BIM dynamic model construction method for pumped storage projects

Abstract

The present application provides a method for constructing a BIM dynamic model for a pumped storage project, comprising the following steps: S1, obtaining BIM model data of the pumped storage project, and mapping it into a three-dimensional space coordinate system to construct a BIM three-dimensional model; S2, obtaining GIS data, and mapping it according to the geographic space location of the pumped storage project to construct a GIS three-dimensional model; S3, fusing the BIM three-dimensional model with the GIS three-dimensional model to obtain a fusion model; S4, obtaining the hydraulic parameters of the pumped storage project, and inputting them into the fusion model of the pumped storage project; S5, obtaining the current reservoir capacity prediction result; The BIM dynamic model is constructed by the above steps S1 to S5, and the method improves the construction accuracy of the BIM dynamic model of the pumped storage project. The patent is stable and reliable in operation and is suitable for guiding the production and operation and maintenance management of the pumped storage project.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic model construction of pumped storage engineering, and in particular to a BIM dynamic model construction method for pumped storage engineering. Background Art

[0002] The current dynamic model involves a lot of data fusion, some of which are difficult to obtain accurately, such as seepage pressure drop or seepage parameters. In addition, some core parameters that have a great impact on the dynamic modeling of pumped storage projects are also difficult to obtain, such as seepage velocity, seepage slope parameters, and seepage volume parameters. The commonly used analysis methods are mainly hydraulics and flow network methods.

[0003] In the prior art, the permeability coefficient is a comprehensive indicator reflecting the seepage characteristics of the soil. The size of the permeability coefficient mainly depends on the particle shape, size, non-uniformity coefficient and water temperature of the soil. It is generally determined by empirical methods, indoor measurement methods and field measurement methods, which is difficult to be accurately obtained, resulting in a large gap between the established dynamic model of the pumped storage project and the actual situation, and it is difficult to use to guide production and operation and maintenance. CN114884101A records a pumped storage scheduling method based on adaptive model control prediction, and proposes to use an adaptive function of real-time load fluctuations to determine the final state storage capacity constraint value of the pumped storage capacity in each scheduling cycle under the real-time power market, and use the final state storage capacity constraint value and model predictive control to perform rolling clearing on the target scheduling clearing model to schedule the output of each unit. However, the document does not give how to obtain the final state storage capacity constraint value and the specific meaning of the final state storage capacity constraint value, that is, the model needs to rely on more accurate storage capacity data, and only with the measured storage capacity data, it is difficult to guide future production in advance. Therefore, a method that can accurately obtain the final state storage capacity data is urgently needed to ensure the more reliable operation of the model. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a BIM dynamic model construction method for pumped storage projects, which can integrate GIS data to construct a high-precision BIM dynamic model that meets the on-site geographical conditions, and can more accurately predict the most important parameter storage capacity data of the pumped storage project to guide the high-efficiency and high-effective production and operation and maintenance management of the pumped storage project, and provide basic data for other management models.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a BIM dynamic model construction method for pumped storage engineering, comprising the following steps:

[0006] S1. Obtain BIM model data of the pumped storage project and construct a BIM three-dimensional model;

[0007] S2, obtain GIS data and construct a GIS three-dimensional model with digital elevation data;

[0008] S3, integrating the BIM 3D model with the GIS 3D model to obtain a fusion model;

[0009] Among them, a reference point is set in the BIM three-dimensional model, and the coordinates of the reference point are converted into coordinates in the unified coordinate system of the GIS three-dimensional model;

[0010] During the fusion process of BIM 3D model and GIS 3D model, addition and intersection operations are performed. Addition operations are performed at positions where the BIM 3D model and GIS 3D model do not interfere with each other, and intersection operations are performed at positions where they interfere with each other, and the data corresponding to the BIM 3D model at the interfering positions is retained;

[0011] S4, obtaining hydraulic parameters of the pumped storage project and inputting them into the fusion model of the pumped storage project;

[0012] Hydraulic parameters include: pumping flow, surface runoff, rainfall, production flow, flood discharge, infiltration and evaporation;

[0013] The hydraulic parameters are used to predict the reservoir capacity of the pumped storage project in a specific period of time in the future. The specific calculation formula is:

[0014] ;

[0015] In the formula, is the storage capacity of the reservoir in a specific period of time, is the pumping flow rate in a specific period of time, is the runoff volume in a specific period of time, is the permeation volume in a specific period of time, is the production flow in a specific period of time, is the evaporation in a specific period of time, is the flood discharge in a specific period of time;

[0016] S5. Obtain the storage capacity of the reservoir within a specific period of time;

[0017] The BIM dynamic model is constructed through the above steps S1 to S5.

