Pumped storage power station batch planning and site selection method based on terrain classification

Abstract

The application discloses a pumped storage power station batch planning and site selection method based on terrain classification. The method comprises the following steps: preparing terrain data of a research area, generating a mean digital elevation model, generating a digital elevation difference model, generating a valley digital elevation model, generating an area water system, segmenting the water system, determining a range of a height ratio, calculating an upper reservoir grid range meeting a condition of the height ratio, generating an upper reservoir grid polygon longest valley, further screening by calculating a gradient of the longest valley, and selecting an upper and lower reservoir with the smallest height ratio by calculating the height ratio again.

Description

Technical Field

[0001] This invention belongs to the field of water conservancy and hydropower engineering technology, specifically relating to a method for batch planning and site selection of pumped storage power stations based on terrain classification. Background Technology

[0002] Pumped storage is an important component of a new type of power system based on new energy sources. It is of great significance for ensuring power supply, ensuring grid security, promoting the consumption of new energy sources, and promoting the construction of a clean, low-carbon, safe, and efficient energy system.

[0003] The first consideration in constructing a pumped storage power station is site selection. This involves comprehensively considering factors such as regional natural conditions, transportation, grid connection, regional economic development plans, and other facilities. Simultaneously, it's crucial to minimize disruption to the existing water system, protect vegetation, reduce earthwork excavation, conserve land, and minimize house demolition and population relocation. Generally, during a regional survey of pumped storage power station sites, the first consideration is whether the region's topography and hydrological conditions meet the construction requirements of the upper and lower reservoirs. Only then are constraints such as ecological red lines, basic farmland, and protected areas considered to further refine the selection of pumped storage power station sites.

[0004] From a hydrological perspective, pumped storage power stations require a certain level of water supply. While pumped storage power stations don't require the large runoff of traditional hydropower stations to operate normally, there must be sufficient water sources near the site to accommodate water storage and losses due to evaporation and seepage. From a topographical perspective, the head between the upper and lower reservoirs of a pumped storage power station is generally between 200 and 800 meters. A head less than 200 meters increases the unit installed capacity construction cost, and due to limitations in the manufacturing capabilities of single-stage mixed-flow reversible turbines, the head should not exceed 800 meters. Furthermore, the ratio of the distance between the upper and lower reservoirs to their elevation should generally not exceed 10. A larger ratio results in a longer water conveyance system, higher construction costs, and greater head loss, leading to lower power generation efficiency.

[0005] Currently, when conducting regional hydrological and topographical condition surveys, it is often necessary to rely on manual methods to conduct a thorough search using software such as Google Earth. This is not only inefficient but may also lead to the omission of sites with favorable hydrological and topographical conditions. In response to this problem, the industry has developed some batch site selection programs, but due to algorithmic defects, the site selection results are not satisfactory and cannot meet the industry's demand for batch screening of pumped storage power station sites. Summary of the Invention

[0006] The purpose of this invention is to provide a method for batch planning and site selection of pumped storage power stations based on terrain classification. This invention enables batch screening of upper and lower reservoir sites for pumped storage power stations in a large area.

