CN110110391A - A kind of flood estimation method in the scarce measuring runoff data basin based on region division - Google Patents

A kind of flood estimation method in the scarce measuring runoff data basin based on region division Download PDF

Info

Publication number
CN110110391A
CN110110391A CN201910305713.1A CN201910305713A CN110110391A CN 110110391 A CN110110391 A CN 110110391A CN 201910305713 A CN201910305713 A CN 201910305713A CN 110110391 A CN110110391 A CN 110110391A
Authority
CN
China
Prior art keywords
basin
flood
region
cluster
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910305713.1A
Other languages
Chinese (zh)
Inventor
杨汉波
杨文聪
杨大文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Original Assignee
Tsinghua University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tsinghua University filed Critical Tsinghua University
Priority to CN201910305713.1A priority Critical patent/CN110110391A/en
Publication of CN110110391A publication Critical patent/CN110110391A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/20Design optimisation, verification or simulation
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02ATECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A10/00TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
    • Y02A10/40Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application is provided in a kind of scarce measuring runoff data river basin flood evaluation method based on region division, including step A, the predetermined region of acquisition, historical flood event data of multiple first basins within first time;B, the basin complex network in predetermined region is constructed;C, using cluster recognizer, region division is carried out to the basin in predetermined region;D, flood estimation model in region is established;E, using flood estimation model in region, flood estimation is designed to the second basin in the presumptive area.The present invention is sorted out the identical basin of flood season property by the concurrent feature of flood, ensure that river basin flood feature homogeney in region, so that regional flood model is more reliable;Using cluster recognizer, it is not necessary that basin classification quantity is manually set, and classification results can filter out the small classification of basin quantity (such as containing only the classification in 1 basin), ensure that the homogeney in basin in big classification.

