Irrigation prescription map partition extraction method based on irrigation node Beidou information

Abstract

The invention discloses an irrigation prescription map partition extraction method based on irrigation node Beidou information. The method comprises the following steps: obtaining a known large-scale irrigation district water demand diagram to be researched and a known large-scale irrigation district water shortage index diagram to be researched in an irrigation district to be researched; beidou positioning information of end domain irrigation terminal nodes of an irrigation area to be researched is acquired, and a time-space database of all the end domain irrigation terminal nodes of the irrigation area to be researched is established; according to the space-time database, the to-be-researched irrigation area is divided into a plurality of partitions by adopting fuzzy C-means clustering, and nodes with the membership degree larger than or equal to a set threshold value in each partition serve as partition initial boundaries; smoothing the initial boundary of each partition to obtain the final partition boundary of each partition; determining the water demand relationship of the subareas based on the water demand diagram of the irrigation area to be researched; according to the water demand diagram and the water shortage index diagram of the irrigation area to be researched and the water demand relation of the subareas, the gross irrigation quota of the subareas is calculated to serve as the irrigation prescription diagram of the subareas.

Description

Technical Field

[0001] This invention relates to the field of smart agricultural irrigation technology, specifically to a method for extracting irrigation prescription maps based on BeiDou information from irrigation nodes. Background Technology

[0002] As China's smart irrigation technology matures, irrigation prescription maps are needed to represent irrigation quotas and serve as a crucial link between air-space-ground sensing of water shortage information and irrigation decisions. Irrigation prescription maps, as a key technological carrier for precision agricultural water management, refer to spatially differentiated irrigation decision guidance maps generated through multi-source data fusion and intelligent analysis. The essence of smart irrigation is precise dynamic irrigation, and the prerequisite for precise dynamic irrigation is field-specific, demand-based irrigation. Therefore, small-scale irrigation water demand in the field is a prerequisite for realizing smart irrigation. Typically, irrigation prescription maps refer to the water demand of the irrigation area under study calculated by combining water shortage information obtained through remote sensing, drones, soil moisture sensors, etc., with the principle of water balance. Conventional irrigation prescription maps are often too large-scale to meet the needs of precision irrigation.

[0003] Therefore, irrigation prescription maps that do not define boundaries and zoning irrigation zones are difficult to adapt to the requirements of smart irrigation and precision irrigation. Summary of the Invention

[0004] To address the aforementioned shortcomings in existing technologies, the irrigation prescription map partitioning extraction method based on BeiDou information of irrigation nodes provided by this invention solves the problem of low irrigation accuracy caused by the excessively large scale of conventional irrigation prescription maps.

[0005] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: A method for extracting irrigation prescription maps based on BeiDou information from irrigation nodes is provided, comprising the following steps: S1. Obtain the known large-scale water demand map and water shortage index map of the irrigation area to be studied; S2. Obtain the BeiDou positioning information of the irrigation terminal nodes in the irrigation area to be studied, and establish a spatiotemporal database of all irrigation terminal nodes in the irrigation area to be studied. S3. Based on the spatiotemporal database, fuzzy C-means clustering is used to divide the irrigation area under study into several partitions, and the nodes in each partition with a membership degree greater than or equal to a set threshold are used as the initial boundaries of the partitions. S4. Smooth the initial boundary of each partition to obtain the final partition boundary of each partition; then determine the water demand relationship of the partition based on the water demand map of the irrigation area to be studied. S5. Based on the water demand map and water shortage index map of the irrigation area to be studied and the water demand relationship of the sub-districts, calculate the gross irrigation quota of the sub-districts as the irrigation prescription map of the sub-districts.

