A hydrological data management method based on data whole-chain space-time coupling quality control

Abstract

The present application relates to a kind of hydrological data governance methods based on data whole chain space-time coupling quality control, solve the current fine hydrological data governance capability, the problem that the hydrological data lacking automatic quality control means of data distortion but not obvious exception.This method includes the following steps: S1, establishing hydrological station standardization sample set;S2, based on hydrological station standardization sample set, carry out hydrological data preliminary screening quality control;S3, based on S1 sample set, S2 data carries out space-time coupling quality control, realizes to data distortion but not obvious exception hydrological data fine quality control, in turn for hydrological station multi-element space-time coupling quality control, river reach same element space-time coupling quality control, basin point surface space-time coupling quality control.The present application can automatically and efficiently distinguish various hydrological data distortion problems by carrying out whole chain space-time coupling quality control to hydrological data, realizes intelligent fine management of hydrological data, provides more accurate and reliable hydrological data support for flood and drought disaster prevention.

Description

Technical Field

[0001] This invention belongs to the field of hydrology and relates to a method for processing hydrological data, and particularly to a method for hydrological data governance based on spatiotemporal coupling quality control of the entire data chain. Background Technology

[0002] Hydrological data is the most direct and fundamental information reflecting the water and rainfall conditions of rivers, lakes, and reservoirs. It is crucial intelligence for flood and drought disaster prevention and control, and plays a vital role in socio-economic development. Especially during typhoons and heavy rains, timely and reliable hydrological data such as peak flood levels and flow rates are essential for flood control scheduling of water conservancy projects and the safe evacuation of residents along rivers. Hydrological data must be accurate and reliable; distorted data conveying incorrect information can have serious consequences. Therefore, quality control of hydrological data is extremely important.

[0003] Currently, hydrological data management primarily occurs at the application database at the end of the data chain. This involves classifying and developing differentiated quality control rules based on different hydrological observation elements such as rainfall and water levels at different stations. Data exceeding normal ranges or variability is identified as anomalous, thus achieving automated quality control. However, this method can only filter out obviously anomalous data such as those exceeding set thresholds or exhibiting significant abrupt changes; it can only be considered a "coarse screening." Obvious anomalous data can often be manually detected and prevented from propagating errors in practical applications. Conversely, data that is distorted but not obviously anomalous is more likely to transmit errors. However, automated quality control methods for this type of data are currently lacking, and no reports or patents documenting related technologies have been found.

[0004] In recent years, with the deepening of digital reform and the rapid development of cutting-edge technologies such as big data and cloud computing, technical conditions have been provided for massive data processing and intelligent decision analysis, making it possible to carry out fully automated and refined hydrological data management. Summary of the Invention

[0005] The purpose of this invention is to address the current shortcomings in the ability to manage refined hydrological data and the lack of automatic quality control methods for hydrological data that is distorted but not obviously abnormal. This invention provides a hydrological data management method based on spatiotemporal coupling quality control of the entire data chain, which can automatically and efficiently identify various types of hydrological data distortion problems and realize intelligent and refined management of hydrological data.

[0006] The technical solution adopted by this invention to solve its technical problem is: a hydrological data management method based on spatiotemporal coupling quality control of the entire data chain, characterized by including the following steps:

[0007] S1. Establish a standardized sample set of hydrological stations and organize the hydrological stations into a standardized array format to facilitate automatic analysis and calculation;

[0008] S2. Based on the standardized sample set of hydrological stations, conduct preliminary screening and quality control of hydrological data, including front-end equipment quality control, equipment joint quality control, and hydrological station data fusion process quality control.

[0009] S3. Based on the S1 sample set and S2 data, spatiotemporal coupling quality control is carried out to achieve refined quality control of hydrological data that is distorted but not obviously abnormal. The methods are: spatiotemporal coupling quality control of multiple elements of hydrological stations, spatiotemporal coupling quality control of the same elements of river sections, and spatiotemporal coupling quality control of watershed points and surfaces.

[0010] Preferably, the establishment of the standardized sample set in step S1 includes the following steps:

[0011] S1.1. Based on the watershed as the unit and the river system classification, establish a standardized hydrological station W, including a sample set G and Z of hydrological water level stations for the main stream, first-level tributaries, ... Nth-level tributaries. 1 ...Z N And the interval rain gauge sample set P;

[0012] W = {G,Z} 1 Z 2 Z 3 ,…,Z N ,P};

[0013] S1.2 Establish a sample set of hydrological water level stations along the main stream, G = {G} j}, G j This represents the j-th hydrological station in the main stream, ordered sequentially from upstream to downstream.

[0014] S1.3 Establish a sample set of hydrological and water level stations for first-order tributaries. For first-order tributaries, G represents the j-th hydrological station on the i-th first-order tributary, ordered sequentially from upstream to downstream. q G represents the first hydrological station downstream of the confluence point of the i-th first-order tributary and the main stream. q ∈G;

[0015] S1.4 Establish a sample set of hydrological water level stations for tributaries below the second level. 2≤k≤N, This represents the j-th hydrological station on the i-th k-th tributary, ordered sequentially from upstream to downstream. The first hydrological station downstream of the confluence point of the i-th k-th tributary and its corresponding k-1-th tributary.

[0016] S1.5. Establish a sample set of interval rainfall stations. Using hydrological water level stations as units, match rainfall stations within the watershed between the upstream hydrological water level station section and the current station section to form a standardized sample set of interval rainfall stations, P = {P x}、P x ={P xl};

[0017] x∈{G,Z 1 Z k}, 2≤k≤N;

[0018] x is one of the hydrological and water level stations along all the main streams and tributaries. P x P represents the interval rainfall station sample set of hydrological water level station x, that is, the set of all rainfall stations between hydrological water level station x and its upstream hydrological water level station; xl These represent the l-th rain gauge station in the x-interval watershed of the hydrological water level station.