Description

A kind of flood estimation method in the scarce measuring runoff data basin based on region division
Technical field
The invention belongs to meteorological model fields, especially flood estimation.
Background technique
Flood is one of most important natural calamity in the whole world, and the flood control of hydraulic engineering downstream is required with hydraulic structure flood control Design flood standard.The key problem of design flood is that the flood frequency in basin calculates, i.e., certain return period (such as a-hundred-year) Corresponding flood magnitude, wherein basin is defined as the collecting area that watershed is surrounded.However, in the (definition of scarce data basin For the insufficient basin of runoff observational data, but there is other information, such as drainage area, elevation, drought index, precipitation), if Meter flood can only be estimated using the information of the observation data of neighbouring River Basin Hydrology website, then develop regional flood in engineering Frequency analysis method, wherein region is the set in multiple basins.The analytic approach is classified by watershed first carries out region division, protects It is similar to demonstrate,prove the same area flooding schedule (such as magnitude, seasonality, incitant);Then one is established with basin gas to each region As (such as precipitation), geographical feature (such as basin dispersed elevation) Calculation of Flood statistical model as input;Finally use the model Flood frequency calculating is carried out to the insufficient basin of data.
Traditional region partitioning method utilizes the geographical feature (such as area, dispersed elevation) in basin, Meteorological Characteristics (such as Nian Ping Equal precipitation), flooding schedule (such as flood season property) construct statistics similarity index, it is poly- using level further according to similarity indices The data mining algorithms such as class, K average cluster carry out basin classification.Clustering algorithm needs to preset classification number, introduces artificial Subjective factor;In addition the similarity relationships between a basin cannot be intuitively presented in clustering algorithm.
Complex network is the network structure connected and composed between node and node by enormous amount, it is a description In system between each element correlation tool, the cluster recognizer in complex network is a kind of point for network structure Class method, and energy automatic identification optimal classification number, suitable for the region division classified based on basin.
Summary of the invention
The present invention constructs the scarce data river basin flood design method based on region division, and this method uses complex web Cluster recognizer in network theory carries out basin classification, so that predetermined region is marked off several big regions, and for each Region design flood appraising model is established in region respectively, and the design flood estimation in data basin is lacked for corresponding region.
The present invention provide it is a kind of for lacking the flood design method in measuring runoff data basin, comprising steps of
A, it obtains in predetermined region, historical flood event data of multiple first basins within first time;
B, the basin complex network in predetermined region is constructed;
C, using cluster recognizer, region division is carried out to the basin in predetermined region;
D, flood estimation model in region is established;
E, using flood estimation model in region, flood estimation is designed to the second basin in the presumptive area;
Wherein, the multiple first basin is to have measuring runoff data, meteorological data, geography within first time The basin of information data.(geographic information data includes drainage area, altitude data.The measuring runoff data are the hydrology Data.)
Further, first time range is preferably no less than 30 years;First basin is the basin with measuring runoff data; Second basin is the basin without measuring runoff data.
Further, the region refers to all regions within the scope of some.The region refers to the territorial scope The set in interior multiple basins with certain same characteristic features.The basin refers to the river gathering ground surrounded by watershed line.
Further, in step, in predetermined region, in the first basin particular time range, each basin is obtained 2N maximum diurnal courses are deep, as flood event;
Wherein, two diurnal courses of the time of origin within 14 days are considered as same field flood event, only take wherein biggish Diurnal courses value;
Further, the time range of the existing hydrological data in the first basin is no less than continuous N, and N is natural number.
Further, step B includes sub-step:
B1, the concurrent ratio of flood between different basins is determined,
B2, by each watershed generalization be a node, if the concurrent ratio of two river basin flood events be greater than threshold value, A line is constructed between the two nodes.
Further, in step C, watershed network carries out cluster division.Define network modularity M be
Wherein lcIt is the number of edges inside cluster c, dcIt is total number of edges that all nodes of cluster c are possessed, ncIt is cluster number, m is Total number of edges of network.The cluster of complex network is divided by maximizing modularity M realization.The effect of division is, in cluster The quantity on the side that node connects from each other is relatively more, and the quantity phase on the side that the node between different clusters connects from each other To less, as shown in Figure 2.The optimization method for maximizing modularity M uses the " cluster_ of R language " igraph " packet Louvain " function is carried out based on modularity greedy optimization algorithm (louvain algorithm).
According to cluster division result using the corresponding basin of each cluster as a region.
Further, step D includes sub-step:
D1, the flood that each basin different reoccurrence is inquired into using Pareto distribution with wide scope;
