang Tao,Wu Renbiao,Li Yuemin.Single sensor multiscale state fusion estimation algorithm[J].Signal Processing,2013,29(8):971-976
[23] 刘明堂,田壮壮,齐慧勤,等.基于Kalman-BP协同融合模型的含沙量测量[J].应用基础与工程科学学报,2016,24(5):970-977 Liu Mingtang,Tian Zhuangzhuang,Qi Huiqin,et al.Cooperative fusion model based on Kalman-BP neural network for suspended sediment concentration measurement[J].Journal of Basic Scienceand Engineering,2016,24(5):970-977
[24] 全学海,丁宣浩,蒋英春.基于EMD和概率神经网络的说话人识别[J].桂林电子科技大学学报,2012,30(2):108-112 Quan Xuehai,Ding Xuanhao,Jiang Yingchun.Speaker recognition based on EMD and probabilistic neural networks[J].Journal of Guilin University of Electronic Technology,2012,30(2):108-112
[25] Chen Xianyue,Zhou Jianzhong,Xiao Han.Fault diagnosis based on comprehensive geometric characteristic and probability neural network[J].Applied Mathematics and Computation,2014,230(3):542-554
[26] 张淑清,徐剑涛,姜安琦,等.基于极点对称模态分解和概率神经网络的轴承故障诊断[J].中国机械工程,2017,28(4):425-431 Zhang Shuqing,Xu Jiantao,Jiang Anqi,et al.Fault diagnosis of bearings based on extreme-point symmetric mode decomposition and probabilistic neural network[J].China Mechanical Engineering,2017,28(4):425-431