[0018] In a preferred solution, the GIS data in step S2 includes: digital elevation data, digital orthophoto data, vector data and oblique photography data;

[0019] Constructing a GIS three-dimensional model also includes: converting the format of GIS data, converting the coordinate system and aligning the data, constructing a unified coordinate system, and establishing a GIS three-dimensional model under the unified coordinate system.

[0020] In a preferred solution, lightweight processing is performed on the fusion model in step S3, including:

[0021] Filter and delete redundant data;

[0022] Reconstruct some fusion models in a topological manner to reduce the data surface;

[0023] Reduce redundant point, line and surface data by clustering;

[0024] Reduce the dimension of complex curves by fitting;

[0025] Simplify the geometric information and attribute information of the fusion model;

[0026] The lightweight fusion model is used to generate a BIM dynamic model, and the BIM dynamic model is used as the basis for the subsequent generation of a digital twin model.

[0027] In the preferred solution, the runoff volume within a specific period of time The calculation formula is:

[0028] ;

[0029] is the surface runoff in a specific period of time, The amount of rainfall during a specific period of time.

[0030] In the preferred solution, in order to calculate the evaporation of water in the reservoir in a specific period of time, the reference evaporation is first calculated. The product of the reference evaporation and the specific period ΔT is the evaporation in the specific period of time. The calculation formula of the reference evaporation is:

[0031] ;

[0032] Where, E is the reference evaporation, is the net radiation, G is the soil heat flux, γ is the humidity calculation constant, T is the mean temperature, is the wind speed at a height of 2 meters, is the saturated vapor pressure, is the actual water vapor pressure, and Δ is the slope of the saturated water vapor pressure curve.

[0033] In the preferred solution, the calculation method of flood discharge depends on the type of flood discharge facility, specifically:

[0034] Spillway: When there is no gate control or the gate is fully open, the spillway discharge formula is:

[0035] ;

[0036] in, is the discharge flow of the spillway, unit: m³ / s; C is the discharge coefficient; B is the net width of the weir top, unit: m; H is the water head above the weir, unit: m;

[0037] Flood discharge tunnel: The discharge flow formula of the flood discharge tunnel is:

[0038] ;

[0039] in, is the discharge flow of the flood discharge tunnel. is the cross-sectional area of ​​the spillway tunnel, and H is the calculated water head.

[0040] In the preferred solution, the method for calculating the infiltration volume of water in the reservoir within a specific period of time is:

[0041] ; In the formula, is the infiltration volume in a specific period of time, K is the infiltration parameter, A is the water flow section, H is the total head loss, L is the length of the seepage path, and ΔT is the specific period of time.

[0042] In a preferred solution, the permeability parameter K is obtained by a field test method, and the field test specifically includes:

[0043] Water injection test: by continuously injecting water into the borehole to maintain a certain water level, the permeability coefficient of the aquifer is calculated based on the relationship between the amount of water injected and the water level rise;

[0044] Pumping test: Calculate the aquifer permeability parameter K using the final stable water level value of the pumping well;

[0045] Water pressure test: By injecting water into the borehole and applying pressure, the permeability parameter K is calculated based on the changes in water pressure and water flow.

[0046] In the preferred solution, the specific steps of the pumping test are:

[0047] S01. Arrange a main pumping hole and several observation holes;

[0048] S02. At the beginning, keep the water volume stable and make the water level change continuously or keep the water level stable and make the water volume change continuously, and detect the change of water level at a certain time interval or the change of water volume over time;

[0049] S03. After the pumping is completed, observe the water level or water volume recovery until the fluctuation stabilizes;

[0050] S04, collecting water level or water volume change data of the pumping hole and the observation hole, and calculating the permeability parameter K by fitting and superposition;

[0051] The water level or water volume data h of each observation well at a specific time point is calculated separately using the Jacobs approximation formula:

[0052] ;

[0053] In the formula: pumping rate Q, distance from observation hole to pumping hole r, water level or water volume data at a specific time point h, aquifer thickness B, constant term C; determine the hydraulic conductivity T and calculate the permeability parameter K:

[0054] .