[0006] Furthermore, methods for smoothing the initial boundaries of each partition include: S41. Based on the boundary points of each partition, the weighting coefficient λ is calculated by minimizing the energy objective function. C , λ L , λ K Energy objective function The expression is: , , in, For length items; For constraint terms; λ is the curvature term; C , λ L , λ K is the weighting coefficient; n is the total number of boundary nodes in the partition; and b is the i-th and i-1-th optimized boundary point on the partition boundary; i Let i be the i-th unoptimized original boundary point on the partition boundary; To represent Euclidean distance; S42. Based on the weighting coefficient λ C , λ L , λ K Calculate the smoothed boundary points for each partition boundary: Among them, A L A is a tridiagonal matrix; C A is a diagonal matrix; K It is a pentagonal matrix; The set of boundary points after partitioning and smoothing. , and These are the 1st, 2nd, and kth boundary points in the boundary point set S, respectively. The set of boundary points before partitioning is smoothed. , and These are the 1st, 2nd, and kth boundary points in the boundary point set B, respectively.

[0007] Furthermore, the method for determining the water demand relationship for each zone is as follows: Determine whether the water supply and water demand of the irrigation district under study are met. If so, the water demand relationship of the zone is a fully irrigated scenario; otherwise, the water demand relationship of the zone is a non-fully irrigated scenario.

[0008] Furthermore, when the water demand relationship of a zone is a fully irrigated scenario, the method for calculating the irrigation prescription map of the zone includes: S511. Calculate the net irrigation quota corresponding to the irrigation zone: in, The net irrigation quota corresponding to the zone; X represents the irrigation water volume corresponding to the irrigation zone; X and Y represent the spatial coordinates of the terminal nodes of the irrigation zone. max and X min The upper and lower limits of X and Y are respectively. max and Y min These are the upper and lower limits of Y, respectively; The irrigation quota corresponding to the spatial coordinate range; This represents the area within the coordinates of the spatial range. S512. Calculate the regional gross irrigation quota as its irrigation prescription map: in, For regional gross irrigation quotas; This refers to the irrigation water utilization coefficient. The irrigation guarantee coefficient is the factor used under fully irrigated conditions. ; When the water demand relationship of a zone is a non-fully irrigated scenario, the methods for calculating the irrigation prescription map of the zone include: S521. Determine the irrigation guarantee factor based on the water shortage index diagram: Where θ is the mean of the water shortage index for the region; θ min Let θ be the minimum value of θ within the irrigation area under study; 0.8 ≤ θ ≤ 1 represents a severely water-deficient area within the irrigation area under study; 0.5 ≤ θ ≤ 1 represents a moderately water-deficient area within the irrigation area under study; θ min ≤θ≤0.5 indicates a slightly water-deficient area within the irrigation district under study; S522. Calculate the initial value of the gross irrigation quota based on the irrigation guarantee coefficient. : ; S523. Calculate the total water demand of the irrigation area under study based on the gross irrigation quota. : in, This is the initial value for the gross irrigation quota of the j-th partition; This represents the total number of partitions. S524. Based on the water demand of the irrigation area under study With water supply Calculate the water distribution coefficient : ; S625, based on water distribution coefficient The final gross irrigation quota for each zone is calculated as its irrigation prescription map: .

[0009] Preferably, step S3 further includes: S31. Based on the spatiotemporal database, extract the latitude and longitude coordinates of all irrigation terminal nodes in the end-domain area to form a point set. : in, Let be the longitude and latitude of the i-th terminal irrigation node; , and Point sets The first, second and qth end-domain irrigation terminal nodes are defined, where q is the total number of end-domain irrigation terminal nodes in the irrigation district to be studied. S32. The objective function is optimized by using the fuzzy C-means clustering method to assign q nodes to C cluster centers, resulting in several partitions; S33. Extract the membership degree of all end-domain irrigation terminal nodes to cluster j from the membership degree matrix u. For each partition j, construct a membership degree surface based on the membership degree value. S34. For each partition j, its boundary is defined by a membership degree equal to a set threshold. The contour lines, i.e., the boundaries of each partition, are: in, Let j be the boundary of the j-th partition; The coordinates of the irrigation terminal nodes in the partition; for The degree of membership.

[0010] Furthermore, construct membership surfaces. The expression is: Where p is the coordinate of any point; φ() is the Gaussian function, expressed as: The coefficient ε = 0.5; To represent Euclidean distance; the coefficient λ i a, b, and c are determined by solving the following system of equations: in, Let be the membership degree of terminal node i in the irrigation domain to cluster j.