[0019] Preferably, the S2 step of conducting preliminary screening and quality control of hydrological data includes the following steps:

[0020] S2.1 Front-end equipment quality control; Based on the effective range of the equipment and the range of water changes, the real-time data of the sensing equipment is quality controlled, and the equipment with obvious abnormal data such as data exceeding the normal range of change, data remaining unchanged for a long time or missing, and data jumping sharply are included in the equipment fault warning.

[0021] S2.2 Joint Equipment Quality Control: A hydrological station may have multiple hydrological observation elements such as water level, rainfall, and flow rate. When there are two or more sensing devices for the same observation element at the station, joint equipment quality control is carried out. When there are two sets of devices for the same element, the differences in real-time data between the two sets of devices at the same moment are compared. The two sets of devices whose errors exceed the reasonable range are included in the equipment fault pending database for further judgment. When there are more than two sets of devices for the same element, if the error of a certain device with two or more sets of devices for the same element exceeds the reasonable range, the device is included in the equipment fault warning. If all devices of the hydrological station are included in the fault warning, the station is included in the station fault warning.

[0022] S2.3, Equipment data is fused with hydrological station data; equipment not included in the equipment fault warning list is fused with station data according to priority order, that is, the real-time station data uses the equipment data with the highest priority first; when the equipment with the highest priority fails, the equipment data with the second highest priority is used, and so on; the priority is set manually according to the difference in equipment accuracy.

[0023] As a preferred option, in step S2.3, the hydrological station data fusion process is quality controlled in conjunction with the station characteristics; the lower limit values ​​of the station characteristics include riverbed elevation, reservoir dead water level, historical lowest water level, and historical minimum flow, and the upper limit values ​​of the station characteristics include river embankment crest elevation, reservoir flood control high water level, reservoir dam crest elevation, historical highest water level, and historical maximum flow. Station data that are significantly lower than the lower limit values ​​or higher than the upper limit values ​​of the station characteristics are judged as abnormal data.

[0024] Preferably, the multi-element spatiotemporal coupling quality control of hydrological stations in S3 includes the following steps:

[0025] S3.1 Perform multi-element spatiotemporal coupling quality control on each hydrological water level station x;

[0026] S3.1.1 Establish the observation element set M for each hydrological water level station. x Taking each hydrological water level station x as a unit, the observation elements of the hydrological water level station are sorted according to causal relationships and included in the observation element set; for example, if a hydrological water level station observes three elements: rainfall a, flow rate b, and water level c, then the observation element set M of that station is... x = (a, b, c);

[0027] S3.1.2 Correlation analysis of various elements of hydrological water level stations, observation element set M x For each observation element, the preceding elements are used as independent variables and the subsequent elements are used as dependent variables. Correlation analysis is performed on the subsequent elements with all their preceding elements. For example, correlation analysis is performed between b and a, and correlation analysis is performed between c and a and b respectively. Considering that the changes of each element may be asynchronous over time, a time displacement variable is added to analyze the correlation of hydrological water level station elements under different time displacements t. Linear regression or nonlinear regression methods are used to fit linear or univariate quadratic regression curves, and the correlation coefficient r1 is calculated respectively.

[0028] S3.1.3 Establish the coupling element set O of the hydrological water level station x For M x For any element, a regression curve with the highest correlation coefficient r1 can be selected, corresponding to another element under a certain time displacement. If the correlation coefficient r1 reaches the manually set coupling allowable value, the spatiotemporal coupling of the two elements is considered successful; otherwise, the coupling fails. This process is repeated for all elements of hydrological water level station x, automatically performing multi-element spatiotemporal coupling analysis. Successfully coupled elements form the coupling element set O of hydrological water level station x. x Coupled element set O x This includes independent variables, dependent variables, time shift values, regression curve equations, and 95% confidence intervals of the regression curves; multiple spatiotemporal coupling elements can correspond to one element, and they are arranged in order of priority from high to low according to the correlation coefficient r1.

[0029] S3.1.4, Conduct multi-element spatiotemporal coupling quality control of hydrological water level stations; based on the coupling element set O of hydrological water level stations. x Based on the real-time data of the independent variable, the spatiotemporal coupling regression value of the dependent variable under the corresponding time shift is found. When the real-time data of the dependent variable deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the dependent variable is distorted.

[0030] Preferably, in step 3.1.2, when analyzing the various elements of the hydrological water level station, the time displacement variable is as follows: the current flow rate b is used to perform correlation analysis with the cumulative rainfall a per unit time before time t1; the current water level c is used to perform correlation analysis with the cumulative rainfall a per unit time before time t2; the current water level c is used to perform correlation analysis with the flow rate b before time t3; and the correlation coefficient r1 is calculated for each of the different time displacement values ​​t1, t2, and t3.

[0031] Preferably, the spatiotemporal coupling quality control of the same element in the river section in S3 includes the following steps:

[0032] S3.2 Conduct spatiotemporal coupling quality control of the same elements in the river section for hydrological water level stations;

[0033] S3.2.1 Correlation analysis of river sections with the same elements; based on the standardized sample set W of hydrological stations, the sample sets G and Z of hydrological water level stations of the main stream, first-order tributaries, second-order tributaries, ..., Nth-order tributaries. 1 Z 2 Z 3 ,…,Z N The correlation analysis of the observation elements of a certain hydrological water level station with the same elements of all hydrological water level stations upstream is carried out. Considering the flood propagation time, the correlation of the same observation elements of each hydrological water level station upstream and downstream under different time displacements t is analyzed. Different time displacements are set between the same elements of upstream and downstream hydrological water level stations. That is, the observation elements of the downstream hydrological water level station at the current time are selected and the same observation elements of the upstream hydrological water level station before time t4 are selected. Linear or univariate quadratic regression curves are fitted, and the correlation coefficient r2 is calculated for different time displacement values.

