1. The rainfall acquisition method at the loss of the rainfall station based on intrinsic orthogonal decomposition is characterized by comprising the following steps:

step 1: selecting a plurality of adjacent rainfall stations according to the positions of the missing stations, sorting rainfall data of the stations, and acquiring rainfall data containing original rainfall field data;

step 2: respectively taking original rainfall field data and rainfall field data subjected to artificial space-time normalization as input, and performing intrinsic orthogonal decomposition on the rainfall field twice to obtain a main mode and a corresponding time coefficient of the rainfall field; the modes correspond to the rain intensity state after the rainfall field is decomposed, and the main modes are those modes with high contribution rate to the energy of the rainfall field;

and step 3: screening out a mode representing the space-time attribute and a corresponding time coefficient according to the decomposition result of the step 2;

and 4, step 4: and (3) eliminating the modes which are screened out in the step (3) and can reflect the rainfall spatio-temporal characteristics from the main modes, and reconstructing the rainfall process of the position of the missing station by using the residual modes by taking the station closest to the position of the missing station as a template to obtain rainfall data of the missing station.

2. The rainfall station deficiency rainfall acquisition method based on intrinsic orthogonal decomposition according to claim 1, wherein:

in step 1, the acquired rainfall data has the following specific requirements: reconstructing the required rainfall data into daily scale in the rainfall process at the missing part of the rainfall station for a long time; reconstructing rainfall data required by the rainfall process at the deficiency position of the medium-long duration rainfall station into an hour scale; and reconstructing the required rainfall data into a minute scale in the rainfall process at the short-duration rainfall station missing part.

3. The rainfall station deficiency rainfall acquisition method based on intrinsic orthogonal decomposition according to claim 1, wherein:

in step 2, performing intrinsic orthogonal decomposition on rainfall sample data of different sites to obtain main modes of the rainfall field and time coefficients corresponding to the modes, wherein the intrinsic orthogonal decomposition specifically operates as follows:

(1) the rainfall field is marked as X ═ v (X, t)¹)v(x，t²)...v(x，tⁿ)]∈R^m×nWherein X represents a rainfall field; t represents time, n is the maximum value of the time sequence, R is a real number field, and m is the number of stations;represents t after subtraction of the time averageⁿ(ii) the intensity of rainfall at the moment; u (x, t)ⁿ) Indicating the rain intensity of the station at time n x,representing the average rain intensity of the x station at all times;

(2) singular value decomposition X-phi sigma psi is carried out on rainfall field X^TWhere phi is a mode matrix, and sigma is a mode energetic diagonal matrix, psi^TIs a time coefficient matrix;

selecting a main mode, and reconstructing a rainfall field:wherein j is a mode number, λ_jIs a modal characteristic value, whose value can reflect the modal energy contribution rate, a_j(t) is the time coefficient vector of the j-th mode, phi_j(x) Is a mode value vector of the jth mode.

4. The rainfall station deficiency rainfall acquisition method based on intrinsic orthogonal decomposition according to claim 1, wherein:

in step 2, the mode with the accumulated energy contribution rate of more than 60% to the rainfall field is taken as the main mode.

5. The rainfall station deficiency rainfall acquisition method based on intrinsic orthogonal decomposition according to claim 1, wherein:

in step 2, the main modes with the contribution rate of the accumulated energy of the rainfall field exceeding 80% are extracted to perform rainfall type reconstruction.

6. The rainfall station deficiency rainfall acquisition method based on intrinsic orthogonal decomposition according to claim 1, wherein:

in step 4, the rainfall process of the closest site is selected as a reconstruction template.