[0011] Furthermore, step S32 further includes: S321. Initialize the membership degree of each irrigation terminal node in the end domain; S322, Update the membership degree of terminal node i to cluster j in the irrigation domain: Among them, v j v is the j-th cluster center; k Let be the k-th cluster center; c is the number of cluster centers; m is the fuzzy index; Euclidean distance; S323. Update the cluster centers based on the updated membership degrees: ; S324. Determine whether the cluster centers before and after the update meet the termination condition. If yes, complete the partitioning; otherwise, return to step S322. The termination condition is: in, This is the termination threshold.

[0012] Furthermore, the expression for the objective function J in step S32 is: in, ; denoted by Euclidean distance; c represents the total number of cluster centers.

[0013] The beneficial effects of this invention are as follows: This scheme establishes a spatiotemporal database of irrigation terminal nodes using BeiDou positioning information of the terminal nodes in the irrigation field. Then, it uses fuzzy C-means clustering (FCM) combined with a set threshold to extract the boundaries. In this way, the irrigation area under study can be divided into multiple small zones. After smoothing, the terminal nodes of the irrigation field at the zone boundaries can be accurately determined to ensure accurate division of the zones. Then, the irrigation guarantee coefficient is determined based on the water shortage index map, thereby ensuring accurate calculation of the irrigation quota for different water supply and demand conditions in the zones. Finally, an irrigation prescription map suitable for field rotation irrigation grouping under small-scale conditions is established to ensure irrigation accuracy. Attached Figure Description

[0014] Figure 1 This is a flowchart of a method for extracting irrigation prescription maps based on BeiDou information from irrigation nodes. Detailed Implementation

[0015] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0016] refer to Figure 1 , Figure 1 A flowchart illustrating a method for extracting irrigation prescription maps based on BeiDou information from irrigation nodes is shown; for example... Figure 1 As shown, the method S includes steps S1 to S5.

[0017] In step S1, the known large-scale water demand map and water shortage index map of the irrigation area to be studied are obtained. The water demand map of the irrigation area is specifically based on irrigation big data such as soil, meteorology, crops, and farmland water supply, and is obtained by using remote sensing or UAVs. It should also include the water shortage index map so that step S5 can determine the irrigation guarantee coefficient for insufficient irrigation. In step S2, the BeiDou positioning information of the irrigation terminal nodes in the irrigation area to be studied is obtained, and a spatiotemporal database of all irrigation terminal nodes in the irrigation area to be studied is established; the irrigation terminal nodes are gate controllers, valve controllers, etc.

[0018] The spatiotemporal database enables efficient spatial querying and temporal tracing, providing a data foundation for subsequent boundary extraction and irrigation decisions. The spatiotemporal database should include information such as the node ID, latitude and longitude, deployment time, and equipment type of each irrigation terminal node in the end domain. A diagram can be found in Table 1.

[0019] Table 1 shows the information on the terminal nodes of the irrigation system. In step S3, based on the spatiotemporal database, fuzzy C-means clustering is used to divide the irrigation area under study into several partitions, and nodes with a membership degree greater than or equal to a set threshold in each partition are used as the initial boundaries of the partitions. S31. Based on the spatiotemporal database, extract the latitude and longitude coordinates of all irrigation terminal nodes in the end-domain area to form a point set. : in, Let be the longitude and latitude of the i-th terminal irrigation node; , and Point sets The first, second and qth end-domain irrigation terminal nodes are defined, where q is the total number of end-domain irrigation terminal nodes in the irrigation district to be studied. S32. The objective function is optimized using fuzzy C-means clustering to assign q nodes to C cluster centers, resulting in several partitions. The expression for the objective function J in this step is: in, ; denoted by Euclidean distance; c represents the total number of cluster centers.

