A method for optimizing the layout of meteorological rainfall stations

Abstract

The present invention discloses a method for optimizing the layout of meteorological rain gauge stations, which belongs to the field of optimizing the layout of meteorological rain gauge stations. The method comprises the following steps: obtaining the existing station network layout; obtaining the first threshold, the second threshold and the third threshold respectively according to the existing station network layout; performing station extraction according to the second threshold and the third threshold to obtain a new station network layout and a second correlation coefficient set; judging whether the correlation coefficients in the second correlation coefficient set are all less than the first threshold, if so, completing the optimization of the layout of meteorological rain gauge stations, otherwise, updating the existing station network layout to the new station network layout, and performing a new round of station extraction. The present invention solves the problems of uneven layout of rain gauge network and redundant precipitation information.

Description

Technical Field

[0001] The invention belongs to the field of meteorological rainfall station site layout optimization, and in particular relates to a meteorological rainfall station site layout optimization method. Background Art

[0002] Precipitation is one of the important elements of meteorological station observation. Especially with the abnormal climate changes in recent years, the accurate monitoring of rainfall by meteorological stations is of great significance for weather forecasting, climate change, disaster prevention and mitigation, service decision-making, flood prevention and disaster relief, etc. The accuracy of rainfall monitoring depends on the scientific, reasonable and efficient station network layout. Although a dense station network density can more accurately monitor the precipitation and intensity changes at each point, the manpower, material and economic costs will also increase accordingly, which is easy to cause waste of resources and redundancy of rainfall information. Although a sparse station network density saves resources to a certain extent, it is easy to provide effective rainfall information for disaster prevention and mitigation and decision-making services due to insufficient detection capabilities.

[0003] In the existing technical solutions, some have proposed application scenarios of different methods by comparing the rationality of the cone method, the accumulation difference method, the rainstorm center control method and the pumping station method; some have introduced the independent and identically distributed central limit theorem into the planning of rainfall station networks, and studied the precipitation distribution in the plum rain area of ​​Anhui; some have used the cone method and the correlation coefficient method based on the spatiotemporal characteristics of historical rainfall, combined with the historical mountain torrent disasters and the needs of early warning and forecasting, to analyze the rationality of the water and rainfall station network in Shunchang County, Fujian Province, and provide an optimized layout plan; some have proposed a reasonable layout plan for the station network in Beijing urban area by combining the comparative analysis of multiple rainfall events using a combination of mathematical and statistical methods; some have studied the rationality of the layout of rainfall station networks in combination with topography, meteorology and other factors; some have studied the relationship between the layout of rainfall stations and the uncertainty of hydrological models, and explored the impact of different layout methods of rainfall station networks on the uncertainty of hydrological models; some have obtained the layout of the ground station network in East China through data assimilation sensitivity tests for typical weather cases of typhoons, plum rains and severe convection in East China. From the above research progress, it can be seen that although the current rainfall station network layout method has been widely used, there is still no unified and reasonable solution for the planning of rainfall station networks. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, a meteorological rain gauge station layout optimization method provided by the present invention solves the problems of uneven layout of rain gauge network and redundant precipitation information.

[0005] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a method for optimizing the layout of meteorological rainfall stations, comprising the following steps:

[0006] S1. Obtain the existing station network layout;

[0007] S2. According to the existing station network layout, a first threshold, a second threshold and a third threshold are obtained respectively;

[0008] S3, extracting stations according to the second threshold and the third threshold to obtain a new station network layout and a second correlation coefficient set;

[0009] S4. Determine whether the correlation coefficients in the second correlation coefficient set are all less than the first threshold value. If so, the optimization of the meteorological rainfall station site layout is completed. Otherwise, the existing station network layout is updated to the new station network layout, and return to step S3.

[0010] The beneficial effects of the present invention are as follows: based on the existing station network layout, the present invention solves the problem of redundant precipitation information caused by dense station layout in some areas through the control range of the meteorological rainfall station, the distance to the nearest station and precipitation correlation analysis, and takes "no loss of precipitation information" as the principle, and effectively realizes the economic, efficient and reasonable layout of the meteorological rainfall station network.