7. A rainfall acquisition system based on intrinsic orthogonal decomposition at the loss of a rainfall station is characterized by comprising:

the acquisition part selects a plurality of adjacent rainfall stations according to the positions of the missing stations, collects and arranges rainfall data of the stations, and acquires original rainfall field data and rainfall field data after artificial space-time normalization;

the rainfall data orthogonal decomposition part is used for respectively inputting original rainfall field data and rainfall field data subjected to artificial space-time normalization and carrying out intrinsic orthogonal decomposition on the rainfall field twice to obtain a main mode and a corresponding time coefficient of the rainfall field; the modes correspond to the rain intensity state after the rainfall field is decomposed, and the main modes are those modes with high contribution rate to the energy of the rainfall field;

the screening part screens out a mode representing the space-time attribute and a corresponding time coefficient according to the decomposition result of the rainfall data orthogonal decomposition part;

the missing rainfall acquisition part is used for eliminating the mode which is screened out by the screening part and can reflect the rainfall spatio-temporal characteristics from the main modes, and reconstructing the rainfall process of the position of the missing station by using the station which is closer to the position of the missing station as a template and the residual mode to obtain the rainfall data of the missing station; and

and the control part is in communication connection with the acquisition part, the rainfall data orthogonal decomposition part, the screening part and the missing rainfall acquisition part and controls the operation of the acquisition part, the rainfall data orthogonal decomposition part, the screening part and the missing rainfall acquisition part.

8. The rainfall station deficiency rainfall acquisition system based on intrinsic orthogonal decomposition of claim 7 further comprising:

and the input display part is in communication connection with the acquisition part, the rainfall data orthogonal decomposition part, the screening part, the missing rainfall acquisition part and the control part and is used for allowing a user to input an operation instruction and performing corresponding display.

Background

Rainfall data is the basis of production convergence calculation, and whether the rainfall data is accurate and sufficient has a significant influence on hydrological calculation. The rainfall process is a highly heterogeneous process and is influenced by multiple factors such as time, space, terrain, wind speed and wind direction, and the non-uniformity of rainfall is one of important sources of uncertainty of a hydrological model.

While accurate simulation of the location and intensity of rainfall is critical to water resource management and hydrologic research, currently, hydrologists put more effort on developing more sophisticated rainfall-runoff simulation methods rather than developing improved techniques to measure and predict rainfall spatiotemporal changes. The rain gauge is the most common instrument for measuring rainfall, but the rain gauge only can provide point measurement results and lacks information on spatial variation. However, a large amount of manpower, material resources and financial resources are required to be invested for building and maintaining a high-density rainfall station network, and the number of the existing rainfall stations in China is limited, so that the requirement of hydrological calculation cannot be met frequently. Therefore, it is common practice to simulate the rainfall process at the missing site by replacing the rainfall process at the missing site with data from neighboring sites, but this practice often causes large errors due to the temporal and spatial inconsistency of the rainfall itself.

Disclosure of Invention

The present invention is made to solve the above problems, and an object of the present invention is to provide a rainfall acquisition method and system based on intrinsic orthogonal decomposition at the loss of a rainfall station.

In order to achieve the purpose, the invention adopts the following scheme:

< method >

The invention provides a rainfall acquisition method at a rainfall station deletion position based on intrinsic orthogonal decomposition, which is characterized by comprising the following steps of:

step 1: selecting adjacent rainfall stations as many as possible according to the positions of the missing stations, sorting the rainfall data of the stations, and acquiring rainfall data containing original rainfall field data;

step 2: respectively taking original rainfall field data and rainfall field data subjected to artificial space-time normalization as input (the original rainfall field data are taken as input in the first intrinsic orthogonal decomposition, and the rainfall field data subjected to artificial space-time normalization are taken as input in the second intrinsic orthogonal decomposition), and performing two times of intrinsic orthogonal decomposition on the rainfall field to obtain a main mode and a corresponding time coefficient of the rainfall field; the modes correspond to the rain intensity state after the rainfall field is decomposed, and the main modes are those modes with high contribution rate to the energy of the rainfall field;

and step 3: screening out a mode representing the space-time attribute and a corresponding time coefficient according to the decomposition result of the step 2;

and 4, step 4: and (3) eliminating the modes which are screened out in the step (3) and can reflect the rainfall spatiotemporal characteristics from the main modes, and reconstructing the rainfall process of the position of the missing station by using the residual modes by taking the station which is closer to the position of the missing station as a template to obtain the rainfall data of the missing station.