D2, it is directed to each region, using the flood R for establishing the chance of k mono- using Random Forest modelk(value of k be 2,5, 10,20,50,100) with the area Area (km in basin2), face dispersed elevation Elev (m), face be averaged average annual runoff coefficientWith And the daily precipitation P that k mono- is metk(mm) relationship.
Further, in step E, for the second basin of a certain scarce Streamflow Data, if being located at some area that step C delimited In domain (be defined as basin outlet hydrometric station be located at be partitioned into the outer envelope in all data basins in the region), then according to should The flood model (formula 5) in region, by the Area, Elev in the basin,And Pk(mm) the design flood R in the basin is soughtk
The present invention also provides a kind of river basin flood estimating system based on region division, comprising:
Data acquisition module, for obtaining in predetermined region, history flood of multiple first basins within first time Water event data;
Basin complex network constructing module, for constructing the basin complex network in predetermined region;
Region division module, for carrying out region division to the basin in predetermined region using cluster recognizer;
Flood estimation model building module, for establishing flood estimation model in region;
Flood estimation module, for being carried out to the second basin in the presumptive area using flood estimation model in region Design flood estimation.
The present invention also provides a kind of electronic equipment, including storage medium, processor,
The processor is adapted for carrying out method above-mentioned.
The present invention also provides a kind of storage mediums, are stored with computer program, which is able to carry out above-mentioned Method.
The present invention has the advantages that
1. sorting out the identical basin of flood season property by the concurrent feature of flood, river basin flood in region ensure that Feature homogeney, so that regional flood model is more reliable;
2. cluster recognizer is without being manually set basin classification quantity, and can to filter out basin quantity small for classification results Classification (such as containing only the classification in 1 basin), ensure that the homogeney in basin in big classification.
Detailed description of the invention
Fig. 1 is the schematic diagram that flood provided herein concurrently defines.
Fig. 2 be this application involves complex network and cluster structure block schematic illustration.
Fig. 3 be the invention relates to constellation space distribution schematic diagram.
Fig. 4 be the invention relates to cluster 1 basin website spatial distribution and selected scarce data basin website show It is intended to.
Specific embodiment
Many details are explained in the following description in order to fully understand the application.But the application can be with Much it is different from other way described herein to implement, those skilled in the art can be without prejudice to the application intension the case where Under do similar popularization, therefore the application is not limited by following public specific implementation.
Method provided by the embodiment of the present invention, including step A: obtaining in predetermined region, and multiple first basins are first Historical flood event data in time range;
Wherein, the multiple first basin is to have measuring runoff data, meteorological data, geography within first time The basin of information data.(geographic information data includes drainage area, altitude data.The measuring runoff data are the hydrology Data.)
Further, the region refers to all regions within the scope of some.The region refers to the territorial scope The set in interior multiple basins with certain same characteristic features.The basin refers to the river gathering ground surrounded by watershed line.
Using U.S.'s Model Parameter Estimation Experiment data set as example, selection includes 284 basins of 1971 to 2000 depth of runoff data day by day, the area in selected basin is in 80km2To 10328km2Between.
In step, in predetermined region (U.S.), in the first basin particular time range (1971 to 2000), 2N maximum diurnal courses for obtaining each basin are deep, as flood event;
Wherein, two diurnal courses of the time of origin within 14 days are considered as same field flood event, only take wherein biggish Diurnal courses value;
Further, the existing hydrology in the first basin, meteorological data time range be no less than continuous N, N is nature Number.
Specifically, step A obtains the historical flood event that there is data basin in the U.S.
In step A, there are in the particular time range of data basin (continuous 30 years) the U.S. is existing, obtain each basin 60 A maximum diurnal courses are deep, as flood event.In order to guarantee the independence of extracted flood event, time of origin 14 days with Two interior diurnal courses regard same field flood event as deeply, only take wherein biggish diurnal courses value.
For each basin, 1971 to 2000 60 independent maximum diurnal courses values are extracted.
Step B: construction basin complex network
B1, the concurrent ratio of flood between different basins is determined.
Flood is concurrently defined as: being no more than 7 days respectively from two flood time of origins difference of two basins i and j, is seen Fig. 1.Concurrent ratio is defined as:
Wherein mijIt is two river basin flood number of concurrent.
B2, it determines company side between Node Contraction in Complex Networks, constructs complex network.The present invention is using each basin as one Node, as shown in Fig. 2, if the concurrent ratio c of the flood event in two basinsij≥c95, then to the node for representing the two basins Lian Bian, wherein concurrent threshold value c95It is all95% quantile, i.e., under the threshold value only have 5% basin node pair There is Bian Xianglian.Side is connected to all qualified two nodes, constructs complex network.
Specifically, using basin as node, concurrent ratio between calculating two-by-two, 95% quantile of all concurrent ratios is c95 =0.35;If two concurrent ratios of river basin flood are not less than concurrent threshold value c95, then a line is added to node for this.Specifically, " igraph " packet of R language can be used in the step.