[0034] S3.2.2 Establish a set of upstream and downstream coupled hydrological and water level stations S; For any observation element of a downstream hydrological and water level station, a regression curve with the highest correlation coefficient r2 can be selected, corresponding to the same observation element of the upstream hydrological and water level station under a certain time displacement condition; If the correlation coefficient r2 reaches the set coupling allowable value, it is considered that the spatiotemporal coupling of the same element of the upstream and downstream hydrological and water level stations is successful, otherwise the coupling is considered to be unsuccessful; By analogy, the spatiotemporal coupling analysis of the same element of the river section is automatically performed on all upstream and downstream hydrological and water level stations. The upstream and downstream hydrological and water level stations that are successfully coupled form a set of upstream and downstream coupled hydrological and water level stations S, which includes the same element of the upstream and downstream hydrological and water level stations, time displacement value, regression curve equation, and 95% confidence interval of the regression curve;

[0035] S3.2.3 Conduct spatiotemporal coupling quality control of the same elements in the river section; Based on the upstream and downstream coupled hydrological and water level station set S, according to the real-time data of the downstream hydrological and water level station, find the spatiotemporal coupling regression value of the same elements of the upstream station under the corresponding time displacement. When the real-time data of the upstream station deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the same elements of both stations may be distorted. Both stations are entered into the distorted station pending database for further judgment.

[0036] S3.2.4 Analyze the undetermined database of distorted stations; when a distorted station has two or more coupled stations, if the real-time data of the station deviates from all coupled stations, the real-time data of the station is determined to be distorted; if the real-time data of the station does not deviate from at least one coupled station, the real-time data of the coupled station that deviated is determined to be distorted; when a station has only one coupled station, the real-time data of both stations may be distorted, pending manual determination.

[0037] Preferably, the spatiotemporal coupling quality control of watershed points and surfaces in S3 includes the following steps:

[0038] S3.3, Conduct spatiotemporal coupling quality control of flow and water level observation elements and rainfall at various hydrological and water level stations in the basin;

[0039] S3.3.1 Calculate the cumulative rainfall of the basin above a certain hydrological station; based on the standardized sample set W of the hydrological station and the interval rainfall station sample set P; the set of all rainfall stations in the basin above a certain hydrological station consists of the interval rainfall station sample set of that station and the interval rainfall station sample sets of all its upstream stations; based on all rainfall stations in the basin above a certain hydrological station, use the arithmetic mean of the rainfall data of all rainfall stations as the areal rainfall value, and calculate the cumulative areal rainfall of the basin above a certain hydrological station for different durations, where different durations are from 1 hour to 3 days;

[0040] S3.3.2 Correlation analysis of watershed points and areas; Considering the differences in runoff generation and confluence time in the watershed, analyze the correlation between the cumulative areal rainfall of various durations above a certain hydrological water level station and the water level and flow elements of the hydrological water level station under different time displacement conditions, fit linear or univariate quadratic regression curves, and calculate the correlation coefficient r3 under different time displacement conditions and different durations.

[0041] S3.3.3 Establish a watershed point-to-surface spatiotemporal coupling element set R; For any hydrological station's water level and discharge elements, a regression curve with the highest correlation coefficient r3 can be selected, corresponding to the cumulative areal rainfall over a certain duration in the watershed above the hydrological station under a certain time displacement; If the correlation coefficient r3 reaches the set coupling allowable value, the spatiotemporal coupling of the two elements is considered successful, otherwise the coupling fails; and so on, automatically perform watershed point-to-surface spatiotemporal coupling analysis on all hydrological station elements. The successfully coupled station elements form a watershed point-to-surface spatiotemporal coupling element set R, which includes water level or discharge elements, cumulative areal rainfall over a certain duration in the watershed above the hydrological station, time displacement value, regression curve equation, and 95% confidence interval of the regression curve;

[0042] S3.3.4. Conduct spatiotemporal coupling quality control of watershed points and surfaces; based on the spatiotemporal coupling element set R of watershed points and surfaces, according to the real-time data of water level and flow at hydrological water level stations, find the spatiotemporal coupling regression value of the cumulative areal rainfall over a certain duration in the watershed above the hydrological water level station under the corresponding time displacement. When the real-time data of the cumulative areal rainfall over a certain duration in the watershed above a certain hydrological water level station deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of water level and flow at a certain hydrological water level station is distorted.

[0043] As a preferred option, in step 3.3.2, different time shifts are set between the cumulative areal rainfall over different durations in the basin above the hydrological water level station and the water level and flow factors of the hydrological water level station. That is, the water level or flow factor of the hydrological water level station at the current time is selected and the cumulative areal rainfall over Δt duration in the basin above the hydrological water level station before time t5 is used to fit a linear or univariate quadratic regression curve. The correlation coefficient r3 is calculated for different time shift values ​​and different cumulative areal rainfall over different durations.

[0044] As a preferred option, in steps S2 and S3, when abnormal / distorted data from hydrological stations are detected, individual abnormal / distorted data are individually deleted or corrected through automatic filtering or manual verification; for continuous abnormal / distorted data, the next-highest priority device is selected for data re-fusion, re-quality control of the fusion process, and re-quality control of spatiotemporal coupling, based on device priority; devices with continuous abnormal / distorted data are included in the equipment fault warning; if all devices at a station are included in the fault warning, then the station is included in the station fault warning.

[0045] This invention enables the automatic and efficient identification of various hydrological data distortion problems by conducting full-chain spatiotemporal coupled quality control on hydrological data, thereby achieving intelligent and refined management of hydrological data and providing more accurate and reliable hydrological data support for flood and drought disaster prevention.

[0046] First, quality control covers the entire hydrological data flow chain. From equipment quality control and joint equipment quality control at the front end of the data chain, to quality control of the data fusion process at the stations, and then to spatiotemporal coupling quality control at the end of the data chain, front-end quality control can promptly detect equipment problems, breaking the traditional situation of only conducting quality control at the end of the data chain. This realizes the transformation from the original "one station, one data" quality control to "one data, multiple sources" quality control, which is conducive to timely and comprehensive detection of data problems.

[0047] Second, it achieves comprehensive quality control through point-line-surface spatiotemporal coupling. From point quality control of multi-element spatiotemporal coupling at monitoring stations, to line quality control of the same element spatiotemporal coupling in river sections, and then to surface quality control of point-surface spatiotemporal coupling in the watershed, it conducts comprehensive and multi-dimensional point-line-surface coupling analysis of hydrological data, intelligently identifying distorted data and breaking through the limitations of traditional quality control methods that can only perform "coarse screening." For monitoring stations where data distortion is detected, it triggers the re-fusion of data from the front-end equipment in the data chain, achieving effective linkage between front-end and back-end quality control and comprehensively improving the data assurance of monitoring stations. Attached Figure Description

[0048] The invention will now be further described with reference to the accompanying drawings.

[0049] Figure 1 This is a diagram illustrating the main steps of the present invention.

[0050] Figure 2 This is a detailed flowchart of the present invention. Detailed Implementation

[0051] The present invention will be further described below with reference to specific embodiments and accompanying drawings.

[0052] Example: A hydrological data management method based on spatiotemporal coupling quality control across the entire data chain, such as... Figure 1 , 2 As shown. Includes the following steps:

[0053] S1. Establish a standardized sample set of hydrological stations and organize the hydrological stations into a standardized array format to facilitate automatic analysis and calculation;

[0054] S1.1. Based on the watershed as the unit and the river system classification, establish a standardized hydrological station W, including a sample set G and Z of hydrological water level stations for the main stream, first-level tributaries, ... Nth-level tributaries. 1 ...Z N And the interval rain gauge sample set P;

[0055] W = {G,Z} 1 Z 2 Z 3 ,…,Z N ,P};

[0056] S1.2 Establish a sample set of hydrological water level stations along the main stream, G = {G} j}, G j This represents the j-th hydrological station in the main stream, ordered sequentially from upstream to downstream.

[0057] S1.3 Establish a sample set of hydrological and water level stations for first-order tributaries. For first-order tributaries, G represents the j-th hydrological station on the i-th first-order tributary, ordered sequentially from upstream to downstream. q G represents the first hydrological station downstream of the confluence point of the i-th first-order tributary and the main stream. q ∈G;

[0058] S1.4 Establish a sample set of hydrological water level stations for tributaries below the second level. 2≤k≤N, This represents the j-th hydrological station on the i-th k-th tributary, ordered sequentially from upstream to downstream. The first hydrological station downstream of the confluence point of the i-th k-th tributary and its corresponding k-1-th tributary.

[0059] S1.5. Establish a sample set of interval rainfall stations. Using hydrological water level stations as units, match rainfall stations within the watershed between the upstream hydrological water level station section and the current station section to form a standardized sample set of interval rainfall stations, P = {P x}、P x ={P xl};

[0060] x∈{G,Z 1 Z k}, 2≤k≤N;

[0061] x is one of the hydrological and water level stations along all the main streams and tributaries. P x P represents the interval rainfall station sample set of hydrological water level station x, that is, the set of all rainfall stations between hydrological water level station x and its upstream hydrological water level station; xl These represent the l-th rain gauge station in the x-interval watershed of the hydrological water level station.

[0062] S2. Based on the standardized sample set of hydrological stations, conduct preliminary screening and quality control of hydrological data, including front-end equipment quality control, equipment joint quality control, and hydrological station data fusion process quality control.

[0063] S2.1 Front-end equipment quality control; Based on the effective range of the equipment and the range of water changes, the real-time data of the sensing equipment is quality controlled, and the equipment with obvious abnormal data such as data exceeding the normal range of change, data remaining unchanged for a long time or missing, and data jumping sharply are included in the equipment fault warning.

[0064] S2.2 Joint Equipment Quality Control: A hydrological station may have multiple hydrological observation elements such as water level, rainfall, and flow rate. When there are two or more sensing devices for the same observation element at the station, joint equipment quality control is carried out. When there are two sets of devices for the same element, the differences in real-time data between the two sets of devices at the same moment are compared. The two sets of devices whose errors exceed the reasonable range are included in the equipment fault pending database for further judgment. When there are more than two sets of devices for the same element, if the error of a certain device with two or more sets of devices for the same element exceeds the reasonable range, the device is included in the equipment fault warning. If all devices of the hydrological station are included in the fault warning, the station is included in the station fault warning.

[0065] S2.3. Equipment data is fused with hydrological station data. Equipment not included in the equipment fault warning list is fused with station data according to priority order, i.e., the real-time station data uses the highest priority equipment data first; when the highest priority equipment fails, the second highest priority equipment data is used, and so on; the priority is manually set based on the differences in equipment accuracy. Quality control is carried out on the hydrological station data fusion process in conjunction with station characteristics; the lower limits of station characteristics include riverbed elevation, reservoir dead water level, historical minimum water level, and historical minimum flow; the upper limits of station characteristics include river embankment crest elevation, reservoir flood control high water level, reservoir dam crest elevation, historical maximum water level, and historical maximum flow. Station data that is significantly lower than the lower limit or higher than the upper limit of station characteristics is judged as abnormal data.

[0066] S3. Based on the S1 sample set and S2 data, spatiotemporal coupling quality control is carried out to achieve refined quality control of hydrological data that is distorted but not obviously abnormal. The methods are: spatiotemporal coupling quality control of multiple elements of hydrological stations, spatiotemporal coupling quality control of the same elements of river sections, and spatiotemporal coupling quality control of watershed points and areas.

[0067] S3.1 Perform multi-element spatiotemporal coupling quality control on each hydrological water level station x;

[0068] S3.1.1 Establish the observation element set M for each hydrological water level station. x Taking each hydrological water level station x as a unit, the observation elements of the hydrological water level station are sorted according to causal relationships and included in the observation element set; for example, if a hydrological water level station observes three elements: rainfall a, flow rate b, and water level c, then the observation element set M of that station is... x = (a, b, c);

[0069] S3.1.2 Correlation analysis of various elements of hydrological water level stations, observation element set M xFor each observed element, the preceding elements are used as independent variables and the subsequent elements as dependent variables. Correlation analysis is performed on the subsequent elements with all their preceding elements; for example, b is correlated with a, and c is correlated with both a and b. Considering that the changes of each element may be asynchronous over time, a time displacement variable is added to analyze the correlation of hydrological water level station elements at different time displacements t. Linear regression or nonlinear regression methods are used to fit linear or quadratic regression curves, and the correlation coefficient r1 is calculated respectively. When analyzing each element of the hydrological water level station, the time displacement variable is: the current flow rate b is used to perform correlation analysis with the cumulative rainfall a per unit time before time t1; the current water level c is used to perform correlation analysis with the cumulative rainfall a per unit time before time t2; and the current water level c is used to perform correlation analysis with the flow rate b before time t3. Correlation coefficient r1 is calculated for each of the different time displacement values ​​t1, t2, and t3.

[0070] S3.1.3 Establish the coupling element set O of the hydrological water level station x For M x For any element, a regression curve with the highest correlation coefficient r1 can be selected, corresponding to another element under a certain time displacement. If the correlation coefficient r1 reaches the manually set coupling allowable value, the spatiotemporal coupling of the two elements is considered successful; otherwise, the coupling fails. This process is repeated for all elements of hydrological water level station x, automatically performing multi-element spatiotemporal coupling analysis. Successfully coupled elements form the coupling element set O of hydrological water level station x. x Coupled element set O x This includes independent variables, dependent variables, time shift values, regression curve equations, and 95% confidence intervals of the regression curves; multiple spatiotemporal coupling elements can correspond to one element, and they are prioritized from high to low according to their correlation coefficient r1.

[0071] S3.1.4, Conduct multi-element spatiotemporal coupling quality control of hydrological water level stations; based on the coupling element set O of hydrological water level stations. x Based on the real-time data of the independent variable, the spatiotemporal coupling regression value of the dependent variable under the corresponding time shift is found. When the real-time data of the dependent variable deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the dependent variable is distorted.

[0072] S3.2 Conduct spatiotemporal coupling quality control of the same elements in the river section for hydrological water level stations;

[0073] S3.2.1 Correlation analysis of river sections with the same elements; based on the standardized sample set W of hydrological stations, the sample sets G and Z of hydrological water level stations of the main stream, first-order tributaries, second-order tributaries, ..., Nth-order tributaries. 1 Z 2 Z 3 ,…,Z NThe correlation analysis of the observation elements of a certain hydrological water level station with the same elements of all hydrological water level stations upstream is carried out. Considering the flood propagation time, the correlation of the same observation elements of each hydrological water level station upstream and downstream under different time displacements t is analyzed. Different time displacements are set between the same elements of upstream and downstream hydrological water level stations. That is, the observation elements of the downstream hydrological water level station at the current time are selected and the same observation elements of the upstream hydrological water level station before time t4 are selected. Linear or univariate quadratic regression curves are fitted, and the correlation coefficient r2 is calculated for different time displacement values.

[0074] S3.2.2 Establish a set of upstream and downstream coupled hydrological and water level stations S; For any observation element of a downstream hydrological and water level station, a regression curve with the highest correlation coefficient r2 can be selected, corresponding to the same observation element of the upstream hydrological and water level station under a certain time displacement condition; If the correlation coefficient r2 reaches the set coupling allowable value, it is considered that the spatiotemporal coupling of the same element of the upstream and downstream hydrological and water level stations is successful, otherwise the coupling is considered to be unsuccessful; By analogy, the spatiotemporal coupling analysis of the same element of the river section is automatically performed on all upstream and downstream hydrological and water level stations. The upstream and downstream hydrological and water level stations that are successfully coupled form a set of upstream and downstream coupled hydrological and water level stations S, which includes the same element of the upstream and downstream hydrological and water level stations, time displacement value, regression curve equation, and 95% confidence interval of the regression curve;

[0075] S3.2.3 Conduct spatiotemporal coupling quality control of the same elements in the river section; Based on the upstream and downstream coupled hydrological and water level station set S, according to the real-time data of the downstream hydrological and water level station, find the spatiotemporal coupling regression value of the same elements of the upstream station under the corresponding time displacement. When the real-time data of the upstream station deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the same elements of both stations may be distorted. Both stations are entered into the distorted station pending database for further judgment.

[0076] S3.2.4 Analyze the undetermined database of distorted stations; when a distorted station has two or more coupled stations, if the real-time data of the station deviates from all coupled stations, the real-time data of the station is determined to be distorted; if the real-time data of the station does not deviate from at least one coupled station, the real-time data of the coupled station that deviated is determined to be distorted; when a station has only one coupled station, the real-time data of both stations may be distorted, pending manual determination.

[0077] S3.3, Conduct spatiotemporal coupling quality control of flow and water level observation elements and rainfall at various hydrological and water level stations in the basin;

[0078] S3.3.1 Calculate the cumulative rainfall of the basin above a certain hydrological station; based on the standardized sample set W of the hydrological station and the interval rainfall station sample set P; the set of all rainfall stations in the basin above a certain hydrological station consists of the interval rainfall station sample set of that station and the interval rainfall station sample sets of all its upstream stations; based on all rainfall stations in the basin above a certain hydrological station, use the arithmetic mean of the rainfall data of all rainfall stations as the areal rainfall value, and calculate the cumulative areal rainfall of the basin above a certain hydrological station for different durations, where different durations are from 1 hour to 3 days;

[0079] S3.3.2, Correlation Analysis of Point-to-Surface Areas in the Watershed: Considering the differences in runoff generation and confluence times in the watershed, the correlation between the cumulative areal rainfall over different durations and the water level and flow factors of the hydrological station above a certain hydrological station under different time displacements is analyzed. Linear or quadratic regression curves are fitted, and the correlation coefficient r3 is calculated under different time displacements and for different durations. Different time displacements are set between the cumulative areal rainfall over different durations and the water level and flow factors of the hydrological station above a certain hydrological station. That is, the water level or flow factor of the hydrological station at the current time is selected and the cumulative areal rainfall over Δt duration of the hydrological station above the hydrological station before time t5 is selected. Linear or quadratic regression curves are fitted, and the correlation coefficient r3 is calculated for different time displacement values ​​and different cumulative areal rainfall over different durations.

[0080] S3.3.3 Establish a watershed point-to-surface spatiotemporal coupling element set R; For any hydrological station's water level and discharge elements, a regression curve with the highest correlation coefficient r3 can be selected, corresponding to the cumulative areal rainfall over a certain duration in the watershed above the hydrological station under a certain time displacement; If the correlation coefficient r3 reaches the set coupling allowable value, the spatiotemporal coupling of the two elements is considered successful, otherwise the coupling fails; and so on, automatically perform watershed point-to-surface spatiotemporal coupling analysis on all hydrological station elements. The successfully coupled station elements form a watershed point-to-surface spatiotemporal coupling element set R, which includes water level or discharge elements, cumulative areal rainfall over a certain duration in the watershed above the hydrological station, time displacement value, regression curve equation, and 95% confidence interval of the regression curve;

[0081] S3.3.4. Conduct spatiotemporal coupling quality control of watershed points and surfaces; based on the spatiotemporal coupling element set R of watershed points and surfaces, according to the real-time data of water level and flow at hydrological water level stations, find the spatiotemporal coupling regression value of the cumulative areal rainfall over a certain duration in the watershed above the hydrological water level station under the corresponding time displacement. When the real-time data of the cumulative areal rainfall over a certain duration in the watershed above a certain hydrological water level station deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of water level and flow at a certain hydrological water level station is distorted.

[0082] In steps S2 and S3, the abnormality / distortion handling methods are as follows: when abnormality / distortion is detected in hydrological station data, individual abnormal / distorted data are individually deleted or corrected through automatic filtering or manual verification; for continuous abnormality / distortion, the next highest priority equipment is selected for station data re-fusion, re-quality control of the fusion process, and re-quality control of spatiotemporal coupling, based on equipment priority; equipment with continuous abnormality / distortion is included in the equipment fault warning; if all equipment at a station is included in the fault warning, then the station is included in the station fault warning.

Claims

1. A hydrological data governance method based on data full-chain spatiotemporal coupling quality control, characterized in that: The method comprises the following steps: S1, establishing a hydrological station standardization sample set; S2, based on the hydrological station standardization sample set, carrying out hydrological data preliminary screening quality control, including front-end equipment quality control, equipment joint quality control, hydrological station data fusion process quality control; S2 carries out hydrological data preliminary screening quality control, including the following steps: S2.1, front-end equipment quality control; according to the effective range of the equipment, the water body change range, the real-time data of the sensing equipment is quality controlled, the data out of the normal change range, the data long-term unchanged or missing, the data large jump data obviously abnormal equipment is listed in the equipment fault early warning; S2.2, equipment joint quality control; a hydrological station can have water level, rainfall, flow and other hydrological observation elements, when the station has two sets and more than two sets of sensing equipment for the same observation element, the equipment joint quality control is carried out; when the same element equipment is two sets, the real-time data difference of the two sets of equipment at the same time is compared, and the two sets of equipment with error exceeding the reasonable range are listed in the equipment fault pending library for further identification; when the same element equipment is more than two sets, if the error of a certain equipment and two sets and more than two sets of the same element equipment exceeds the reasonable range, the equipment is listed in the equipment fault early warning; if all the equipment of the hydrological station is listed in the fault early warning, the station is listed in the station fault early warning; S2.3, equipment data to hydrological station data fusion; the equipment not listed in the equipment fault early warning is subjected to station data fusion according to the priority order, that is, the station real-time data adopts the highest priority equipment data; when the highest priority equipment fails, the second highest priority equipment data is adopted, and so on; the priority is set according to the difference of the equipment accuracy; S3, based on S1 sample set and S2 data, time-space coupling quality control is carried out to realize fine quality control of hydrological data distortion but not obvious abnormality, which is in turn hydrological station multi-element time-space coupling quality control, river section same element time-space coupling quality control, and basin point-surface time-space coupling quality control.

2. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 1, characterized in that: The establishment of the step S1 standardization sample set comprises the following steps: S1.1, Establish a standard sample set W of hydrological stations, including sample sets G, Z of hydrological water level stations of the main stream, the first-order tributaries, … N-order tributaries, and sample set P of interval rainfall stations, based on river system classification, taking a river basin as a unit 1 N ​​ W = {G, Z 1 ,Z 2 ,Z 3 ,…, Z N ,P} ; S1.2, Establishing a sample set of trunk hydrological water level stations, G = {G j}, G j represents the jthhydrological water level station in the upstream-to-downstream order of the trunk river S1.3, establishing a sample set of hydrological water level stations of the primary tributaries, for the primary tributaries, represents the jth hydrological water level station of the ith primary tributary in order from upstream to downstream, G q represents the first hydrological water level station in the downstream direction of the confluence point of the ith primary tributary into the main stream, G q ∈G; S1.4, a sample set of hydrological water level stations of the secondary and below tributaries is established, for the secondary and below tributaries, 2≤k≤N, denotes the jth hydrological water level station of the ith k-level tributary in the order from upstream to downstream, is the first hydrological water level station downstream of the confluence point of the ith k-level tributary into the corresponding k-1-level tributary, S1.5, Establish interval rainfall station sample set, take hydrological water level station as unit, match rainfall station in interval basin below section of upstream hydrological water level station and above section of this station, form interval rainfall station standardized sample set, P={P x}, P x ={P xl} ; x e {G, Z 1 ,Z k}, 2≤k≤N; x is one of all hydrological water level stations of the main stream and tributaries, P x The interval rainfall station sample set representing the hydrological water level station x, that is, all rainfall station sets between the hydrological water level station x and its upstream hydrological water level station;P xl Respectively represents the lth rainfall station on the interval basin of the hydrological water level station x.

3. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 1, characterized in that: In step S2.3, the hydrological station data fusion process is quality controlled in combination with the station characteristics; the lower limit value of the station characteristics is the river bottom elevation, the reservoir dead water level, the historical minimum water level and the historical minimum flow, and the upper limit value of the station characteristics is the river embankment top elevation, the reservoir flood control high water level, the reservoir dam top elevation, the historical maximum water level and the historical maximum flow; the station data obviously lower than the lower limit value of the station characteristics or higher than the upper limit value of the station characteristics is determined as abnormal data.

4. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 2 or 3, characterized in that: The hydrological station multi-element time-space coupling quality control in S3 comprises the following steps: S3.1, carrying out station multi-element time-space coupling quality control on each hydrological water level station x; S3.1.1, Establishing the observation element set M of each hydrological water level station x ; the observation elements of the hydrological water level station are sorted according to the causal relationship and listed in the observation element set; for example, the observation elements of a hydrological water level station are rain a, flow b and water level c, then the observation element set M of the station is x = (a, b, c); S3.1.2, correlation analysis of each element of hydrological water level station, observation element set M x Each observation element, the former element as the independent variable, the latter element as the dependent variable, the latter element and all its former elements are analyzed for correlation; such as b and a are analyzed for correlation, c is analyzed for correlation with a and b respectively; considering that the changes of each element may not be synchronized in time, a time displacement variable is added to analyze the correlation of the elements of the hydrological water level station at different time displacements t, linear regression or nonlinear regression method is used to fit linear or one-dimensional quadratic regression curve, and correlation coefficient r1 is calculated respectively; S3.1.3 Establish the coupling element set O of the hydrological water level station x For M x For any element, a regression curve with the highest correlation coefficient r1 can be selected, corresponding to another element under a certain time displacement. If the correlation coefficient r1 reaches the manually set coupling allowable value, the spatiotemporal coupling of the two elements is considered successful; otherwise, the coupling fails. This process is repeated for all elements of hydrological water level station x, automatically performing multi-element spatiotemporal coupling analysis. Successfully coupled elements form the coupling element set O of hydrological water level station x. x Coupled element set O x This includes independent variables, dependent variables, time shift values, regression curve equations, and 95% confidence intervals of the regression curves; multiple spatiotemporal coupling elements can correspond to one element, and they are arranged in order of priority from high to low according to the correlation coefficient r1. S3.1.4, carry out hydrological water level station multi-element space-time coupling quality control; based on the coupled element set O of the hydrological water level station x According to the real-time data of the independent variable, the strain variable space-time coupling regression value under the corresponding time displacement is found, and when the real-time data of the strain variable deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the strain variable is distorted.

5. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 4, characterized in that: In step 3.1.2, when analyzing each element of the hydrological water level station, the time displacement variable is: selecting the flow b at the current time and the unit time cumulative rainfall a before t1 time for correlation analysis; selecting the water level c at the current time and the unit time cumulative rainfall a before t2 time for correlation analysis, and selecting the water level c at the current time and the flow b before t3 time for correlation analysis; for different time displacement values t1, t2 and t3, the respective correlation coefficients r1 are calculated.

6. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 4, characterized in that: The river section same element space-time coupling quality control in S3 includes the following steps: S3.2, river section same element space-time coupling quality control is carried out on the hydrological water level station; S3.2.1, correlation analysis of the same element of the river section; based on the standardized sample set W of the main stream, the first-order tributary, the second-order tributary, …, the N-order tributary hydrological water level station sample set G, Z 1 2 3 N , the observed elements of a certain hydrological water level station and the same elements of all upstream hydrological water level stations are respectively analyzed, the correlation of the same observed elements of the upstream and downstream hydrological water level stations is analyzed under different time displacements t, different time displacements are set between the same elements of the upstream and downstream hydrological water level stations, that is, the observed elements of the downstream hydrological water level station at the current time are selected, and the same observed elements of the upstream hydrological water level station before t4 time are selected, linear or monomial quadratic regression curves are fitted, and different time displacement values are used to calculate the correlation coefficients r2;​​​ S3.2.2, the upstream and downstream coupled hydrological water level station set S is established;For the observation element of any downstream hydrological water level station, a regression curve with the highest correlation coefficient r2 can be selected, which corresponds to the same observation element of the upstream hydrological water level station under certain time displacement condition;If the correlation coefficient r2 reaches the set coupling allowed value, it is considered that the upstream and downstream hydrological water level stations are successfully coupled in the same element space-time, otherwise the coupling fails;In this way, the river section same element space-time coupling analysis is automatically carried out on all upstream and downstream hydrological water level stations, and the upstream and downstream coupled hydrological water level station set S is formed, which includes the same element of the upstream and downstream hydrological water level stations, the time displacement value, the regression curve equation and the 95% confidence interval of the regression curve; S3.2.3, river section same element space-time coupling quality control is carried out;Based on the upstream and downstream coupled hydrological water level station set S, the real-time data of the downstream hydrological water level station is used to find the space-time coupling regression value of the upstream station under the corresponding time displacement, and when the real-time data of the upstream station deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the two stations may be distorted, and both stations enter the distorted station pending library for further determination; S3.2.4, analysis is carried out on the distorted station pending library;When there are two or more coupling stations for a certain distorted station, if the real-time data of the station deviates from all coupling stations, it is determined that the real-time data of the station is distorted;If the real-time data of the station does not deviate from at least one coupling station, it is determined that the real-time data of the deviated coupling station is distorted;When there is only one coupling station for a certain station, it is determined that the real-time data of both stations may be distorted, which is determined manually.

7. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 2 or 3, characterized in that: The basin point and plane space-time coupling quality control in S3 includes the following steps: S3.3, basin point and plane space-time coupling quality control is carried out on the flow and water level observation elements of the hydrological water level station and rainfall; S3.3.1, the cumulative rainfall of the basin area above a certain hydrological water level station is calculated;The interval rainfall station sample set P of the hydrological station standardization sample set W is used;The rainfall station set of the basin area above a certain hydrological water level station is composed of the interval rainfall station sample set of the station and the interval rainfall station sample set of all upstream stations;The arithmetic mean of the rainfall data of all rainfall stations is used as the plane rainfall value, and the cumulative plane rainfall of the basin area above a certain hydrological water level station is calculated under different durations, where the different durations are 1 hour to 3 days; S3.3.2, basin point and plane correlation analysis;Considering the time difference of basin production and runoff, the correlation between the cumulative plane rainfall of the basin area above a certain hydrological water level station and the water level and flow elements of the hydrological water level station under different time displacement conditions is analyzed, linear or monomial quadratic regression curves are fitted, and the correlation coefficients r3 under different time displacement conditions and different durations are calculated. S3.3.3, establish the basin point-surface spatio-temporal coupling element set R; for any hydrological water level station water level element and flow element, a regression curve with the highest correlation coefficient r3 can be selected, corresponding to the certain time displacement condition of the certain duration cumulative surface rainfall above the hydrological water level station; if the correlation coefficient r3 reaches the set coupling allowed value, it is considered that the two elements are successfully coupled in space and time, otherwise the coupling fails; similarly, all hydrological water level station elements are automatically analyzed for basin point-surface spatio-temporal coupling, and the successfully coupled station elements form the basin point-surface spatio-temporal coupling element set R, R includes water level or flow element, certain duration cumulative surface rainfall above the hydrological water level station, time displacement value, regression curve equation, and 95% confidence interval of the regression curve; S3.3.4, perform basin point-surface spatio-temporal coupling quality control; based on the basin point-surface spatio-temporal coupling element set R, according to the real-time data of the water level and flow of the hydrological water level station, the corresponding time displacement condition of the certain duration cumulative surface rainfall above the hydrological water level station is searched, and when the real-time data of the certain duration cumulative surface rainfall above the hydrological water level station deviates from the 95% confidence interval of the regression curve, it is determined that the real-time data of the water level and flow of the hydrological water level station is distorted.

8. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 7, characterized in that: In step 3.3.2, different time displacements are set between the various different duration cumulative surface rainfalls above the hydrological water level station and the water level elements and flow elements of the hydrological water level station, i.e. the current time water level or flow element of the hydrological water level station is selected and the Δt duration cumulative surface rainfall above the hydrological water level station before t5 time is fitted to obtain a linear or one-dimensional quadratic regression curve, and different time displacement values and different duration cumulative surface rainfalls are used to calculate the correlation coefficient r3.

9. The hydrological data governance method based on data full-chain space-time coupling quality control according to claim 3, characterized in that: In steps S2 and S3, when abnormal / distorted data of the hydrological station is found, the individual abnormal / distorted data is deleted or corrected by automatic filtering or manual checking method; for continuous abnormal / distorted data, according to the priority of the equipment, the next priority equipment is selected to perform station data re-fusion, re-fusion process quality control and spatio-temporal coupling quality control; The equipment with continuous abnormal / distorted data is listed in the equipment fault warning; If all the equipment of the station is listed in the fault warning, the station is listed in the station fault warning. If all the equipment of the station is listed in the fault warning, the station is listed in the station fault warning.

Images