Preferably, the rainfall acquisition method based on the intrinsic orthogonal decomposition at the rainfall station missing position provided by the invention can also have the following characteristics: in step 1, the acquired rainfall data has the following specific requirements: reconstructing rainfall data required by the rainfall station loss in a rainfall process for a long time (15 days and 7 days) into daily scale; reconstructing rainfall data required by the rainfall station loss position in the medium-long duration (3 days and 1 day) to be an hour scale; the short-duration (2 hours and 1 hour) rainfall station loss rainfall process reconstructs the needed rainfall data into a minute scale.

Preferably, the rainfall acquisition method based on the intrinsic orthogonal decomposition at the rainfall station missing position provided by the invention can also have the following characteristics: in step 2, performing intrinsic orthogonal decomposition on rainfall sample data of different sites to obtain main modes of a rainfall field and time coefficients corresponding to the modes, wherein the intrinsic orthogonal decomposition specifically operates as follows:

(1) the rainfall field is marked as X ═ v (X, t)¹)v(x，t²)...v(x，tⁿ)]∈R^m×nWherein X represents a rainfall field; t represents time, n is the maximum value of the time sequence, R is a real number field, and m is the number of stations;represents t after subtraction of the time average_n(ii) the intensity of rainfall at the moment; u (x, t)ⁿ) Indicating the rain intensity of the station at time n x,representing the average rain intensity of the x station at all times;

(2) singular value decomposition X-phi sigma psi is carried out on rainfall field X^TWhere phi is a mode matrix, and sigma is a mode energetic diagonal matrix, psi^TIs a time coefficient matrix;

selecting a dominant mode, i.e. a mode whose cumulative energy contribution rate can be approximated to 1Reconstructing a rainfall field:wherein j is a mode number, λ_jIs a modal characteristic value, whose value can reflect the modal energy contribution rate, a_j(t) is the time coefficient vector of the j-th mode, phi_j(x) Is a mode value vector of the jth mode.

Preferably, the rainfall acquisition method based on the intrinsic orthogonal decomposition at the rainfall station missing position provided by the invention can also have the following characteristics: in step 2, the mode with the accumulated energy contribution rate of more than 60% to the rainfall field is taken as the main mode.

Preferably, the rainfall acquisition method based on the intrinsic orthogonal decomposition at the rainfall station missing position provided by the invention can also have the following characteristics: in step 2, the main modes with the contribution rate of the accumulated energy of the rainfall field exceeding 80% are extracted to carry out rainfall type reconstruction.

Preferably, the rainfall acquisition method based on the intrinsic orthogonal decomposition at the rainfall station missing position provided by the invention can also have the following characteristics: in step 4, the rainfall process of the closest site is selected as a reconstruction template.

< System >

Further, the invention also provides a rainfall acquisition system based on intrinsic orthogonal decomposition at the loss of the rainfall station, which is characterized by comprising:

the acquisition part selects a plurality of adjacent rainfall stations according to the positions of the missing stations, collects and arranges rainfall data of the stations, and acquires original rainfall field data and rainfall field data after artificial space-time normalization;

the rainfall data orthogonal decomposition part is used for respectively inputting original rainfall field data and rainfall field data subjected to artificial space-time normalization and carrying out intrinsic orthogonal decomposition on the rainfall field twice to obtain a main mode and a corresponding time coefficient of the rainfall field; the modes correspond to the rain intensity state after the rainfall field is decomposed, and the main modes are those modes with high contribution rate to the energy of the rainfall field;

the screening part screens out a mode representing the space-time attribute and a corresponding time coefficient according to the decomposition result of the rainfall data orthogonal decomposition part;

the missing rainfall acquisition part is used for eliminating the mode which is screened out by the screening part and can reflect the rainfall spatio-temporal characteristics from the main modes, and reconstructing the rainfall process of the position of the missing station by using the station which is closer to the position of the missing station as a template and the residual modes to obtain the rainfall data of the missing station; and

and the control part is in communication connection with the acquisition part, the rainfall data orthogonal decomposition part, the screening part and the missing rainfall acquisition part and controls the operation of the acquisition part, the rainfall data orthogonal decomposition part, the screening part and the missing rainfall acquisition part.

Preferably, the fracture core permeation grouting real-time filling effect detection system provided by the invention further comprises: and the input display part is in communication connection with the acquisition part, the rainfall data orthogonal decomposition part, the screening part, the missing rainfall acquisition part and the control part and is used for enabling a user to input an operation instruction and performing corresponding display.

Action and Effect of the invention

According to the rainfall station missing rainfall acquisition method and system based on intrinsic orthogonal decomposition, original data and data after artificial space-time normalization are respectively used as input, intrinsic orthogonal decomposition is carried out, and a mode representing space-time attributes and a corresponding time coefficient are screened out; and reconstructing the rainfall process of the position of the missing station by using the rest main modes by taking the station closest to the position of the missing station as a template. The invention introduces intrinsic orthogonal decomposition into the rainfall process reconstruction at the missing station, effectively solves the error problem possibly caused by space-time difference in the direct transplanting process of the station based on a large amount of measured data of different adjacent stations, ensures that the deduced rainfall process is closer to the actual situation, can obtain a more accurate and reasonable calculation result with high precision, and provides a new idea for acquiring rainfall data at the missing rainfall station. And basic data can be provided for production confluence calculation, and the runoff process of the research area is further obtained.

Drawings

FIG. 1 is a flow chart of a rainfall acquisition method at the loss of a rainfall station based on intrinsic orthogonal decomposition according to the present invention;

FIG. 2 is a graph of modal and temporal coefficients prior to spatio-temporal normalization, as contemplated in an embodiment of the present invention;

FIG. 3 is a graph of the spatiotemporal normalized modal and temporal coefficients involved in an embodiment of the present invention;

FIG. 4 is a graph of modal energy contribution ratios involved in an embodiment of the present invention, with the cumulative contribution ratio of the first 7 dominant modes exceeding 75%;

fig. 5 is a diagram showing a comparison of rainfall process between simulation results and actual measurement results of different methods according to the embodiment of the present invention.

Detailed Description

The following describes in detail specific embodiments of a rainfall acquisition method and system based on intrinsic orthogonal decomposition at the absence of a rainfall station according to the present invention with reference to the accompanying drawings.

< example >

As shown in fig. 1, the method for acquiring rainfall at the missing part of the rainfall station based on intrinsic orthogonal decomposition provided by this embodiment includes the following steps:

step 1: assuming that the mountainous flood area (114.375 ° E,30.5 ° N) in wuhan city lacks rainfall data, 55 rainfall stations (station latitudes and longitudes of 109.375 ° E,30.5 ° N110 ° E,30 ° N110 ° E,30.5 ° N109.375 ° E,29.5 ° N108.75 ° E,30.5 ° N108.75 ° E,30 ° N109.375 ° E,30 ° N110.625 ° E,31.5 ° N110 ° E,31.5 ° N113.125 ° E,30.5 ° N115 ° E,30.5 ° N115.625 ° E,31 ° N115.625 ° E,30.5 ° N115.625 ° E,30 ° N115 ° E,30 ° N113.125 ° E,31 ° N112.5 ° E,30.5 ° N112.5 ° E,31.5 ° N113.125 ° E,30 ° N113.125 ° E, 365 ° N3932 ° E, 3632 ° N9632 ° E,33 ° N110 ° E,33 ° N111.25 ° E,33 ° N110 ° E,32.5 ° N110 ° E,32 ° N113.75 ° E,31.5 ° N113.125 ° E,32 ° N113.125 ° E,31.5 ° N114.375 ° E,31 ° N115 ° E,31 ° N113.75 ° E,29.5 ° N114.375 ° E,29.5 ° N114.375 ° E,30 ° N111.25 ° E,32 ° N111.25 ° E,31.5 ° N111.875 ° E,32.5 ° N111.875 ° E,31.5 ° N111.875 ° E,32 ° N112.5 ° E,32 ° N114.375 ° E,31.5 ° N111.875 ° E,31 ° N111.25 ° E,30.5 ° N4 ° E,30.5 ° N111.875 ° N114.375 ° E,31.5 ° N3675 ° N113.75 ° E), precipitation stations are subjected to the rainfall data;

step 2: respectively taking original rainfall field data and rainfall field data subjected to artificial space-time normalization as input, and performing intrinsic orthogonal decomposition on the rainfall field twice to obtain main modes of the rainfall field and time coefficients (shown in figures 2 and 3) corresponding to the modes; the modes correspond to the rain intensity state after the rainfall field decomposition, the main modes are those with larger contribution rate to the energy of the rainfall field, and the mode with the accumulated energy contribution rate exceeding 75% is selected as the main mode (fig. 4);

and step 3: and (3) comparing and screening the modes representing the space-time attributes and the corresponding time coefficients according to the decomposition results in the step (2), and comparing with a graph in fig. 2 and fig. 3 to obtain the main modes representing the space-time characteristics, namely the second mode, the third mode and the sixth mode.

And 4, step 4: and (3) removing the second, third and sixth modes from the main modes, taking the sites of the New continent area (115.0 degrees E and 31.0 degrees N) in Wuhan city as a template, and reconstructing the rainfall process of the positions of the missing sites by using the residual modes to obtain the rainfall data of the missing sites.

The nash efficiency coefficient (NSE) is generally used to verify the performance of the simulation result of the hydrological model, and is calculated as follows:

in the formula, Q_oAs an observed value, Q_mTo an analog value, Q^tThe value of the time t is the value of the time t,the average of the observations. The NSE value is negative infinity to 1, the NSE is close to 1, the simulation quality is good, and the model reliability is high; NSE is close to 0, which means that the simulation result is close to the average level of the observation value, namely the overall result is credible; NSE is much less than 0, the model is not trusted.

By calculation, the nash efficiency coefficient of the method of the conventional method of directly transplanting the rainfall process of the new continent area as the rainfall process of the flood area is-1.73, while the nash efficiency coefficient using the method of the present invention is 0.3 (fig. 5).

Further, the present embodiment also provides a system capable of automatically implementing the above process, where the system includes: the rainfall monitoring device comprises an acquisition unit, a rainfall data orthogonal decomposition unit, a screening unit, a missing rainfall acquisition unit, an input display unit and a control unit.

The acquisition part can select a plurality of adjacent rainfall stations according to the positions of the missing stations, collect and arrange rainfall data of the stations, and accordingly acquire original rainfall field data and rainfall field data after artificial space-time normalization.

The rainfall data orthogonal decomposition part respectively takes original rainfall field data and rainfall field data after artificial space-time normalization as input, and performs intrinsic orthogonal decomposition on the rainfall field twice to obtain a main mode and a corresponding time coefficient of the rainfall field; the modes correspond to the rain intensity state after the rainfall field is decomposed, and the main modes are those with high contribution rate to the energy of the rainfall field.

The screening part screens out the mode representing the space-time attribute and the corresponding time coefficient according to the decomposition result of the rainfall data orthogonal decomposition part.

The missing rainfall acquisition part eliminates the modes which are screened out by the screening part and can reflect the rainfall spatiotemporal characteristics from the main modes, takes the station which is closer to the position of the missing station as a template, and reconstructs the rainfall process of the position of the missing station by using the residual modes to obtain the rainfall data of the missing station.

The input display part is in communication connection with the acquisition part, the rainfall data orthogonal decomposition part, the screening part and the missing rainfall acquisition part and is used for enabling a user to input an operation instruction and performing corresponding display. For example, the input display unit may display the rainfall data at the missing site reconstructed by the missing rainfall acquisition unit.

And the control part is in communication connection with the acquisition part, the rainfall data orthogonal decomposition part, the screening part, the missing rainfall acquisition part and the input display part and controls the operation of the acquisition part, the rainfall data orthogonal decomposition part, the screening part, the missing rainfall acquisition part and the input display part.

The above embodiments are merely illustrative of the technical solutions of the present invention. The method and system for acquiring rainfall at the absence of rainfall station based on intrinsic orthogonal decomposition according to the present invention are not limited to the description in the above embodiments, but rather are subject to the scope defined by the claims. Any modification or supplement or equivalent replacement made by a person skilled in the art on the basis of this embodiment is within the scope of the invention as claimed in the claims.