[0011] Furthermore, the step S2 is specifically as follows:

[0012] S201, using the Thiessen polygon method, respectively calculating the control area of ​​each meteorological rainfall station in the existing station network layout to obtain a first control area set;

[0013] S202, using a K-nearest neighbor algorithm, respectively calculating the neighboring stations of each meteorological rainfall station in the existing station network layout to obtain a first neighboring station set;

[0014] S203, according to the first adjacent station set, respectively calculating the distance between each meteorological rainfall station and the adjacent station in the existing station network layout to obtain a first distance set;

[0015] S204, obtaining and calculating the correlation coefficient between each meteorological rainfall station and the adjacent stations in the existing station network layout according to the precipitation data of each meteorological rainfall station in the existing station network layout, and obtaining a first correlation coefficient set:

[0016]

[0017] Among them, R xy is the correlation coefficient between the xth meteorological rainfall station and its neighboring stations in the first correlation coefficient set; P i is the i-th precipitation value of the x-th meteorological rainfall station; is the average precipitation of the xth meteorological rain gauge station; i is the precipitation value of the i-th station adjacent to the x-th meteorological rainfall station; is the average precipitation of the neighboring stations of the x-th meteorological rainfall station; i is the precipitation value number; N is the precipitation sequence;

[0018] S205. Obtain a first threshold, a second threshold, and a third threshold respectively according to the first control area set, the first distance set, and the first correlation coefficient set.

[0019] The beneficial effect of the above further scheme is: by calculating the control area of ​​each meteorological rainfall station in the existing station network layout, the distance between each meteorological rainfall station and the adjacent station in the existing station network layout, and the correlation coefficient between each meteorological rainfall station and the adjacent station in the existing station network layout, based on the actual relevant data, the first threshold, the second threshold and the third threshold are set respectively, which provides a cutoff condition for the method and indicates the optimization direction.

[0020] Furthermore, the step S205 is specifically as follows:

[0021] S2051. Establish a first linear equation according to the first distance set and the first correlation coefficient set:

[0022] R=aL+b

[0023] Where L is the distance between the station and the adjacent station; a is a constant; b is the distance coefficient; R is the correlation coefficient between the meteorological rainfall station and the adjacent station;

[0024] S2052. Establish a second linear equation based on the first control area set and the first correlation coefficient set:

[0025] R=cS+d

[0026] Where S is the control area of ​​the site; c is a constant; d is the area coefficient;

[0027] S2053, taking the median of the first correlation coefficient set as a first threshold;

[0028] S2054, using the first threshold as the correlation coefficient between the meteorological rainfall station and the adjacent station, substituting it into the first linear equation to solve the distance between the station and the adjacent station, and obtaining the second threshold;

[0029] S2055. Substituting the first threshold as the correlation coefficient between the meteorological rainfall station and the adjacent stations into the second linear equation to solve the control area of ​​the station, thereby obtaining the third threshold.

[0030] The beneficial effects of the above further scheme are: establishing the first linear equation and the second linear equation respectively, quantifying the relationship between the distance between the station and the adjacent station and the correlation coefficient between the meteorological rainfall station and the adjacent station, as well as the control area of ​​the station and the correlation coefficient between the meteorological rainfall station and the adjacent station, so that the setting of the second threshold and the third threshold is more reasonable and more in line with the actual situation, thereby improving the optimization effect of the layout of the meteorological rainfall station.

[0031] Furthermore, the step S3 is specifically as follows:

[0032] S301, using the Thiessen polygon method, respectively calculating the control area of ​​each meteorological rainfall station in the existing station network layout to obtain a second control area set;

[0033] S302, using the K-nearest neighbor algorithm, respectively calculating the neighboring stations of each meteorological rainfall station in the existing station network layout to obtain a second neighboring station set;

[0034] S303, according to the second adjacent station set, respectively calculating the distance between each meteorological rainfall station and the adjacent station in the existing station network layout to obtain a second distance set;

[0035] S304, obtaining and calculating the correlation coefficient between each meteorological rainfall station and the adjacent station in the existing station network layout according to the precipitation data of each meteorological rainfall station in the existing station network layout, and obtaining a second correlation coefficient set;

[0036] S305, obtaining the rainfall station to be deleted according to the second control area set, the second distance set and the second correlation coefficient set;

[0037] S306. Eliminate the rainfall stations to be deleted from the existing station network layout to obtain a new station network layout.