[0007] The technical solution of this invention is: a method for batch planning and site selection of pumped storage power stations based on terrain classification, comprising the following steps: Step 1: Convert the topographic data of the study area into a digital elevation model (sourceDEM). The accuracy of the sourceDEM raster cells is A. Step 2: Set the neighborhood square window R=A*A, use this window to traverse all cells of the sourceDEM raster, calculate the mean value of the raster cells in the window, and assign it to the current cell to finally form the mean digital elevation model meanDEM; Step 3: Calculate the digital elevation difference model (DEM): VarianceDEM=sourceDEM-meanDEM; Step 4: Classify the terrain based on the varianceDEM, dividing the terrain into valleys, flatlands, and ridges. Areas with varianceDEM>0 are ridges and are excluded. After exclusion, a flat valley digital elevation model pgDEM is formed. Step 5: Based on the digital elevation model sourceDEM, perform fill and cut, flow direction, cumulative flow and reclassification calculations in sequence to generate the regional water system (river); when performing cumulative flow calculations, set the catchment area generation threshold CatchmentA. Step 6: Set the river segment length reachL, divide the regional river system into segments based on the river segment length reachL, generate river segments reach_n, obtain the center point centerPoint_n of each river segment reach_n, and obtain the coordinates and elevation of centerPoint_n; Step 7: Assume the horizontal distance between the upper and lower pumped storage reservoirs is L, the height difference is H, and the ratio of distance to height difference is K, K=L / H; Step 8: Using the center point centerPoint_n of each river segment reach_n as the center point, form a search ring with the range of height differences as the radius. Calculate the distance-to-height ratio K between each grid cell of the Pinggu Digital Elevation Model pgDEM within the ring and centerPoint_n. Eliminate grid cells in the Pinggu Digital Elevation Model pgDEM whose distance-to-height ratio K does not meet the requirements. Traverse all river segments to generate a distance-to-height digital elevation model jgDEM. Step 9: Generate the longest valley (R) for each patch in the jgDEM digital elevation model. Step 10: Calculate the average gradient J of the longest valley (longestR) of each patch in the distance height digital elevation model jgDEM, and use it as a threshold parameter to further filter out distance height digital elevation models jgDEM with an average gradient J of the longest valley (longestR) that is less than a preset value, and generate a gradient digital elevation model J_DEM. Step 11: Locate the centroid C_n of each map segment in the gradient digital elevation model J_DEM. Using the centroid C of each map segment as the center point and the height difference range described in Step 8 as the radius, form a search ring. Calculate the distance-to-height ratio Kn between the ring and the center point centerPoint_n of each river segment. Select the river segment with the minimum value mincenterPoint_n and connect it with the corresponding map segment in the gradient digital elevation model J_DEM to form an upper and lower reservoir pair. This identifies the pumped storage power station site in the study area that meets the hydrological and topographic conditions.

[0008] In step 7 of the aforementioned method for batch planning and site selection of pumped storage power stations based on terrain classification, the distance-to-height ratio K is 2~10, the height difference H is 200~800m, and the horizontal distance L is 400~8000m.

[0009] In step 9 of the aforementioned method for batch planning and site selection of pumped storage power stations based on terrain classification, the specific method for generating the longest valley (longestR) for each patch in the jgDEM (digital elevation model) is as follows: (1) Convert each patch of the distance-height digital elevation model jgDEM into a vector boundary jgBoudary, and use the vector boundary jgBoudary to clip the cumulative flow raster generated in step 5 to generate the clipped cumulative flow raster clip_Acc. (2) Use the vector boundary jgBoudary to clip the flow raster generated in step 5 to generate the clipped flow raster clip_Dir; (3) Locate the maximum value raster cell maxcell_Acc of each pattern in the cumulative flow raster clip_Acc, and combine it with the flow direction raster clip_Dir to find the next cell nextcell_Acc that flows into this raster cell and has the largest cumulative flow raster value; (4) Based on nextcell_Acc, repeat step (3) until the longest valley (longestR) of each patch in the jgDEM digital elevation model is generated. Beneficial effects

[0010] This invention innovatively adopts a terrain classification method for batch selection of pumped storage power station sites. Combined with the height ratio of pumped storage power stations and the gradient of the longest valley in the upper reservoir, it ensures automated site selection without overlooking sites with good conditions. This solves the industry's need for batch screening of pumped storage power station sites. Compared with traditional manual screening, the screening efficiency is increased by more than 10 times, and the omission of high-quality sites can be greatly avoided. Attached Figure Description

[0011] Figure 1 This is a flowchart of the batch planning and site selection method for pumped storage power stations based on terrain classification according to the present invention.

[0012] Figure 2 This is a map showing the results of using this method to plan and select sites for pumped storage power stations in a certain study area. Implementation

[0013] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this should not be construed as limiting the present invention.