Step C: utilizing cluster recognizer, carries out region division to the basin in the region.
Define network modularity M be
Wherein lcIt is the number of edges inside cluster c, dcIt is total number of edges that all nodes of cluster c are possessed, ncIt is cluster number, m is Total number of edges of network.
The cluster of complex network is divided by maximizing modularity M realization.
The effect of division is that the quantity on the side that the node in cluster connects from each other is relatively more, and between different clusters The quantity on side that connects from each other of node it is relatively fewer, as shown in Figure 2.
The optimization method for maximizing modularity M is carried out using " cluster_louvain " function of R language " igraph " packet Based on modularity greedy optimization algorithm (louvain algorithm).
According to cluster division result using the corresponding basin of each cluster as a region.
Specifically, being used " cluster_louvain " of R language " igraph " packet for the network constructed in step B Function carries out the cluster based on modularity greedy optimization algorithm (louvain algorithm) and identifies, obtains 18 clusters, wherein 12 small-scale clusters (cluster 7 arrives cluster 18) are containing only not more than 6 nodes;6 extensive cluster (cluster 1 arrives cluster 6) institutes It is not less than 22 containing number of nodes, they contain 260 nodes altogether, account for the 91.5% of node total number 284.The same big cluster In node representated by spatially all neighbour, the spatial distribution in their included basins have respectively constituted 6 regions in basin, The spatial distribution in cluster 1 to the basin of cluster 6 is shown in Fig. 3.
Step D: region design flood appraising model is established.
Step D includes sub-step:
D1, design flood is calculated by measured data.Each basin is used in the field the 2N flood event of selected historical period It is fitted Pareto distribution with wide scope:
Wherein R is one day depth of runoff of the basin in participated in the election of historical period, and u is minimum diurnal courses in selected 2N flood Deep, σ and ξ are distribution parameters, and Pr represents probability.
Calculate different reoccurrence flood:
K is the return period, is commonly 2,5,10,20,50,100 in engineering, meets within corresponding 2 years one, meets within 5 years one, 10 years one It meets, 20 years a chances, fifty year return period, once-in-a-century flood.
D2, training Random Forest model.
To all basins in each region, the flood R of the chance of k mono- is established using Random Forest modelk(value of k be 2, 5,10,20,50,100) with the area Area (km in basin2), face dispersed elevation Elev (m), face be averaged average annual runoff coefficient And the daily precipitation P that k mono- is metk(mm) relationship.Model is estimated to obtain the design flood that the return period is k:
Rk=f (Area, Elev, Rcoef,Pk) (5)
Wherein runoff coefficientR is that average annual runoff is deep (mm), and P is that more average annual rainfalls are deep (mm).
Specifically, quasi- using 60 flood samples of " fevd " function in R language " extRemes " packet to each basin Pareto distribution with wide scope is closed, distribution parameter is estimated using linear moment mode.Each basin can be acquired by Pareto distribution with wide scope Observed flood design value obtains 2 years one chances, 5 years chances, 10 years chances, 20 years chances, fifty year return period, a-hundred-year Diurnal courses value.The face of the identical return period daily rain amount value that is averaged also is inquired into same method.
This example pays close attention to 6 extensive clusters: cluster 1 arrives cluster 6.For each cluster return period k=2,5,10,20, 50,100 years floods, using using R language caret packet training Random Forest model formula (5).The accuracy rate of model passes through intersection Verifying is extracted the sample equal with basin quantity using bootstrap (bootstrap) and is trained to assess, and not with remainder The basin of extraction so repeats 25 times, the median of the deterministic coefficient of 25 test sets is as final standard as test set True rate.The accuracy rate in each region is shown in Table 1.It can be seen that R squares of simulation of each cluster each return period is big other than cluster 3 and cluster 4 There is preferable simulation effect in part 0.5 or more.
The accuracy rate (R squares) of each cluster regions design flood appraising model of table 1
Step E: flood estimation is designed using the model of corresponding region to scarce data basin.
For the second basin of a certain scarce Streamflow Data, (it is defined as basin to go out if being located in some region that step C delimited Saliva text erect-position is in the outer envelope for being partitioned into all data basins in the region), then according to the flood model (formula in the region 5), by the Area, Elev in the basin,And Pk(mm) the design flood R in the basin is soughtk
As an example, (this website does not exist a scarce data website 06817500 of the selection in the constituted region of cluster 1 Within 284 websites for constructing complex network), position is as shown in Figure 4.
With the Random Forest model trained in the constituted region of cluster 1 in step D, the variable in the basin is inputted: Area=3211.6km2, Elev (m)=1240m,P2=53.36mm, P5=64.36mm, P10= 73.70mm P20=84.02mm, P50=99.34mm, P100=112.35mm.Acquire design flood are as follows: R2=11.09mm, R5= 14.11mm R10=16.65mm, R20=17.17mm, R50=22.49mm, R100=24.77mm.
The technical principle of the invention is described above in combination with a specific embodiment.These descriptions are intended merely to explain of the invention Principle, and shall not be construed in any way as a limitation of the scope of protection of the invention.Based on the explanation herein, the technology of this field Personnel can associate with other specific embodiments of the invention without creative labor, these modes are fallen within Within protection scope of the present invention.