[0038] The beneficial effect of the above further scheme is: station extraction based on the existing station network layout solves the problem of redundant precipitation information caused by dense station layout in some areas, and effectively realizes the economic, efficient and reasonable layout of meteorological rainfall station network.

[0039] Furthermore, the step S305 is specifically as follows:

[0040] S3051. According to the second control area set, obtain meteorological rainfall station sites whose control areas are smaller than a third threshold value to obtain a first rainfall station site set;

[0041] S3052. According to the second distance set, obtain the meteorological rainfall station sites in the first rainfall station site set whose distances to the adjacent stations are less than the second threshold value, to obtain the second rainfall station site set;

[0042] S3053, sorting the rainfall stations in the second rainfall station site set in descending order based on the second correlation coefficient set to obtain a rainfall station site sorting table;

[0043] S3054. Set the number of stations to be drawn at a time, and select meteorological rain gauge stations in turn according to the rain gauge station ranking table and the number of stations to be drawn at a time, to obtain the rain gauge stations to be deleted.

[0044] The beneficial effect of the above further scheme is: by setting the number of single station extraction and performing station extraction in small quantities and multiple times, the adjacent stations of each meteorological rain gauge station, the control area of ​​each meteorological rain gauge station, the distance between each meteorological rain gauge station and the adjacent station, and the correlation coefficient between each meteorological rain gauge station and the adjacent station can be recalculated according to the results of each station extraction, so that the layout optimization direction can be continuously adjusted to improve the optimization effect.

[0045] Furthermore, the number of rainfall stations to be deleted in step S3054 is the same as the number of stations extracted in a single time, and the meteorological rainfall stations in the rainfall stations to be deleted are not adjacent to each other.

[0046] The beneficial effects of the above further scheme are: by setting the number of single sampling stations, sampling stations are carried out in small quantities and multiple times, so that the layout optimization direction is constantly adjusted to improve the optimization effect; and in the single sampling results, the meteorological rainfall stations are not adjacent stations to each other, avoiding accidental deletion of stations and affecting the optimization effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The present invention is a flow chart of the method.

[0048] Figure 2 This is the original station network layout diagram of Ya'an in the embodiment of the present invention.

[0049] Figure 3 It is a box plot of the controlled area S, the distance L to the nearest station and the correlation coefficient R of each rainfall station in Ya'an in the embodiment of the present invention.

[0050] Figure 4 Schematic diagram of the relationship between the correlation coefficient between the nearest stations in Ya'an and the distance between stations in an embodiment of the present invention.

[0051] Figure 5 Schematic diagram of the relationship between the correlation coefficient between the nearest sites in Ya'an and the site control area in an embodiment of the present invention.

[0052] Figure 6 This is a diagram showing the surface rainfall error results of the Ya'an rain gauge network after each pumping and the original network in an embodiment of the present invention.

[0053] Figure 7 Schematic diagram of the spatial distribution of daily precipitation in Ya'an based on six rainfall processes of the original station network in an embodiment of the present invention.

[0054] Figure 8 This is a schematic diagram of the spatial distribution of daily precipitation in six rainfall processes after Ya'an conducted 11 pumping stations based on the present invention in an embodiment of the present invention.

[0055] Fig. 9 Schematic diagram of the change of absolute error of surface rainfall during six rainfall processes in Ya'an in an embodiment of the present invention.

[0056] Fig.10 Schematic diagram of the relative error change of surface rainfall during six rainfall processes in Ya'an in an embodiment of the present invention.

[0057] Fig.11 This is a station network layout diagram after the Ya'an pumping station is completed in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0059] like Figure 1 As shown, in one embodiment of the present invention, a method for optimizing the layout of meteorological rainfall stations includes the following steps:

[0060] S1. Obtain the existing station network layout;

[0061] S2. According to the existing station network layout, a first threshold, a second threshold and a third threshold are obtained respectively;

[0062] S3, extracting stations according to the second threshold and the third threshold to obtain a new station network layout and a second correlation coefficient set;

[0063] S4. Determine whether the correlation coefficients in the second correlation coefficient set are all less than the first threshold value. If so, the optimization of the meteorological rainfall station site layout is completed. Otherwise, the existing station network layout is updated to the new station network layout, and return to step S3.

[0064] In this embodiment, taking Ya'an as an example, Figure 2 As shown, before 2019, there were 265 rain gauges in Ya'an.

[0065] The step S2 is specifically as follows:

[0066] S201, using the Thiessen polygon method, respectively calculating the control area of ​​each meteorological rainfall station in the existing station network layout to obtain a first control area set;

[0067] S202, using a K-nearest neighbor algorithm, respectively calculating the neighboring stations of each meteorological rainfall station in the existing station network layout to obtain a first neighboring station set;

[0068] S203, according to the first adjacent station set, respectively calculating the distance between each meteorological rainfall station and the adjacent station in the existing station network layout to obtain a first distance set;

[0069] S204, obtaining and calculating the correlation coefficient between each meteorological rainfall station and the adjacent stations in the existing station network layout according to the precipitation data of each meteorological rainfall station in the existing station network layout, and obtaining a first correlation coefficient set:

[0070]

[0071] Among them, R xy is the correlation coefficient between the xth meteorological rainfall station and its neighboring stations in the first correlation coefficient set; P i is the i-th precipitation value of the x-th meteorological rainfall station; is the average precipitation of the xth meteorological rain gauge station; i is the precipitation value of the i-th station adjacent to the x-th meteorological rainfall station; is the average precipitation of the neighboring stations of the x-th meteorological rainfall station; i is the precipitation value number; N is the precipitation sequence;

[0072] S205. Obtain a first threshold, a second threshold, and a third threshold respectively according to the first control area set, the first distance set, and the first correlation coefficient set.

[0073] In this embodiment, Figure 3 As shown, based on the correlation coefficient set between the Ya'an rainfall station and the nearest station, the first threshold is determined to be 0.9.

[0074] The step S205 is specifically as follows:

[0075] S2051. Establish a first linear equation according to the first distance set and the first correlation coefficient set:

[0076] R=aL+b

[0077] Where L is the distance between the station and the adjacent station; a is a constant; b is the distance coefficient; R is the correlation coefficient between the meteorological rainfall station and the adjacent station;

[0078] S2052. Establish a second linear equation based on the first control area set and the first correlation coefficient set:

[0079] R=cS+d

[0080] Where S is the control area of ​​the site; c is a constant; d is the area coefficient;

[0081] S2053, taking the median of the first correlation coefficient set as a first threshold;

[0082] S2054, using the first threshold as the correlation coefficient between the meteorological rainfall station and the adjacent station, substituting it into the first linear equation to solve the distance between the station and the adjacent station, and obtaining the second threshold;

[0083] S2055. Substituting the first threshold as the correlation coefficient between the meteorological rainfall station and the adjacent stations into the second linear equation to solve the control area of ​​the station, thereby obtaining the third threshold.

[0084] In this embodiment, Figure 4 and Figure 5 As shown, the first linear equation is R = -0.0111L + 0.935; the second linear equation is R = -0.0003S + 0.9132; the second threshold value obtained according to the first linear equation and the second linear equation is L = 3.13km; the third threshold value is S = 44km 2 .

[0085] The step S3 is specifically as follows:

[0086] S301, using the Thiessen polygon method, respectively calculating the control area of ​​each meteorological rainfall station in the existing station network layout to obtain a second control area set;

[0087] S302, using the K-nearest neighbor algorithm, respectively calculating the neighboring stations of each meteorological rainfall station in the existing station network layout to obtain a second neighboring station set;

[0088] S303, according to the second adjacent station set, respectively calculating the distance between each meteorological rainfall station and the adjacent station in the existing station network layout to obtain a second distance set;

[0089] S304, obtaining and calculating the correlation coefficient between each meteorological rainfall station and the adjacent station in the existing station network layout according to the precipitation data of each meteorological rainfall station in the existing station network layout, and obtaining a second correlation coefficient set;

[0090] S305, obtaining the rainfall station to be deleted according to the second control area set, the second distance set and the second correlation coefficient set;

[0091] S306. Eliminate the rainfall stations to be deleted from the existing station network layout to obtain a new station network layout.

[0092] The step S305 is specifically as follows:

[0093] S3051. According to the second control area set, obtain meteorological rainfall station sites whose control areas are smaller than a third threshold value to obtain a first rainfall station site set;

[0094] S3052. According to the second distance set, obtain the meteorological rainfall station sites in the first rainfall station site set whose distances to the adjacent stations are less than the second threshold value, to obtain the second rainfall station site set;

[0095] S3053, sorting the rainfall stations in the second rainfall station site set in descending order based on the second correlation coefficient set to obtain a rainfall station site sorting table;

[0096] S3054. Set the number of stations to be drawn at a time, and select meteorological rain gauge stations in turn according to the rain gauge station ranking table and the number of stations to be drawn at a time, to obtain the rain gauge stations to be deleted.

[0097] The number of rainfall stations to be deleted in step S3054 is the same as the number of stations extracted in a single operation, and the meteorological rainfall stations among the rainfall stations to be deleted are not adjacent to each other.

[0098] In this embodiment, the first control area set, the first distance set and the first correlation coefficient set are respectively the same as the second control area set, the second distance set and the second correlation coefficient set before the first pumping station. The second control area set, the second distance set and the second correlation coefficient set after the pumping station are changing, and the second control area set, the second distance set and the second correlation coefficient set at this time are not the same as the first control area set, the first distance set and the first correlation coefficient set. The calculation of the first control area set, the first distance set and the first correlation coefficient set is to determine the first threshold, the second threshold and the third threshold.

[0099] In this embodiment, after the stations are drawn one by one, the number of stations with high correlation gradually decreases, indicating that the redundant information between stations is gradually reduced, and the distance between stations and the control area of ​​stations are gradually increased, indicating that the distribution of stations tends to be uniform. In terms of mutual information and station distribution, the station network layout tends to be reasonable. The statistical results of the station control area, the distance between adjacent stations and the correlation coefficient before and after the Ya'an rainfall station is drawn are shown in Table 1.

[0100] Table 1

[0101]

[0102] The relative error of the network rainfall after each pumping station and the original network rainfall, whether it is the monthly rainfall or the daily rainfall, is mostly within 1%, and the absolute error is also within 0.3mm (the specific results are shown in Figure 6 As shown in Figure 2 ). From the changes in surface rainfall during the six precipitation processes in Ya'an, the rainfall areas and rainfall ranges of various magnitudes did not change much after each pumping station, and the heavy rainfall center and area could be captured (the specific results are shown in Figure 2 ). Figure 7 and Figure 8 The average absolute errors after each extraction station were 0.12mm, 0.05mm, 0.16mm, 0.05mm, 0.06mm, and 0.04mm, respectively, and the relative errors were mostly within 0.1% (the specific results are shown in Fig. 9 and Fig.10 As shown). In general, after each pumping, the precipitation area and the range of precipitation levels are basically consistent, with small errors. The station network has good detection and capture capabilities for precipitation, indicating that the station network layout optimization method proposed in the present invention is more reasonable.

[0103] In this embodiment, the Thiessen polygon distribution diagram of the rainfall station after the station extraction is completed is as follows: Fig.11 As shown, according to the original station network in Ya'an, station extraction was carried out based on the present invention to obtain the final station network layout diagram, and a total of 11 station extractions were carried out.

[0104] The basic idea of ​​the present invention is to solve the problem of redundant precipitation information in some areas due to dense station layout, effectively realize the economic, efficient and reasonable layout of meteorological rainfall station network, based on the control range of the station, the distance between the nearest stations and precipitation correlation analysis, with the principle of "no loss of precipitation information".

Claims

1. A method for optimizing the layout of meteorological rainfall stations, characterized in that: The following steps are involved: S1. Obtain the existing station network layout; S2. According to the existing station network layout, a first threshold, a second threshold and a third threshold are obtained respectively; the step S2 is specifically as follows: S201, using the Thiessen polygon method, respectively calculating the control area of ​​each meteorological rainfall station in the existing station network layout to obtain a first control area set; S202, using a K-nearest neighbor algorithm, respectively calculating the neighboring stations of each meteorological rainfall station in the existing station network layout to obtain a first neighboring station set; S203, according to the first adjacent station set, respectively calculating the distance between each meteorological rainfall station and the adjacent station in the existing station network layout to obtain a first distance set; S204, obtaining and calculating the correlation coefficient between each meteorological rainfall station and the adjacent stations in the existing station network layout according to the precipitation data of each meteorological rainfall station in the existing station network layout, and obtaining a first correlation coefficient set: in, is the first correlation coefficient set The correlation coefficient between a meteorological rainfall station and its neighboring stations; For the The first meteorological rainfall station precipitation value; For the The average precipitation at the meteorological rain gauge station; For the The neighboring station of the meteorological rainfall station precipitation value; For the The average precipitation of the neighboring stations of a meteorological rain gauge station; Number the precipitation values; is the precipitation series; S205, obtaining a first threshold, a second threshold and a third threshold respectively according to the first control area set, the first distance set and the first correlation coefficient set; the step S205 is specifically as follows: S2051. Establish a first linear equation according to the first distance set and the first correlation coefficient set: in, is the distance between the station and the adjacent station; is a constant; is the distance coefficient; is the correlation coefficient between the meteorological rainfall station and the adjacent stations; S2052. Establish a second linear equation based on the first control area set and the first correlation coefficient set: in, is the control area of ​​the site; is a constant; is the area coefficient; S2053, taking the median of the first correlation coefficient set as a first threshold; S2054, using the first threshold as the correlation coefficient between the meteorological rainfall station and the adjacent station, substituting it into the first linear equation to solve the distance between the station and the adjacent station, and obtaining the second threshold; S2055, using the first threshold as the correlation coefficient between the meteorological rainfall station and the adjacent station, substituting it into the second linear equation to solve the control area of ​​the station, and obtaining the third threshold; S3, extracting stations according to the second threshold and the third threshold to obtain a new station network layout and a second correlation coefficient set; the step S3 is specifically as follows: S301, using the Thiessen polygon method, respectively calculating the control area of ​​each meteorological rainfall station in the existing station network layout to obtain a second control area set; S302, using the K-nearest neighbor algorithm, respectively calculating the neighboring stations of each meteorological rainfall station in the existing station network layout to obtain a second neighboring station set; S303, according to the second adjacent station set, respectively calculating the distance between each meteorological rainfall station and the adjacent station in the existing station network layout to obtain a second distance set; S304, obtaining and calculating the correlation coefficient between each meteorological rainfall station and the adjacent station in the existing station network layout according to the precipitation data of each meteorological rainfall station in the existing station network layout, and obtaining a second correlation coefficient set; S305, obtaining the rainfall station to be deleted according to the second control area set, the second distance set and the second correlation coefficient set; S306, removing the rainfall stations to be deleted from the existing station network layout to obtain a new station network layout; S4. Determine whether the correlation coefficients in the second correlation coefficient set are all less than the first threshold value. If so, the optimization of the meteorological rainfall station site layout is completed. Otherwise, the existing station network layout is updated to the new station network layout, and return to step S3.

2. The method for optimizing the layout of meteorological rainfall stations according to claim 1, characterized in that: The step S305 is specifically as follows: S3051. According to the second control area set, obtain meteorological rainfall station sites whose control areas are smaller than a third threshold value to obtain a first rainfall station site set; S3052. According to the second distance set, obtain the meteorological rainfall station sites in the first rainfall station site set whose distances to the adjacent stations are less than the second threshold value, to obtain the second rainfall station site set; S3053, sorting the rainfall stations in the second rainfall station site set in descending order based on the second correlation coefficient set to obtain a rainfall station site sorting table; S3054. Set the number of stations to be drawn at a time, and select meteorological rain gauge stations in turn according to the rain gauge station ranking table and the number of stations to be drawn at a time, to obtain the rain gauge stations to be deleted.

3. The method for optimizing the layout of meteorological rainfall stations according to claim 2, characterized in that: The number of rainfall stations to be deleted in step S3054 is the same as the number of stations extracted in a single operation, and the meteorological rainfall stations among the rainfall stations to be deleted are not adjacent to each other.

Images