[0020] In one embodiment of the present invention, step S32 further includes: S321. Initialize the membership degree of each irrigation terminal node in the end domain; S322, Update the membership degree of terminal node i to cluster j in the irrigation domain: Among them, v j v is the j-th cluster center; k Let be the k-th cluster center; c is the number of cluster centers; m is the fuzzy index; Euclidean distance; S323. Update the cluster centers based on the updated membership degrees: ; S324. Determine whether the cluster centers before and after the update meet the termination condition. If yes, complete the partitioning; otherwise, return to step S322. The termination condition is: in, and These are the cluster centers before and after the update of the j-th partition, respectively; This is the termination threshold.

[0021] S33. Extract the membership degree of all end-domain irrigation terminal nodes to cluster j from the membership degree matrix u. For each partition j, construct a membership degree surface based on the membership degree value. During implementation, this scheme preferably constructs membership surfaces. The expression is: Where p is the coordinate of any point; φ() is the Gaussian function, expressed as: The coefficient ε = 0.5; To represent Euclidean distance; the coefficient λ i a, b, and c are determined by solving the following system of equations: in, Let be the membership degree of terminal node i in the irrigation domain to cluster j.

[0022] S34. For each partition j, its boundary is defined by a membership degree equal to a set threshold. The contour lines, i.e., the boundaries of each partition, are: in, Let j be the boundary of the j-th partition; The coordinates of the irrigation terminal nodes in the partition; for The degree of membership.

[0023] In step S4, the initial boundary of each partition is smoothed to obtain the final partition boundary of each partition; then, the water demand relationship of the partition is determined based on the water demand map of the irrigation area to be studied. In implementation, the preferred method for smoothing the initial boundaries of each partition in this scheme includes: S41. Based on the boundary points of each partition, the weighting coefficient λ is calculated by minimizing the energy objective function. C , λ L , λ K Energy objective function The expression is: , , in, For length items; For constraint terms; λ is the curvature term; C , λ L , λ K is the weighting coefficient; n is the total number of boundary nodes in the partition; and b is the i-th and i-1-th optimized boundary point on the partition boundary; i Let i be the i-th unoptimized original boundary point on the partition boundary; To represent Euclidean distance; S42. Based on the weighting coefficient λ C , λ L , λ K Calculate the smoothed boundary points for each partition boundary: Among them, A L A is a tridiagonal matrix; C A is a diagonal matrix; K It is a pentagonal matrix; The set of boundary points after partitioning and smoothing. , and These are the 1st, 2nd, and kth boundary points in the boundary point set S, respectively. The set of boundary points before partitioning is smoothed. , and These are the 1st, 2nd, and kth boundary points in the boundary point set B, respectively.

[0024] This solution uses the above method for smoothing, aiming to reduce noise and unnecessary details while maintaining the overall shape of the boundary, and ensuring that the boundary is smooth and conforms to geometric rules.

[0025] During implementation, the preferred method for determining the water demand relationship for each zone in this scheme is as follows: Determine whether the water supply and water demand of the irrigation district under study are met. If so, the water demand relationship of the zone is a fully irrigated scenario; otherwise, the water demand relationship of the zone is a non-fully irrigated scenario.

[0026] In step S5, based on the water demand map and water shortage index map of the irrigation area to be studied and the water demand relationship of the sub-areas, the gross irrigation quota of the sub-area is calculated as the irrigation prescription map of the sub-area.