Claims (9)

1. a kind of river basin flood evaluation method based on region division, comprising steps of
A, it obtains in predetermined region, historical flood event data of multiple first basins within first time;
B, the basin complex network in predetermined region is constructed;
C, using cluster recognizer, region division is carried out to the basin in predetermined region;
D, flood estimation model in region is established;
E, using flood estimation model in region, flood estimation is designed to the second basin in the presumptive area;
Wherein, the multiple first basin is to have measuring runoff data, meteorological data, geography information within first time The basin of data.
2. the method according to claim 1, wherein
In step A, in predetermined region, in the first basin particular time range, the 2N maximum diurnal courses in each basin are obtained It is deep, as flood event;
Wherein, two diurnal courses of the time of origin within 14 days are considered as same field flood event, only take wherein biggish day diameter Flow valuve;
Preferably, the time range of the existing hydrological data in the first basin is no less than continuous N, and N is natural number.
3. the method according to claim 1, wherein
It include sub-step in step B,
B1, the concurrent ratio of flood between different basins is determined,
B2, by each watershed generalization be a node, if the concurrent ratio of two river basin flood events be greater than threshold value, two A line is constructed between a node.
4. the method according to claim 1, wherein
In step C, the modularity M by maximizing drainage network carries out region division,
Wherein lcIt is the number of edges inside cluster c, dcIt is total number of edges that all nodes of cluster c are possessed, ncIt is cluster number, m is network Total number of edges, modularity M be represent cluster divide effect index.
5. the method according to claim 1, wherein
In step D, including sub-step
D1, the flood that each basin different reoccurrence is inquired into using Pareto distribution with wide scope;
D2, it is directed to each region, using the flood R for establishing the chance of k mono- using Random Forest modelk(value of k be 2,5,10, 20,50,100) with the area Area (km in basin2), face dispersed elevation Elev (m), face be averaged average annual runoff coefficientAnd k The daily precipitation P that year one meetsk(mm) relationship.
6. the method according to claim 1, wherein
In step E, for the second basin, if being located in some region that step C delimited, the Forecasting Flood in the region is used Model is averaged average annual runoff coefficient by the area Area in the basin, face dispersed elevation Elev, faceAnd k mono- is met Daily precipitation Pk(mm) the flood R for asking the k mono- in second basin to meetk
7. a kind of river basin flood estimating system based on region division, comprising:
Data acquisition module, for obtaining in predetermined region, historical flood thing of multiple first basins within first time Number of packages evidence;Basin complex network constructing module, for constructing the basin complex network in predetermined region;
Region division module, for carrying out region division to the basin in predetermined region using cluster recognizer;
Flood estimation model building module, for establishing flood estimation model in region;
Flood estimation module, for being designed to the second basin in the presumptive area using flood estimation model in region Flood estimation.
8. a kind of electronic equipment, including storage medium, processor,
The processor is adapted for carrying out method described in claim 1-6.
9. a kind of storage medium, is stored with computer program, which is able to carry out side described in claim 1-6 Method.
CN201910305713.1A 2019-04-16 2019-04-16 A kind of flood estimation method in the scarce measuring runoff data basin based on region division Pending CN110110391A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910305713.1A CN110110391A (en) 2019-04-16 2019-04-16 A kind of flood estimation method in the scarce measuring runoff data basin based on region division

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910305713.1A CN110110391A (en) 2019-04-16 2019-04-16 A kind of flood estimation method in the scarce measuring runoff data basin based on region division

Publications (1)

Publication Number Publication Date
CN110110391A true CN110110391A (en) 2019-08-09

Family

ID=67485529

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910305713.1A Pending CN110110391A (en) 2019-04-16 2019-04-16 A kind of flood estimation method in the scarce measuring runoff data basin based on region division

Country Status (1)

Country Link
CN (1) CN110110391A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111027763A (en) * 2019-12-06 2020-04-17 中国水利水电科学研究院 Basin flood response similarity analysis method based on machine learning
CN114240106A (en) * 2021-12-06 2022-03-25 中国水利水电科学研究院 Basin flood response similarity analysis method based on hydrologic data mining

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103488871A (en) * 2013-08-27 2014-01-01 国家电网公司 Flood forecasting method for area without runoff data
CN103914619A (en) * 2014-03-31 2014-07-09 河海大学 River network modeling method
KR20160130612A (en) * 2015-05-04 2016-11-14 부경대학교 산학협력단 Method for selecting meteorological observation stations by using multifractal detrended fluctuation analysis and network theory
CN107133398A (en) * 2017-04-28 2017-09-05 河海大学 A kind of river ethic Forecasting Methodology based on complex network