[0014] Example 1. A method for batch planning and site selection of pumped storage power stations based on terrain classification, see [link to example]. Figure 1 It includes the following steps: Step 1: Obtain the topographic data of the study area and convert it into a digital elevation model (sourceDEM). The accuracy of the sourceDEM raster cell is A (m). Step 2: Set the neighborhood square window R=A*A, use this window to traverse all cells of the sourceDEM raster, calculate the mean value of the raster cells in the window, and assign it to the current cell to finally form the mean digital elevation model meanDEM; Step 3: Calculate the digital elevation difference model (DEM): VarianceDEM = sourceDEM - meanDEM Step 4: Based on the varianceDEM (Digital Elevation Difference Model), the terrain is classified into valleys, flatlands, and ridges. Areas suitable for constructing upper reservoirs of pumped storage power stations are generally valleys and flatlands with gentle slopes. Ridges have steeper slopes and are not suitable for upper reservoirs. Areas with varianceDEM > 0 are ridges and are excluded. After exclusion, a pgDEM (Digital Elevation Model for Flat Valleys) is formed. The upper reservoir area for pumped storage power stations will be further selected from the pgDEM. Step 5: Based on the digital elevation model (sourceDEM), and using existing algorithms, perform fill / cut, flow direction, cumulative flow, and reclassification calculations sequentially to generate the regional river system. When calculating cumulative flow, set a catchment area generation threshold (CatchmentA (km²)). 2 ) Step 6: Set the river segment length reachL (km), divide the regional river system into segments based on the river segment length reachL, generate river segments reach_n, obtain the center point centerPoint_n of each river segment reach_n, and obtain the coordinates (X_n, Y_n, Z_n) and elevation of centerPoint_n; Step 7: Assume that the horizontal distance between the upper and lower reservoirs of the pumped storage is L, the height difference is H, and the ratio of distance to height difference (distance-to-height ratio) is K, K=L / H. The distance-to-height ratio K is generally 2~10, and the height difference H is generally between 200~800m. Then the horizontal distance L is between 400~8000m. Step 8: Using the center point centerPoint_n of each river segment reach_n as the center point, form a search ring with the range of height differences as the radius. Calculate the distance-to-height ratio K between each grid cell of the Pinggu Digital Elevation Model pgDEM within the ring and centerPoint_n. Eliminate grid cells in the Pinggu Digital Elevation Model pgDEM whose distance-to-height ratio K does not meet the requirements. Traverse all river segments to generate a distance-to-height digital elevation model jgDEM. The upper reservoir area of ​​pumped storage will be further filtered from the distance-to-height digital elevation model jgDEM. Step 9: Generate the longest valley (longestR) for each patch in the jgDEM digital elevation model. The specific method is as follows: (1) Convert each patch of the distance-height digital elevation model jgDEM into a vector boundary jgBoudary, and use the vector boundary jgBoudary to clip the cumulative flow raster generated in step 5 to generate the clipped cumulative flow raster clip_Acc. (2) Use the vector boundary jgBoudary to clip the flow raster generated in step 5 to generate the clipped flow raster clip_Dir; (3) Locate the maximum value raster cell maxcell_Acc of each pattern in the cumulative flow raster clip_Acc, and combine it with the flow direction raster clip_Dir to find the next cell nextcell_Acc that flows into this raster cell and has the largest cumulative flow raster value; (4) Based on nextcell_Acc, repeat step (3) until the longest valley (longestR) of each patch in the jgDEM digital elevation model is generated; Step 10: Calculate the average gradient J of the longest valley (longestR) of each patch in the distance-to-height digital elevation model (jgDEM). Use this as a threshold parameter to further filter out distance-to-height digital elevation models whose average gradient J of the longest valley (longestR) is less than a certain value, and generate the gradient digital elevation model J_DEM. Step 11: Locate the centroid C_n of each map section in the digital elevation model J_DEM of the gradient. Using the centroid C of each map section as the center point and the height difference range described in Step 8 as the radius, form a search ring. Calculate the distance-to-height ratio Kn between the ring and the center point centerPoint_n of each river segment. Select the river segment with the minimum value mincenterPoint_n and connect it with the corresponding map section of the digital elevation model J_DEM of the gradient to form an upper and lower reservoir pair (upper reservoir C_n, lower reservoir mincenterPoint_n). This identifies pumped storage power station sites in the study area that meet the hydrological and topographical conditions. Based on this, further investigation of geological and sensitive factors is conducted to determine available sites.

[0015] Example 2. A method for batch site selection of pumped storage power stations based on terrain classification. For example... Figure 1 As shown, the method process covers the preparation of topographic data of the study area, generation of mean digital elevation model, generation of digital elevation difference model, generation of flat valley digital elevation model, generation of regional water system, segmentation of water system, determination of distance-to-elevation ratio range, calculation of upper reservoir raster range that meets the distance-to-elevation ratio condition, generation of the longest valley of upper reservoir raster patch, calculation of the gradient of the longest valley for further screening, recalculation of distance-to-elevation ratio and selection of upper and lower reservoirs with the smallest distance-to-elevation ratio.

[0016] The specific operation steps are illustrated below: Step 1: Obtain the topographic data of the study area and convert it into a digital elevation model (sourceDEM). The sourceDEM raster cell accuracy is 30 (m). Step 2: Set the neighborhood square window R=30*30m, use this window to traverse all cells of the sourceDEM raster, calculate the mean value of the raster cells in the window, and assign it to the current cell to finally form the mean digital elevation model meanDEM; Step 3: Calculate the digital elevation difference model (DEM): VarianceDEM = sourceDEM - meanDEM Step 4: Based on the varianceDEM (Digital Elevation Difference Model), the terrain is classified into valleys, flatlands, and ridges. Areas suitable for constructing upper reservoirs of pumped storage power stations are generally valleys and flatlands with gentle slopes. Ridges have steeper slopes and are not suitable for upper reservoirs. Areas with varianceDEM > 0 are ridges and are excluded. After exclusion, a pgDEM (Digital Elevation Model for Flat Valleys) is formed. The upper reservoir area for pumped storage power stations will be further selected from the pgDEM. Step 5: Based on the digital elevation model sourceDEM, using existing algorithms, perform fill and cut calculations, flow direction calculations, cumulative flow calculations, and reclassification calculations sequentially to generate the regional river system. When performing cumulative flow calculations, set the catchment area generation threshold CatchmentA = 10 (km²). Step 6: Set the river segment length reachL=1 (km), divide the regional river system into segments based on the river segment length reachL, generate river segments reach_n, obtain the center point centerPoint_n of each river segment reach_n, and obtain the coordinates and elevation (X_n,Y_n,Z_n) of centerPoint_n. Step 7: Assume that the horizontal distance between the upper and lower reservoirs of the pumped storage is L, the height difference is H, and the ratio of distance to height difference (distance-to-height ratio) is K, K=L / H. The distance-to-height ratio K is generally 2~10, and the height difference H is generally between 200~800m. Then the horizontal distance L is between 400~8000m. Step 8: Using the center point centerPoint_n of each river segment reach_n as the center point, form a search ring with a radius of 200~800m for the height difference range. Calculate the distance-to-height ratio K between each grid cell of the Pinggu Digital Elevation Model pgDEM within the ring and centerPoint_n. Eliminate grid cells in the Pinggu Digital Elevation Model pgDEM whose distance-to-height ratio K value does not meet the requirements. Traverse all river segments to generate a distance-to-height digital elevation model jgDEM. The upper reservoir area of ​​pumped storage will be further filtered from the distance-to-height digital elevation model jgDEM. Step 9: Generate the longest valley (longestR) for each patch in the jgDEM digital elevation model. The specific method is as follows: (1) Convert each patch of the distance-height digital elevation model jgDEM into a vector boundary jgBoudary, and use the vector boundary jgBoudary to clip the cumulative flow raster generated in step 5 to generate the clipped cumulative flow raster clip_Acc. (2) Use the vector boundary jgBoudary to clip the flow raster generated in step 5 to generate the clipped flow raster clip_Dir; (3) Locate the maximum value raster cell maxcell_Acc of each pattern in the cumulative flow raster clip_Acc, and combine it with the flow direction raster clip_Dir to find the next cell nextcell_Acc that flows into this raster cell and has the largest cumulative flow raster value; (4) Based on nextcell_Acc, repeat step (3) until the longest valley (longestR) of each patch in the jgDEM digital elevation model is generated; Step 10: Calculate the average gradient J of the longest valley (longestR) of each patch in the distance-to-height digital elevation model jgDEM, and use it as a threshold parameter to further filter out distance-to-height digital elevation models whose average gradient J of the longest valley (longestR) is less than 0.1, and generate the gradient digital elevation model J_DEM. Step 11: Locate the centroid C_n of each map section in the digital elevation model J_DEM (Digital Elevation Model of Gradient). Using the centroid C of each map section as the center point and the height difference range described in Step 8 as the radius, form a search ring. Calculate the distance-to-height ratio Kn between the ring and the center point centerPoint_n of each river segment. Select the river segment with the minimum value mincenterPoint_n and connect it with the corresponding map section in the digital elevation model J_DEM to form an upper and lower reservoir pair (upper reservoir C_n, lower reservoir mincenterPoint_n). This identifies pumped storage power station sites within the study area that meet the hydrological and topographical conditions. See [link to relevant documentation]. Figure 2 Based on this, further investigations will be conducted on geological conditions and sensitive factors to determine usable site locations.

Claims

1. A method for batch planning and site selection of pumped storage power stations based on terrain classification, characterized in that, Includes the following steps: Step 1: Convert the topographic data of the study area into a digital elevation model (sourceDEM). The accuracy of the sourceDEM raster cells is A. Step 2: Set the neighborhood square window R=A*A, use this window to traverse all cells of the sourceDEM raster, calculate the mean value of the raster cells in the window, and assign it to the current cell to finally form the mean digital elevation model meanDEM; Step 3: Calculate the digital elevation difference model (DEM): VarianceDEM=sourceDEM-meanDEM; Step 4: Classify the terrain based on the varianceDEM, dividing the terrain into valleys, flatlands, and ridges. Areas with varianceDEM>0 are ridges and are excluded. After exclusion, a flat valley digital elevation model pgDEM is formed. Step 5: Based on the digital elevation model sourceDEM, perform fill and cut, flow direction, cumulative flow and reclassification calculations in sequence to generate the regional water system (river); when performing cumulative flow calculations, set the catchment area generation threshold CatchmentA. Step 6: Set the river segment length reachL, divide the regional river system into segments based on the river segment length reachL, generate river segments reach_n, obtain the center point centerPoint_n of each river segment reach_n, and obtain the coordinates and elevation of centerPoint_n; Step 7: Assume the horizontal distance between the upper and lower pumped storage reservoirs is L, the height difference is H, and the ratio of distance to height difference is K, K=L / H; Step 8: Using the center point centerPoint_n of each river segment reach_n as the center point, form a search ring with the range of height differences as the radius. Calculate the distance-to-height ratio K between each grid cell of the Pinggu Digital Elevation Model pgDEM within the ring and centerPoint_n. Eliminate grid cells in the Pinggu Digital Elevation Model pgDEM whose distance-to-height ratio K does not meet the requirements. Traverse all river segments to generate a distance-to-height digital elevation model jgDEM. Step 9: Generate the longest valley (R) for each patch in the jgDEM digital elevation model. Step 10: Calculate the average gradient J of the longest valley (longestR) of each patch in the distance height digital elevation model jgDEM, and use it as a threshold parameter to further filter out distance height digital elevation models jgDEM with an average gradient J of the longest valley (longestR) that is less than a preset value, and generate a gradient digital elevation model J_DEM. Step 11: Locate the centroid C_n of each map segment in the gradient digital elevation model J_DEM. Using the centroid C of each map segment as the center point and the height difference range described in Step 8 as the radius, form a search ring. Calculate the distance-to-height ratio Kn between the ring and the center point centerPoint_n of each river segment. Select the river segment with the minimum value mincenterPoint_n and connect it with the corresponding map segment in the gradient digital elevation model J_DEM to form an upper and lower reservoir pair. This identifies the pumped storage power station site in the study area that meets the hydrological and topographic conditions.

2. The method for batch planning and site selection of pumped storage power stations based on terrain classification according to claim 1, characterized in that, In step 7, the distance-to-height ratio K is 2~10, the height difference H is 200~800m, and the horizontal distance L is 400~8000m.

3. The method for batch planning and site selection of pumped storage power stations based on terrain classification according to claim 1, characterized in that, In step 9, the specific method for generating the longest valley (longestR) for each patch in the jgDEM digital elevation model is as follows: (1) Convert each patch of the distance-height digital elevation model jgDEM into a vector boundary jgBoudary, and use the vector boundary jgBoudary to clip the cumulative flow raster generated in step 5 to generate the clipped cumulative flow raster clip_Acc. (2) Use the vector boundary jgBoudary to clip the flow raster generated in step 5 to generate the clipped flow raster clip_Dir; (3) Locate the maximum value raster cell maxcell_Acc of each pattern in the cumulative flow raster clip_Acc, and combine it with the flow direction raster clip_Dir to find the next cell nextcell_Acc that flows into this raster cell and has the largest cumulative flow raster value; (4) Based on nextcell_Acc, repeat step (3) until the longest valley (longestR) of each patch in the jgDEM digital elevation model is generated.

Images