[0027] In one embodiment of the present invention, when the water demand relationship of a zone is a fully irrigated scenario, the method for calculating the irrigation prescription map of the zone includes: S511. Calculate the net irrigation quota corresponding to the irrigation zone: in, The net irrigation quota corresponding to the zone; X represents the irrigation water volume corresponding to the irrigation zone; X and Y represent the spatial coordinates of the terminal nodes of the irrigation zone. max and X min The upper and lower limits of X and Y are respectively. max and Y min These are the upper and lower limits of Y, respectively; The irrigation quota corresponding to the spatial coordinate range; This represents the area within the coordinates of the spatial range. S512. Calculate the regional gross irrigation quota as its irrigation prescription map: in, For regional gross irrigation quotas; This refers to the irrigation water utilization coefficient. The irrigation guarantee coefficient is the factor used under fully irrigated conditions. ; When the water demand relationship of a zone is a non-fully irrigated scenario, the methods for calculating the irrigation prescription map of the zone include: S521. Determine the irrigation guarantee factor based on the water shortage index diagram: Where θ is the mean of the water shortage index for the region; θ min Let θ be the minimum value of θ within the irrigation area under study; 0.8 ≤ θ ≤ 1 represents a severely water-deficient area within the irrigation area under study; 0.5 ≤ θ ≤ 1 represents a moderately water-deficient area within the irrigation area under study; θ min ≤θ≤0.5 indicates a slightly water-deficient area within the irrigation district under study; S522. Calculate the initial value of the gross irrigation quota based on the irrigation guarantee coefficient. : ; S523. Calculate the total water demand of the irrigation area under study based on the gross irrigation quota. : in, This is the initial value for the gross irrigation quota of the j-th partition; This represents the total number of partitions. S524. Based on the water demand of the irrigation area under study With water supply Calculate the water distribution coefficient : ; S625, based on water distribution coefficient The final gross irrigation quota for each zone is calculated as its irrigation prescription map: .

[0028] In summary, this solution addresses the problem of low irrigation accuracy caused by the excessively large scale of conventional irrigation prescription maps, and establishes an intelligent zoning method based on BeiDou satellite positioning and fuzzy pattern recognition, which can provide technical support for variable-based precise irrigation decisions.

Claims

1. A method for extracting irrigation prescription maps based on BeiDou information from irrigation nodes, characterized in that, Including the following steps: S1. Obtain the known large-scale water demand map and water shortage index map of the irrigation area to be studied; S2. Obtain the BeiDou positioning information of the irrigation terminal nodes in the irrigation area to be studied, and establish a spatiotemporal database of all irrigation terminal nodes in the irrigation area to be studied. S3. Based on the spatiotemporal database, fuzzy C-means clustering is used to divide the irrigation area under study into several partitions, and the nodes in each partition with a membership degree greater than or equal to a set threshold are used as the initial boundaries of the partitions. S4. Smooth the initial boundary of each partition to obtain the final partition boundary of each partition; then determine the water demand relationship of the partition based on the water demand map of the irrigation area to be studied. S5. Based on the water demand map and water shortage index map of the irrigation area to be studied and the water demand relationship of the sub-districts, calculate the gross irrigation quota of the sub-districts as the irrigation prescription map of the sub-districts.

2. The irrigation prescription map zoning extraction method according to claim 1, characterized in that, Methods for smoothing the initial boundaries of each partition include: S41. Based on the boundary points of each partition, the weighting coefficient λ is calculated by minimizing the energy objective function. C , λ L , λ K Energy objective function The expression is: ， ， in, For length items; For constraint terms; λ is the curvature term; C , λ L , λ K is the weighting coefficient; n is the total number of boundary nodes in the partition; and b is the i-th and i-1-th optimized boundary point on the partition boundary; i Let i be the i-th unoptimized original boundary point on the partition boundary; To represent Euclidean distance; S42. Based on the weighting coefficient λ C , λ L , λ K Calculate the smoothed boundary points for each partition boundary: Among them, A L A is a tridiagonal matrix; C A is a diagonal matrix; K It is a pentagonal matrix; The set of boundary points after partitioning and smoothing. , and These are the 1st, 2nd, and kth boundary points in the boundary point set S, respectively. The set of boundary points before partitioning is smoothed. , and These are the 1st, 2nd, and kth boundary points in the boundary point set B, respectively.

3. The irrigation prescription map zoning extraction method according to claim 1, characterized in that, The method for determining the water demand relationship of each zone is as follows: Determine whether the water supply and water demand of the irrigation district under study are met. If so, the water demand relationship of the zone is a fully irrigated scenario; otherwise, the water demand relationship of the zone is a non-fully irrigated scenario.

4. The irrigation prescription map zoning extraction method according to claim 3, characterized in that, When the water demand relationship of a zone is a fully irrigated scenario, the methods for calculating the irrigation prescription map of the zone include: S511. Calculate the net irrigation quota corresponding to the irrigation zone: in, The net irrigation quota corresponding to the zone; X represents the irrigation water volume corresponding to the irrigation zone; X and Y represent the spatial coordinates of the terminal nodes of the irrigation zone. max and X min The upper and lower limits of X and Y are respectively. max and Y min These are the upper and lower limits of Y, respectively; The irrigation quota corresponding to the spatial coordinate range; This represents the area within the coordinates of the spatial range. S512. Calculate the regional gross irrigation quota as its irrigation prescription map: in, For regional gross irrigation quotas; This refers to the irrigation water utilization coefficient. The irrigation guarantee coefficient is the factor used under fully irrigated conditions. ; When the water demand relationship of a zone is a non-fully irrigated scenario, the methods for calculating the irrigation prescription map of the zone include: S521. Determine the irrigation guarantee factor based on the water shortage index diagram: Where θ is the mean of the water shortage index for the region; θ min Let θ be the minimum value of θ within the irrigation area under study; 0.8 ≤ θ ≤ 1 represents a severely water-deficient area within the irrigation area under study; 0.5 ≤ θ ≤ 1 represents a moderately water-deficient area within the irrigation area under study; θ min ≤θ≤0.5 indicates a slightly water-deficient area within the irrigation district under study; S522. Calculate the initial value of the gross irrigation quota based on the irrigation guarantee coefficient. : ； S523. Calculate the total water demand of the irrigation area under study based on the gross irrigation quota. : in, This is the initial value for the gross irrigation quota of the j-th partition; This represents the total number of partitions. S524. Based on the water demand of the irrigation area under study With water supply Calculate the water distribution coefficient : ； S625, based on water distribution coefficient The final gross irrigation quota for each zone is calculated as its irrigation prescription map: 。 5. The irrigation prescription map zoning extraction method according to claim 1, characterized in that, Step S3 further includes: S31. Based on the spatiotemporal database, extract the latitude and longitude coordinates of all irrigation terminal nodes in the end-domain area to form a point set. : in, Let be the longitude and latitude of the i-th terminal irrigation node; , and Point sets The first, second and qth end-domain irrigation terminal nodes are defined, where q is the total number of end-domain irrigation terminal nodes in the irrigation district to be studied. S32. The objective function is optimized by using the fuzzy C-means clustering method to assign q nodes to C cluster centers, resulting in several partitions; S33. Extract the membership degree of all end-domain irrigation terminal nodes to cluster j from the membership degree matrix u. For each partition j, construct a membership degree surface based on the membership degree value. S34. For each partition j, its boundary is defined by a membership degree equal to a set threshold. The contour lines, i.e., the boundaries of each partition, are: in, Let j be the boundary of the j-th partition; The coordinates of the irrigation terminal nodes in the partition; for The degree of membership.

6. The irrigation prescription map zoning extraction method according to claim 5, characterized in that, Constructing membership surfaces The expression is: Where p is the coordinate of any point; φ() is the Gaussian function, expressed as: The coefficient ε = 0.5; To represent Euclidean distance; the coefficient λ i a, b, and c are determined by solving the following system of equations: in, Let be the membership degree of terminal node i in the irrigation domain to cluster j.

7. The irrigation prescription map zoning extraction method according to claim 5, characterized in that, Step S32 further includes: S321. Initialize the membership degree of each irrigation terminal node in the end domain; S322, Update the membership degree of terminal node i to cluster j in the irrigation domain: Among them, v j v is the j-th cluster center; k Let be the k-th cluster center; c is the number of cluster centers; m is the fuzzy index; Euclidean distance; S323. Update the cluster centers based on the updated membership degrees: ； S324. Determine whether the cluster centers before and after the update meet the termination condition. If yes, complete the partitioning; otherwise, return to step S322. The termination condition is: in, This is the termination threshold.

8. The irrigation prescription map zoning extraction method according to any one of claims 5-7, characterized in that, The expression for the objective function J in step S32 is: in, ; denoted by Euclidean distance; c represents the total number of cluster centers.

Images