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103488871A (en) * 2013-08-27 2014-01-01 国家电网公司 Flood forecasting method for area without runoff data
CN103914619A (en) * 2014-03-31 2014-07-09 河海大学 River network modeling method
KR20160130612A (en) * 2015-05-04 2016-11-14 부경대학교 산학협력단 Method for selecting meteorological observation stations by using multifractal detrended fluctuation analysis and network theory
CN107133398A (en) * 2017-04-28 2017-09-05 河海大学 A kind of river ethic Forecasting Methodology based on complex network

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
ALEXANDER Y. SUN ET.AL: "Patterns of precipitation and soil moisture extremes in Texas, US: A complex network analysis", 《ADVANCES IN WATER RESOURCES》 *
吴孝情等: "珠江流域非平稳性降雨极值时空变化特征及其成因", 《水利学报》 *
吴学文等: "基于复杂网络的水文PUB预测", 《电子测量技术》 *
周芬: "设计洪水估算方法的比较研究", 《中国优秀博硕士学位论文全文数据库 (硕士)工程科技Ⅱ辑(季刊)》 *
李新杰: "河川径流时间序列的非线性特征识别与分析", 《中国博士学位论文全文数据库基础科学辑(月刊)》 *
汪伟: "气候变化情景下水库调度对湄公河洪水的影响研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑(月刊)》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111027763A (en) * 2019-12-06 2020-04-17 中国水利水电科学研究院 Basin flood response similarity analysis method based on machine learning
CN114240106A (en) * 2021-12-06 2022-03-25 中国水利水电科学研究院 Basin flood response similarity analysis method based on hydrologic data mining

Similar Documents

Publication Publication Date Title
Sagarika et al. Pacific Ocean SST and Z500 climate variability and western US seasonal streamflow
Mohamed et al. Water demand forecasting in Umm Al-Quwain (UAE) using the IWR-MAIN specify forecasting model
KR102073768B1 (en) Drought information supply system based on portal
KR20170005553A (en) Floods, drought assessment and forecasting techniques development for intelligent service
Mishra et al. Variability in Canadian seasonal streamflow information and its implication for hydrometric network design
Rahman et al. Regional extreme rainfall mapping for Bangladesh using L-moment technique
Ravazzani et al. Review of time-of-concentration equations and a new proposal in Italy
Srivastava et al. A unified approach to evaluating precipitation frequency estimates with uncertainty quantification: Application to Florida and California watersheds
CN111539597A (en) Gridding drainage basin social and economic drought assessment method
Carvalho Oliveira et al. A climatology of the annual cycle of river discharges into the Brazilian continental shelves: from seasonal to interannual variability
Yuan et al. Research and application of an intelligent networking model for flood forecasting in the arid mountainous basins
Dehghan et al. Spatial clustering of maximum 24‐h rainfall over Urmia Lake Basin by new weighting approaches
CN110110391A (en) A kind of flood estimation method in the scarce measuring runoff data basin based on region division
Sahu et al. Density-based spatial clustering of application with noise approach for regionalisation and its effect on hierarchical clustering
Tursunova et al. Water resources of Kazakhstan in conditions of uncertainty
Bajamgnigni Gbambie et al. Added value of alternative information in interpolated precipitation datasets for hydrology
CN113723703A (en) Water quality prediction method and system based on multi-source data fusion and deep learning
Kratzert et al. A glimpse into the Unobserved: Runoff simulation for ungauged catchments with LSTMs
Lin et al. Hurricane freshwater flood risk assessment model for residential buildings in southeast US coastal states considering climate change
Dariane et al. Forecasting of rainfall using different input selection methods on climate signals for neural network inputs
Wang et al. Optimizing multidam releases in large river basins by combining distributed hydrological inflow predictions with rolling-horizon decision making
Welles Verification of river stage forecasts
Tu et al. Flood hazard and risk assessment of Hoang Long River basin, Vietnam
Villalba et al. Selection of multiple donor gauges via Graphical Lasso for estimation of daily streamflow time series
Karami et al. Investigating the inter-annual precipitation changes of Iran

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination