Python-based pumped storage reservoir intelligent site selection method and system

Abstract

The application provides a kind of python-based pumped storage reservoir intelligent site selection method and system, method includes: obtaining the Dem basic terrain data and remote sensing image of preselected area;Fill the pit processing;River section identification submodule carries out river section identification;Sub-basin identification submodule carries out sub-basin identification;Generate contour line;Using reservoir identification module, preliminary determine several reservoirs in preselected area;Using pumped storage power station reservoir matching identification module, the upper and lower reservoir matching of each reservoir preliminarily determined is carried out, all reservoir combinations meeting constraint conditions are identified, and different reservoir intelligent site selection combination schemes are formed.The application can effectively improve the efficiency of reservoir site selection, realize the automation and intelligentization of reservoir site selection process;In addition, the application can also reduce the traditional manpower investment and experience error, improve the accuracy and reliability of site selection result, thereby effectively support large-scale multi-region pumped storage power station site selection and engineering decision.

Description

Technical Field

[0001] The present invention belongs to the technical field of water conservancy and hydropower engineering, and particularly relates to a Python-based intelligent site selection method and system for pumped storage reservoirs. Background Art

[0002] Currently, the development of pumped-storage power stations is entering a construction boom. Large-scale site selection is required in the early stages of construction, and the site selection for pumped-storage power stations is crucial. Specifically, site selection for pumped-storage power stations requires consideration of water source conditions, head, reservoir capacity, the distance-to-height ratio between the upper and lower reservoirs, geographic location, and geological conditions. This is a time-consuming, labor-intensive task with numerous influencing factors. Currently, manual site selection on topographic maps is the primary method used. This manual site selection method primarily relies on topographic orientation, aiming for the largest possible reservoir capacity and the smallest distance-to-height ratio between the upper and lower reservoirs as site selection criteria. However, manual site selection is time-consuming, can lead to errors due to empirical judgment, and can be incomplete in its selection process. Therefore, there is an urgent need to improve site selection efficiency and enhance the accuracy and reliability of site selection results. Summary of the Invention

[0003] In view of the defects of the existing technology, the present invention provides a Python-based pumped storage reservoir intelligent site selection method and system, which can effectively solve the above problems.

[0004] The technical solution adopted in the present invention is as follows:

[0005] The first aspect of the present invention provides a Python-based intelligent site selection method for pumped storage reservoirs, comprising the following steps:

[0006] Step S1, obtaining Dem basic terrain data and remote sensing images of a preselected area;

[0007] Step S2, based on the Dem basic terrain data, using a data preprocessing module to perform depression-filling processing on the remote sensing image to obtain a remote sensing image vector layer after the depression-filling processing;

[0008] Step S3, using the river section identification submodule in the water system identification module, based on pre-set river section identification related parameters, to identify and extract the river sections in the remote sensing image vector layer, thereby generating a river section distribution vector layer of the pre-selected area;

[0009] Step S4, using the sub-watershed identification submodule in the water system identification module, based on pre-set sub-watershed identification related parameters, to identify and extract the sub-watersheds in the river section distribution vector layer of the pre-selected area, thereby generating a sub-watershed distribution vector layer of the pre-selected area;

[0010] Step S5, generating contour lines in the sub-watershed distribution vector layer of the pre-selected area to obtain a sub-watershed distribution vector layer with contour lines;

[0011] Step S6, using a reservoir identification module to perform reservoir site selection analysis on the sub-basin distribution vector layer with contour lines, and preliminarily determine several reservoirs in the pre-selected area;

[0012] In step S7, a pumped storage power station reservoir matching and identification module is used to match the upper and lower reservoirs of each initially determined reservoir, identify all reservoir combinations that meet the constraint conditions, and form a variety of different reservoir intelligent site selection combination schemes.

[0013] Preferably, step S3 is specifically as follows:

[0014] Step S3.1, pre-setting the river section identification related parameters including the river section identification scale threshold;

[0015] Step S3.2, preliminarily identifying a number of river sections that meet the river section identification scale threshold in the remote sensing image vector layer;

[0016] Step S3.3: Preset river section screening conditions, including the elevation range of the downstream outlet of the river section and the elevation difference range of the river section; use the river section screening conditions to filter the initially identified river sections, filter out river sections that do not meet the river section screening conditions, and obtain a plurality of filtered river sections;

[0017] Step S3.4: For each filtered river section, perform node segmentation:

[0018] For each filtered river section, take the downstream endpoint of the river section as the first node, and mark multiple nodes in sequence along the river section from downstream to upstream according to the preset node segmentation scale;

[0019] In step S3.5, all river sections after the marked nodes are used to generate a river section distribution vector layer for the preselected area.

[0020] Preferably, step S4 is specifically as follows:

[0021] Step S4.1, using the sub-basin identification sub-module in the water system identification module, using the Dem basic terrain data and the D8 algorithm, analyze the flow direction of each river section in the river section distribution vector layer to determine the flow direction of each river section;

[0022] Step S4.2: Analyze and determine several river system intersections based on the flow direction of each river section; use each river system intersection as a discharge outlet, determine the river sections associated with the river system intersection, and thus divide and form several sub-basins;

[0023] Step S4.3: extract each sub-watershed of the pre-selected area, thereby generating a sub-watershed distribution vector layer of the pre-selected area.

[0024] Preferably, step S6 is specifically as follows:

[0025] Step S6.1, determining the relevant constraint parameters for reservoir identification, including the preset thresholds of reservoir characteristic parameters and the scanning section; the scanning section is the target length ratio of the river section from downstream to upstream to identify the reservoir;

[0026] Step S6.2, performing reservoir site selection analysis on each sub-basin in parallel in the sub-basin distribution vector layer with contour lines;

[0027] For each sub-basin, the reservoir site selection analysis method is as follows: traverse each river section of the sub-basin in sequence from downstream to upstream. For the currently traversed river section, the reservoir site selection analysis method is as follows:

[0028] Step S6.2.1, determining the target length interval of the current river section based on the scanned section;

[0029] Step S6.2.2: In the target length interval of the river section, the second node of the river section is used as the scanning starting node. At the second node, an initial dam line is given in a direction perpendicular to the river section. Then, it is determined whether the initial dam line intersects with the river section contour line to form a closed area. If so, step S6.2.3 is executed. If not, the dam line is moved along the river section nodes with the divided nodes as the step length, so that the dam line is moved to the third node, and a dam line perpendicular to the river section direction is formed at the third node. Then, it is determined whether the dam line intersects with the river section contour line to form a closed area. If so, step S6.2.3 is executed. If not, the dam line is continued to be moved along the river section nodes until it moves to the last node of the target length interval of the river section, which means that the river section is not suitable for reservoir site selection, and the analysis of the current river section is terminated.

[0030] Step S6.2.3, determining the nearest intersection point between the current dam line and the left and right sides of the river section, and connecting the two intersection points to form a dam line segment; the dam line segment and the river section contour line form a closed polygonal area, which is initially used as the determined reservoir surface;

[0031] Step S6.2.4, using the area Dem of the reservoir surface, the surface volume method is used to calculate the reservoir characteristic parameters;

[0032] Step S6.2.5, determine whether the reservoir characteristic parameters meet the preset threshold values ​​of the reservoir characteristic parameters. If they meet the threshold values, retain the currently analyzed reservoir site. If they do not meet the threshold values, continue scanning the next node of the current river section until the last node of the target length interval of the river section is scanned, indicating that the river section is not suitable as a reservoir site, and the analysis of the current river section is ended.

[0033] Preferably, the preset thresholds of the reservoir characteristic parameters include preset thresholds of dam length, dam height, dam top elevation, dam bottom elevation, reservoir surface area and reservoir capacity, as well as adjacent multipliers.

[0034] Preferably, when conducting reservoir site selection analysis for each river section, the following is also included:

[0035] In the process of identifying and calculating the reservoir capacity, the reservoir capacity range threshold is used as a constraint. When the calculated reservoir capacity meets the reservoir capacity range threshold, the river section where the current reservoir is located is jumped out, and from the current reservoir dam line position, the river section is extended by the maximum length of the diagonal of the range where the current reservoir area is located multiplied by the length of the adjacent multiplier, and the river section is automatically jumped to the position of the end of the extended river section to continue the reservoir site analysis and identification.

[0036] Preferably, step S7 is specifically as follows:

[0037] The constraints are determined to be the range of distance-to-height ratio and the range of reservoir capacity assessment coefficient;

[0038] The distance-to-height ratio is the horizontal distance between the center points of the dam sites of the two reservoirs / the difference in normal water level elevation between the upper and lower reservoirs;

[0039] The storage capacity assessment coefficient is calculated by the following formula:

[0040]

[0041] Where: S is the reservoir capacity assessment coefficient; v is the reservoir capacity; L is the dam length; H is the dam height.

[0042] The second aspect of the present invention provides a Python-based pumped storage reservoir intelligent site selection system, comprising:

[0043] Acquisition module, used to obtain Dem basic terrain data and remote sensing images of pre-selected areas;

[0044] A depression filling processing module is used to perform depression filling processing on the remote sensing image based on the Dem basic terrain data using the data preprocessing module to obtain a vector layer of the remote sensing image after the depression filling processing;

[0045] Water system identification module, including river section identification submodule and sub-basin identification submodule;

[0046] The river section identification submodule is used to identify and extract river sections in the remote sensing image vector layer based on preset river section identification related parameters, thereby generating a river section distribution vector layer of a preselected area;

[0047] The sub-watershed identification submodule is used to identify and extract the sub-watersheds in the river section distribution vector layer of the pre-selected area based on pre-set sub-watershed identification related parameters, thereby generating the sub-watershed distribution vector layer of the pre-selected area;

[0048] A contour line generation module is used to generate contour lines in the sub-basin distribution vector layer of the pre-selected area to obtain a sub-basin distribution vector layer with contour lines;

[0049] A reservoir identification module is used to perform reservoir site selection analysis on the sub-basin distribution vector layer with contour lines, and preliminarily determine a number of reservoirs in the pre-selected area;

[0050] The pumped storage power station reservoir matching and identification module is used to match the upper and lower reservoirs of each initially determined reservoir, identify all reservoir combinations that meet the constraint conditions, and form a variety of different reservoir intelligent site selection combination plans.

[0051] A third aspect of the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the Python-based intelligent site selection method for pumped storage reservoirs by executing the computer instructions.

[0052] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the Python-based intelligent site selection method for pumped storage reservoirs.

[0053] The Python-based intelligent site selection method and system for pumped storage reservoirs provided by the present invention have the following advantages:

[0054] The present invention can effectively improve the efficiency of reservoir site selection and realize the automation and intelligence of the reservoir site selection process; in addition, the present invention can also reduce traditional manpower input and experience errors, improve the accuracy and reliability of site selection results, and thus effectively support the site selection and engineering decision-making of large-scale and multi-regional pumped-storage power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The drawings that constitute part of the specification of the present application are used to provide a further understanding of the present application. The illustrative embodiments of the present application and their descriptions are only used to explain the present application and do not constitute an improper limitation on the present application.

[0056] Figure 1 Schematic diagram of the process of the intelligent site selection method for a pumped storage reservoir according to an embodiment of the present invention;

[0057] Figure 2 Schematic diagram of reservoir identification and matching results of the intelligent site selection system for pumped storage reservoirs described in an embodiment of the present invention;

[0058] Figure 3 Schematic diagram of the process of the intelligent site selection method for pumped storage reservoirs described in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] To further clarify the objectives, technical solutions, and advantages of the embodiments of the present invention, the following will provide a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. The described embodiments are only a portion of the embodiments of the present invention, not all of them. The specific implementation methods described are intended only to illustrate the present invention and are not intended to limit it. All other embodiments derived by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0061] The present invention provides a method and system for intelligent site selection for pumped-storage reservoirs that is simple to operate, flexible, efficient, and stable, and can realize automatic identification of water system reservoirs in a large-scale regional range and automatic matching of upper and lower reservoirs. The present invention can effectively improve the efficiency of reservoir site selection and realize the automation and intelligence of the reservoir site selection process; in addition, the present invention can also reduce traditional manpower input and experience errors, improve the accuracy and reliability of site selection results, and thus effectively support the site selection and engineering decision-making of large-scale and multi-regional pumped-storage power stations.

[0062] See Figure 1 The present invention provides a Python-based intelligent site selection method for pumped storage reservoirs, comprising the following steps:

[0063] Step S1, obtaining Dem basic terrain data and remote sensing images of a preselected area;

[0064] Step S2, based on the Dem basic terrain data, using a data preprocessing module to perform depression-filling processing on the remote sensing image to obtain a remote sensing image vector layer after the depression-filling processing;

[0065] Step S3, using the river section identification submodule in the water system identification module, based on pre-set river section identification related parameters, to identify and extract the river sections in the remote sensing image vector layer, thereby generating a river section distribution vector layer of the pre-selected area;

[0066] Step S3 is specifically as follows:

[0067] Step S3.1, pre-setting the river section identification related parameters including the river section identification scale threshold;

[0068] Step S3.2, preliminarily identifying a number of river sections that meet the river section identification scale threshold in the remote sensing image vector layer;

[0069] Step S3.3: Preset river section screening conditions, including the elevation range of the downstream outlet of the river section and the elevation difference range of the river section; use the river section screening conditions to filter the initially identified river sections, filter out river sections that do not meet the river section screening conditions, and obtain a plurality of filtered river sections;

[0070] Step S3.4: For each filtered river section, perform node segmentation:

[0071] For each filtered river section, take the downstream endpoint of the river section as the first node, and mark multiple nodes in sequence along the river section from downstream to upstream according to the preset node segmentation scale;

[0072] In step S3.5, all river sections after the marked nodes are used to generate a river section distribution vector layer for the preselected area.

[0073] Step S4, using the sub-watershed identification submodule in the water system identification module, based on pre-set sub-watershed identification related parameters, to identify and extract the sub-watersheds in the river section distribution vector layer of the pre-selected area, thereby generating a sub-watershed distribution vector layer of the pre-selected area;

[0074] Step S4 is specifically as follows:

[0075] Step S4.1, using the sub-basin identification sub-module in the water system identification module, using the Dem basic terrain data and the D8 algorithm, analyze the flow direction of each river section in the river section distribution vector layer to determine the flow direction of each river section;

[0076] Step S4.2: Analyze and determine several river system intersections based on the flow direction of each river section; use each river system intersection as a discharge outlet, determine the river sections associated with the river system intersection, and thus divide and form several sub-basins;

[0077] Step S4.3: extract each sub-watershed of the pre-selected area, thereby generating a sub-watershed distribution vector layer of the pre-selected area.

[0078] Step S5, generating contour lines in the sub-watershed distribution vector layer of the pre-selected area to obtain a sub-watershed distribution vector layer with contour lines;

[0079] Step S6, using a reservoir identification module to perform reservoir site selection analysis on the sub-basin distribution vector layer with contour lines, and preliminarily determine several reservoirs in the pre-selected area;

[0080] Step S6 is specifically as follows:

[0081] Step S6.1, determine the relevant constraint parameters for reservoir identification, including the preset thresholds and scanning sections of reservoir characteristic parameters; the scanning section is the target length ratio of the river section from downstream to upstream for identifying the reservoir; the preset thresholds of the reservoir characteristic parameters include the preset thresholds of dam length, dam height, dam top elevation, dam bottom elevation, reservoir surface area and reservoir capacity, as well as the adjacent multiplier.

[0082] Step S6.2, performing reservoir site selection analysis on each sub-basin in parallel in the sub-basin distribution vector layer with contour lines;

[0083] For each sub-basin, the reservoir site selection analysis method is as follows: traverse each river section of the sub-basin in sequence from downstream to upstream. For the currently traversed river section, the reservoir site selection analysis method is as follows:

[0084] Step S6.2.1, determining the target length interval of the current river section based on the scanned section;

[0085] Step S6.2.2: In the target length interval of the river section, the second node of the river section is used as the scanning starting node. At the second node, an initial dam line is given in a direction perpendicular to the river section. Then, it is determined whether the initial dam line intersects with the river section contour line to form a closed area. If so, step S6.2.3 is executed. If not, the dam line is moved along the river section nodes with the divided nodes as the step length, so that the dam line is moved to the third node, and a dam line perpendicular to the river section direction is formed at the third node. Then, it is determined whether the dam line intersects with the river section contour line to form a closed area. If so, step S6.2.3 is executed. If not, the dam line is continued to be moved along the river section nodes until it moves to the last node of the target length interval of the river section, which means that the river section is not suitable for reservoir site selection, and the analysis of the current river section is terminated.

[0086] Step S6.2.3, determining the nearest intersection point between the current dam line and the left and right sides of the river section, and connecting the two intersection points to form a dam line segment; the dam line segment and the river section contour line form a closed polygonal area, which is initially used as the determined reservoir surface;

[0087] Step S6.2.4, using the area Dem of the reservoir surface, the surface volume method is used to calculate the reservoir characteristic parameters;

[0088] Step S6.2.5, determine whether the reservoir characteristic parameters meet the preset threshold values ​​of the reservoir characteristic parameters. If they meet the threshold values, retain the currently analyzed reservoir site. If they do not meet the threshold values, continue scanning the next node of the current river section until the last node of the target length interval of the river section is scanned, indicating that the river section is not suitable as a reservoir site, and the analysis of the current river section is ended.

[0089] In practical applications, when conducting reservoir site selection analysis for each river section, the following also needs to be considered:

[0090] In the process of identifying and calculating the reservoir capacity, the reservoir capacity range threshold is used as a constraint. When the calculated reservoir capacity meets the reservoir capacity range threshold, the river section where the current reservoir is located is jumped out, and from the current reservoir dam line position, the river section is extended by the maximum length of the diagonal of the range where the current reservoir area is located multiplied by the length of the adjacent multiplier, and the river section is automatically jumped to the position of the end of the extended river section to continue the reservoir site analysis and identification.

[0091] In step S7, a pumped storage power station reservoir matching and identification module is used to match the upper and lower reservoirs of each initially determined reservoir, identify all reservoir combinations that meet the constraint conditions, and form a variety of different reservoir intelligent site selection combination schemes.

[0092] In this step, the constraints are determined to be the distance-to-height ratio range and the reservoir capacity assessment coefficient range;

[0093] The distance-to-height ratio is the horizontal distance between the center points of the dam sites of the two reservoirs / the difference in normal water level elevation between the upper and lower reservoirs;

[0094] The storage capacity assessment coefficient is calculated by the following formula:

[0095]

[0096] Where: S is the reservoir capacity assessment coefficient; v is the reservoir capacity; L is the dam length; H is the dam height.

[0097] The present invention also provides a Python-based pumped storage reservoir intelligent site selection system, comprising:

[0098] Acquisition module, used to obtain Dem basic terrain data and remote sensing images of pre-selected areas;

[0099] A depression filling processing module is used to perform depression filling processing on the remote sensing image based on the Dem basic terrain data using the data preprocessing module to obtain a vector layer of the remote sensing image after the depression filling processing;

[0100] Water system identification module, including river section identification submodule and sub-basin identification submodule;

[0101] The river section identification submodule is used to identify and extract river sections in the remote sensing image vector layer based on preset river section identification related parameters, thereby generating a river section distribution vector layer of a preselected area;

[0102] The sub-watershed identification submodule is used to identify and extract the sub-watersheds in the river section distribution vector layer of the pre-selected area based on pre-set sub-watershed identification related parameters, thereby generating the sub-watershed distribution vector layer of the pre-selected area;

[0103] A contour line generation module is used to generate contour lines in the sub-basin distribution vector layer of the pre-selected area to obtain a sub-basin distribution vector layer with contour lines;

[0104] A reservoir identification module is used to perform reservoir site selection analysis on the sub-basin distribution vector layer with contour lines, and preliminarily determine a number of reservoirs in the pre-selected area;

[0105] The pumped storage power station reservoir matching and identification module is used to match the upper and lower reservoirs of each initially determined reservoir, identify all reservoir combinations that meet the constraint conditions, and form a variety of different reservoir intelligent site selection combination plans.

[0106] The present invention also provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the aforementioned Python-based intelligent site selection method for pumped storage reservoirs by executing the computer instructions.

[0107] The present invention also provides a computer-readable storage medium, in which computer instructions are stored. The computer instructions are used to enable a computer to execute the aforementioned Python-based intelligent site selection method for pumped storage reservoirs.

[0108] Two embodiments are described below:

[0109] Example 1:

[0110] This embodiment of the present invention discloses a Python-based intelligent site selection method for pumped-storage reservoirs, including modules for automatic water system identification, automatic watershed (sub-basin) division, automatic reservoir identification, and automatic identification of pumped-storage matching. The main steps are as follows:

[0111] S1: Obtain vector data, such as DEM and remote sensing images, for the preselected area. Load these DEM and remote sensing images into the intelligent site selection system, enabling it to intelligently select a site for the pumped-storage reservoir based on the underlying DEM terrain data. The DEM terrain data can be downloaded from a website. The developed system supports DEM data in both UTM (WGS1984) and Gauss-Kruger (CGCS2000) projections.

[0112] S2: Use the data preprocessing module to fill the dem terrain data to eliminate the impact of potholes on the system model operation.

[0113] S3: Automatically identify river sections in pre-selected areas.

[0114] By using the system's water system identification module and setting the relevant parameters for water system identification, the system can automatically identify the water system in the entire region with one click, extract river section feature data such as the number of river sections and river section length, and generate a vector layer.

[0115] Specifically, when automatically identifying river segments, the river network is determined and extracted based on a threshold for segment identification size. The smaller the threshold, the more river segments are automatically generated. The automatic river system identification module includes a segment filtering model that can filter river segments based on their length, downstream outlet elevation, and elevation difference. River segments that do not meet the filtering criteria are automatically deleted, while the target segments are retained, thus ensuring the quality of the automatically generated river network.

[0116] S4: Divide the watershed (sub-basin) based on the generated global water system.

[0117] The system's sub-basin identification module automatically divides the watershed into several sub-basins, using the water system intersection as the discharge outlet. It extracts sub-basin characteristic data such as the number and area of ​​sub-basins and generates a vector layer.

[0118] Specifically, when automatically identifying catchment areas (sub-basins), the system uses the D8 algorithm, utilizing DEM grid data, to perform flow direction analysis and watershed extraction. Watershed identification begins at the second node of each river segment to avoid node overlap. Parallel computation is performed across the entire domain to identify each sub-basin, with a 5-second wait interval for each sub-basin.

[0119] S5: Use dem to generate contour lines in the entire domain and extract the contour lines of each sub-basin.

[0120] S6: Based on the above steps, automatic identification of reservoirs in the entire region is carried out, and basic parameter data such as dam length, dam height, dam top and bottom elevation, dam top and bottom coordinates, reservoir water surface area, and storage capacity assessment coefficient of each reservoir are calculated, identified, and extracted.

[0121] Specifically, for automatic reservoir identification, the relevant constraint parameters of the reservoir identification calculation must be determined before the identification calculation, including the storage capacity range, scanning section, adjacent magnification, maximum dam height, maximum dam length, etc.; among which, the scanning section is the length ratio of the river section from upstream to downstream or from downstream to upstream for identifying the site of the reservoir.

[0122] The identification process involves using multi-segment node recognition technology. Based on each sub-basin's river section, the system automatically scans along the river section from downstream to upstream, starting with the second node as the starting node. The system then moves the dam line perpendicular to the river section, gradually moving it along the river section nodes with a step size equal to the number of nodes until a closed area intersecting with the contour lines within the river section is found. The dam line is then identified by the closest intersection points on the left and right sides of the river section. The dam line and the contour lines form a closed polygon, which serves as the initial reservoir surface. Using this surface area (dem), the system calculates reservoir characteristic values ​​and other related parameters, including dam length, height, crest and bottom elevations, reservoir surface area, and storage capacity, using the surface volume method. The system then determines whether these parameters meet preset thresholds for reservoir characteristic parameters. If so, the reservoir is retained; otherwise, the system continues searching for further reservoirs until all sub-basins have been identified. Once the current sub-basin has been identified, the system automatically jumps to the next sub-basin and continues searching for reservoirs using the same method until all sub-basins have been identified. At this point, the system realizes automatic identification and calculation of reservoirs across the entire region with one click.

[0123] S7: After identifying all reservoirs in the region, automatic matching and identification calculations are performed for the upper and lower reservoirs. Using distance-to-height ratio, head difference between the upper and lower reservoirs, and horizontal distance between the upper and lower reservoirs as constraints and screening criteria, all reservoirs are automatically paired. All combinations that meet the constraints are identified as the final pumped storage matching results, completing the final site selection.

[0124] Specifically, for automatic identification of pumped-storage matching, one of the main screening conditions is the distance-to-height ratio: the horizontal distance between the center points of the dam sites of the two reservoirs / the difference in normal water level elevation between the upper and lower reservoirs.

[0125] This embodiment also provides a Python-based pumped storage reservoir intelligent site selection system, including the following modules:

[0126] The geographic image acquisition and processing module is used to acquire and load terrain elevation data, establish a geographic information system database, and process the terrain to provide basic spatial data for the intelligent site selection of pumped storage reservoirs.

[0127] The modeling module is used to construct a water system automatic identification model, a catchment area automatic identification model, a reservoir automatic identification model, an upper and lower reservoir matching model, etc. during the intelligent site selection process, and ultimately screen out the ideal reservoir site combination in the region.

[0128] This system independently developed several Python-based pumped storage site selection model libraries, including Contour.py, ContourClip.py, DeleteResv.py, DEM_Fill.py, Rch_Detect.py, Wsh_Detect.py, WSmainM.py, and WSmainS.py. Key application programming interfaces (APIs) were encapsulated. The Python-based libraries primarily include a data processing module (for loading and processing various data), a hydrological calculation module, an optimization module (for implementing multi-objective optimization algorithms), and a results visualization module (using libraries such as matplotlib and seaborn for visualization). Furthermore, the open-source MapWinGIS engine enables comprehensive map interaction.

[0129] Example 2:

[0130] This embodiment takes a pumped storage reservoir pre-selected area of ​​the present invention as an example and further explains with reference to the accompanying drawings:

[0131] Based on the pumped storage site selection requirements, determine the pre-selected area for the construction of the pumped storage power station.

[0132] Get the DEM data of the pre-selected area. The pre-selected site area of ​​a pumped storage reservoir is 413km 2 , the DEM data with an accuracy of 7.26m and a projection coordinate system of CGCS2000 is used as the basic terrain data for intelligent site selection.

[0133] After loading the dem data into this system, the preprocessing module is used to fill the dem data. The purpose of this step is to eliminate the impact of potholes on the system model operation.

[0134] The main process of the pumped storage reservoir site selection embodiment of the present invention is:

[0135] Automatic identification of water systems: The system obtains topographic and geological characteristics and water system characteristics based on DEM data, and uses the water system identification module to identify and extract the water system distribution and basic data of the pre-selected area, including the number of river sections, river section length, river distribution, etc.

[0136] In the water system identification process of the embodiment, a threshold for river segment identification must first be determined. The smaller the threshold, the denser the identified river network and the more river segments generated. In this embodiment, when the threshold is 30,000, 113 river segments are generated.

[0137] Automatic sub-basin identification: Automatically delineates catchment areas (sub-basins) based on the generated global water system. Using the system's sub-basin identification tool, the system automatically delineates sub-basins within pre-selected areas, using water system intersections as discharge outlets. It then extracts characteristic data, including sub-basin number, area, and distribution, along with vector layers.

[0138] In the sub-watershed identification process of the embodiment, the system adopts parallel computing to significantly improve the calculation and identification speed. In this embodiment, based on the above-mentioned water system identification, a total of 110 sub-watersheds are generated.

[0139] Automatic reservoir identification. Before reservoir identification, the pre-processing tool in the reservoir identification module uses DEM (Descriptive Interpretation) to generate contour lines across the entire region and extract contour lines for each sub-basin to be identified. After contour extraction, the system uses the Reservoir Identification tool and multi-segment node recognition technology. Based on the selected river section of each sub-basin, the system automatically scans along the river section from downstream to upstream, starting with the second node as the starting node. The system then gradually moves the dam line along the nodes of the selected river section, perpendicular to the river section, until it finds a closed area intersecting with the contour lines within the river section. The closest intersection point to the left and right sides of the river section is used as the final dam line. When the dam line and the contour lines form a closed area, it is considered the initially identified reservoir surface. Using DEM (Descriptive Interpretation) of this surface area, the surface volume method is used to calculate reservoir characteristic values ​​and other relevant parameters, including dam length, height, crest and bottom elevations, reservoir surface area, and storage capacity. The system determines whether the parameters meet the preset thresholds for reservoir characteristic parameters. If so, the reservoir is retained. Otherwise, the system continues searching for new reservoirs until all sub-basins have been identified. Once the current sub-basin has been identified, the system automatically jumps to the next sub-basin and continues searching for reservoirs until all sub-basins have been identified. This allows the system to automatically identify and calculate reservoirs across the entire region with one click.

[0140] In the embodiment described, before the reservoir identification process, it is necessary to perform node segmentation on the generated river network in the water system identification stage. Then, when identifying the reservoir, on the selected river section of each river basin to be identified, the dam line is gradually moved along the nodes on the selected river section with the node interval as the step size to carry out the reservoir identification calculation process.

[0141] In the automatic reservoir identification stage of the embodiment, it is necessary to set the relevant constraint parameters for reservoir identification calculation before identification, that is, the preset thresholds of reservoir characteristic parameters, including storage capacity range, scanning section, maximum dam height, maximum dam length, adjacent multiplier, etc.; wherein, the scanning section is the length ratio of the river section from upstream to downstream or from downstream to upstream to identify the reservoir. In the process of identifying and calculating storage capacity, the storage capacity range threshold is used as a constraint. When the calculated reservoir capacity meets the threshold requirement, the river section where the current reservoir is located is jumped out, and from the current reservoir dam line position, along the selected river section, the river section is extended by the maximum length of the diagonal of the range where the current reservoir area is located multiplied by the adjacent multiplier, and the river section is automatically jumped to the position where the end of the extended river section is located to continue identification calculation. Until the identification of reservoirs in the basin is completed. In this embodiment, the reservoir selection range is 1-20 million m 3 The scanning section is 100% scanning from downstream to upstream, with a maximum dam height of 200m and a maximum dam length of 1500m. After identification, a total of 77 reservoirs meeting the reservoir conditions were identified in this embodiment.

[0142] The reservoir automatic identification module of the embodiment uses parallel computing. Parallel computing involves building a computing cluster system within the system, comprising a main controller and multiple computing nodes. The main controller can be implemented using a master computer, which includes an engine management center. The entire system is managed by a main program to manage the number of engines and the execution of task functions. The master controller automatically splits the computing tasks of each stage into several subtasks, automatically assigns them to each computing cluster node, and retrieves the calculation results of each node in real time. This parallel computing method significantly improves the efficiency of reservoir site selection, shortens the time required for reservoir site selection, and solves the problem of high computing workload and extremely low reservoir site selection efficiency.

[0143] Pumping matching identification:

[0144] This example uses a maximum distance-to-height ratio of 18, a head difference between the upper and lower reservoirs of 50-750m, and a maximum horizontal distance of 10km as constraints and screening criteria. Using the pumped-storage matching identification tool, all reservoirs are automatically paired. Pairs that don't meet the screening criteria are automatically filtered out, and all pairs that meet the constraints are retained as the pumped-storage matching identification results, completing the preliminary site selection process.

[0145] After pumping and storage matching identification, this embodiment screened out a total of 12 pairs of reservoir combinations that meet the pumping and storage conditions.

[0146] The pumped-storage matching identification module described in this embodiment effectively addresses the survey and planning issues for pumped-storage reservoir combinations, reducing the workload of manual initial reservoir matching calculations and site selection surveys, accelerating site selection, and facilitating faster and more accurate acquisition of optimal site combinations. This intelligent site selection system and method can be used as technical support and assistance in the early stages of site selection for actual projects. Subsequently, the specific design requirements and kinetic energy parameter requirements of pumped-storage power plants can be combined to further develop pumped-storage project construction.

[0147] The above embodiment provides a Python-based pumped storage reservoir intelligent site selection method and system, which has broad application prospects.

[0148] This invention provides a Python-based method and system for intelligent site selection for pumped-storage reservoirs. These methods and systems are not limited by diverse terrain or regional scope. They only require DEM data. Based on this data, they develop and employ a multi-terrain intelligent site selection model for reservoirs, enabling automatic identification of water systems and reservoirs within the selected area, as well as pumped-storage matching identification. This system integrates multiple data analysis technologies, significantly improving site selection efficiency. This reduces the human effort and empirical errors associated with traditional manual site selection, while also improving the accuracy and reliability of site selection results. This system provides project decision makers with a scientific and reliable basis for site selection, optimizing the decision-making process for pumped-storage project construction.

[0149] The Python-based intelligent site selection method and system for pumped storage reservoirs described in the present invention proposes and adopts multi-segment node recognition technology in the intelligent identification calculation of reservoirs, realizing rapid calculation and identification of reservoirs on each selected river section, significantly improving the reservoir identification speed.

[0150] The Python-based intelligent site selection method and system for pumped storage reservoirs described in the present invention calculates the reservoir capacity curve based on DEM data using the grid method, which has higher calculation accuracy than the traditional contour method.

[0151] The embodiments described above are merely examples of the present invention. The scope of protection of the present invention is not limited by the embodiments. The present invention can be modified and varied in various ways. Any modifications, equivalent replacements, improvements, etc. made to designs that are the same or similar to the present invention within the spirit and principles of the present invention are within the scope of protection of the present invention.

Claims

1. A Python-based intelligent site selection method for pumped storage reservoirs, characterized in that: The following steps are involved: Step S1, obtaining Dem basic terrain data and remote sensing images of a preselected area; Step S2, based on the Dem basic terrain data, using a data preprocessing module to perform depression-filling processing on the remote sensing image to obtain a remote sensing image vector layer after the depression-filling processing; Step S3, using the river section identification submodule in the water system identification module, based on pre-set river section identification related parameters, to identify and extract the river sections in the remote sensing image vector layer, thereby generating a river section distribution vector layer of the pre-selected area; Step S4, using the sub-watershed identification submodule in the water system identification module, based on pre-set sub-watershed identification related parameters, to identify and extract the sub-watersheds in the river section distribution vector layer of the pre-selected area, thereby generating a sub-watershed distribution vector layer of the pre-selected area; Step S5, generating contour lines in the sub-watershed distribution vector layer of the pre-selected area to obtain a sub-watershed distribution vector layer with contour lines; Step S6, using a reservoir identification module to perform reservoir site selection analysis on the sub-basin distribution vector layer with contour lines, and preliminarily determine several reservoirs in the pre-selected area; Step S6 is specifically as follows: Step S6.1, determining the relevant constraint parameters for reservoir identification, including the preset thresholds of reservoir characteristic parameters and the scanning section; the scanning section is the target length ratio of the river section from downstream to upstream to identify the reservoir; Step S6.2, performing reservoir site selection analysis on each sub-basin in parallel in the sub-basin distribution vector layer with contour lines; For each sub-basin, the reservoir site selection analysis method is as follows: traverse each river section of the sub-basin in sequence from downstream to upstream. For the currently traversed river section, the reservoir site selection analysis method is as follows: Step S6.2.1, determining the target length interval of the current river section based on the scanned section; Step S6.2.2: In the target length interval of the river section, the second node of the river section is used as the scanning starting node. At the second node, an initial dam line is given in a direction perpendicular to the river section. Then, it is determined whether the initial dam line intersects with the river section contour line to form a closed area. If so, step S6.2.3 is executed. If not, the dam line is moved along the river section nodes with the divided nodes as the step length, so that the dam line is moved to the third node, and a dam line perpendicular to the river section direction is formed at the third node. Then, it is determined whether the dam line intersects with the river section contour line to form a closed area. If so, step S6.2.3 is executed. If not, the dam line is continued to be moved along the river section nodes until it moves to the last node of the target length interval of the river section, which means that the river section is not suitable for reservoir site selection, and the analysis of the current river section is terminated. Step S6.2.3, determining the nearest intersection point between the current dam line and the left and right sides of the river section, and connecting the two intersection points to form a dam line segment; the dam line segment and the river section contour line form a closed polygonal area, which is initially used as the determined reservoir surface; Step S6.2.4, using the area Dem of the reservoir surface, the surface volume method is used to calculate the reservoir characteristic parameters; Step S6.2.5: Determine whether the reservoir characteristic parameters meet the preset thresholds for reservoir characteristic parameters. If so, retain the currently analyzed reservoir site. If not, continue scanning the next node of the current river section until the last node of the target river section length interval is scanned, indicating that the river section is not suitable for reservoir site selection, and end the analysis of the current river section. When conducting reservoir site selection analysis for each river section, the following should also be included: In the process of identifying and calculating the reservoir capacity, the reservoir capacity range threshold is used as a constraint. When the calculated reservoir capacity meets the reservoir capacity range threshold, the river section where the current reservoir is located is jumped out, and the river section is extended from the current reservoir dam line position by the maximum length of the diagonal line of the current reservoir area range multiplied by the length of the adjacent multiplier, and the river section is automatically jumped to the position where the extended river section is located to continue the reservoir site analysis and identification; In step S7, a pumped storage power station reservoir matching and identification module is used to match the upper and lower reservoirs of each initially determined reservoir, identify all reservoir combinations that meet the constraint conditions, and form a variety of different reservoir intelligent site selection combination schemes.

2. The Python-based intelligent site selection method for pumped storage reservoirs according to claim 1 is characterized in that: Step S3 is specifically as follows: Step S3.1, pre-setting the river section identification related parameters including the river section identification scale threshold; Step S3.2, preliminarily identifying a number of river sections that meet the river section identification scale threshold in the remote sensing image vector layer; Step S3.3: Preset river section screening conditions, including the elevation range of the downstream outlet of the river section and the elevation difference range of the river section; use the river section screening conditions to filter the initially identified river sections, filter out river sections that do not meet the river section screening conditions, and obtain a plurality of filtered river sections; Step S3.4: For each filtered river section, perform node segmentation: For each filtered river section, take the downstream endpoint of the river section as the first node, and mark multiple nodes in sequence along the river section from downstream to upstream according to the preset node segmentation scale; In step S3.5, all river sections after the marked nodes are used to generate a river section distribution vector layer for the preselected area.

3. The Python-based intelligent site selection method for pumped storage reservoirs according to claim 1 is characterized in that: Step S4 is specifically as follows: Step S4.1, using the sub-basin identification sub-module in the water system identification module, using the Dem basic terrain data and the D8 algorithm, analyze the flow direction of each river section in the river section distribution vector layer to determine the flow direction of each river section; Step S4.2: Analyze and determine several river system intersections based on the flow direction of each river section; use each river system intersection as a discharge outlet, determine the river sections associated with the river system intersection, and thus divide and form several sub-basins; Step S4.3: extract each sub-watershed of the pre-selected area, thereby generating a sub-watershed distribution vector layer of the pre-selected area.

4. The method for intelligent site selection of pumped storage reservoirs based on Python according to claim 1, characterized in that: The preset thresholds of the reservoir characteristic parameters include preset thresholds of dam length, dam height, dam top elevation, dam bottom elevation, reservoir surface area and reservoir capacity, as well as adjacent multipliers.

5. The Python-based intelligent site selection method for pumped storage reservoirs according to claim 1 is characterized in that: Step S7 is specifically as follows: The constraints are determined to be the range of distance-to-height ratio and the range of reservoir capacity assessment coefficient; The distance-to-height ratio is the horizontal distance between the center points of the dam sites of the two reservoirs / the difference in normal water level elevation between the upper and lower reservoirs; The storage capacity assessment coefficient is calculated by the following formula: ， Where: S is the storage capacity assessment coefficient; is the storage capacity; For the dam chief; The dam is high.

6. A Python-based pumped storage reservoir intelligent site selection system, characterized in that: The Python-based intelligent site selection system for pumped storage reservoirs is used to implement the Python-based intelligent site selection method for pumped storage reservoirs according to any one of claims 1 to 5. The system includes: Acquisition module, used to obtain Dem basic terrain data and remote sensing images of pre-selected areas; A depression filling processing module is used to perform depression filling processing on the remote sensing image based on the Dem basic terrain data using the data preprocessing module to obtain a vector layer of the remote sensing image after the depression filling processing; Water system identification module, including river section identification submodule and sub-basin identification submodule; The river section identification submodule is used to identify and extract river sections in the remote sensing image vector layer based on preset river section identification related parameters, thereby generating a river section distribution vector layer of a preselected area; The sub-watershed identification submodule is used to identify and extract the sub-watersheds in the river section distribution vector layer of the pre-selected area based on pre-set sub-watershed identification related parameters, thereby generating the sub-watershed distribution vector layer of the pre-selected area; A contour line generation module is used to generate contour lines in the sub-basin distribution vector layer of the pre-selected area to obtain a sub-basin distribution vector layer with contour lines; A reservoir identification module is used to perform reservoir site selection analysis on the sub-basin distribution vector layer with contour lines, and preliminarily determine a number of reservoirs in the pre-selected area; The pumped storage power station reservoir matching and identification module is used to match the upper and lower reservoirs of each initially determined reservoir, identify all reservoir combinations that meet the constraint conditions, and form a variety of different reservoir intelligent site selection combination plans.

7. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the Python-based intelligent site selection method for pumped storage reservoirs according to any one of claims 1 to 5 by executing the computer instructions.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, which are used to enable a computer to execute the Python-based intelligent site selection method for pumped storage reservoirs described in any one of claims 1 to 5.

Images