[0055] In the preferred solution, the specific steps of the water pressure test are:

[0056] S11. Select a representative test section, isolate the test section from other parts, and maintain the closedness of the test section;

[0057] S12. Inject high-pressure water into the borehole, gradually increase the pressure, record the water flow rate under different pressures, and calculate the permeability parameter K:

[0058] ;

[0059] Where: the amount of water passing through per unit time Q, the length of the test section L, the cross-sectional area of ​​the test section A, the applied fluid pressure P, and the atmospheric pressure .

[0060] The present invention provides a BIM dynamic model construction method for pumped storage engineering. Compared with the prior art, the present invention has the following beneficial effects:

[0061] 1. The present invention integrates the high-precision BIM three-dimensional model with the GIS three-dimensional model including the geographic spatial location data of the pumped storage project, which facilitates the realization of accurate calculations based on actual production conditions. In particular, in the real physical operation process, simulation and simulation close to the real situation can be carried out. Through the iteration of artificial intelligence algorithms, more accurate data can be gradually approached, thereby providing key data support for refined, efficient and intelligent management.

[0062] 2. The present invention integrates the refinement of the BIM three-dimensional model with the globalization of the GIS three-dimensional model. During the integration process, the BIM three-dimensional model is retained as the preferred structure to obtain a BIM three-dimensional model with the geographic spatial location data of the pumping and storage project, which is convenient for the subsequent direct collection of data in the BIM three-dimensional model for calculation and simulation.

[0063] 3. The present invention introduces water conservancy parameters into the BIM three-dimensional model to predict future reservoir capacity data and provide more accurate reference data for production scheduling. Through the iteration and correction of artificial intelligence, the predicted data can continuously approach the measured data.

[0064] 4. The present invention can realize a digital twin model display that is close to reality, and provide production data and image data for realizing a digital twin model that is close to the actual production conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a flow chart of the BIM dynamic model building method in the present invention.

[0066] Figure 2 It is the GIS three-dimensional model diagram in the present invention.

[0067] Figure 3 It is the BIM three-dimensional model diagram in the present invention.

[0068] Figure 4 It is a fusion model diagram of the pumped storage project in the present invention.

[0069] Figure 5 It is a schematic diagram of the storage capacity change in the fusion model of the present invention. DETAILED DESCRIPTION

[0070] like Figure 1 As shown, a BIM dynamic model construction method for a pumped storage project includes the following steps:

[0071] S1. Obtain the BIM model data of the pumped storage project and construct a BIM three-dimensional model, such as Figure 3 As shown in;

[0072] S2, obtain GIS data and construct GIS 3D model with digital elevation data, such as Figure 2 As shown in;

[0073] S3, integrating the BIM 3D model with the GIS 3D model to obtain a fusion model; Figure 4 As shown in;

[0074] Among them, a reference point is set in the BIM 3D model, and the coordinates of the reference point are converted into coordinates in the unified coordinate system of the GIS 3D model; addition and intersection operations are performed during the fusion process of the BIM 3D model and the GIS 3D model, and addition operations are performed at positions where the BIM 3D model and the GIS 3D model do not interfere with each other, specifically:

[0075] Union(BIM, GIS) ={( x , y , z )|(( x , y , z )∈BIM)∪(( x , y , z )∈GIS)};

[0076] in,( x , y , z) represents the coordinate point in three-dimensional space, and Union(BIM, GIS) represents the union of BIM and GIS.

[0077] Perform intersection operations at the mutually interfering positions and retain the data corresponding to the interfering positions and the BIM 3D model, specifically:

[0078] Intersection(BIM, GIS) ={( x , y , z )|(( x , y , z )∈BIM)∩(( x , y , z )∈GIS)};

[0079] in,( x , y , z ) represents a coordinate point in three-dimensional space, and Intersection(BIM, GIS) represents the intersection of BIM and GIS.

[0080] Preferably, at the location where the BIM 3D model and the GIS 3D model interfere with each other, the BIM 3D model is subjected to outer contour tracking processing, that is, a simple outer contour is used to fit the BIM 3D model at the interference location, and then the fitted contour is intersected with the GIS 3D model. This solution greatly reduces the amount of calculation. After this processing, the fusion model of the pumped storage project is manually modified to eliminate defects.

[0081] Preferably, after the intersection and addition operations are performed, the data corresponding to the BIM 3D model is retained, and after the intersection operation, the data attributes in the BIM 3D model are inherited to the corresponding elements in the fused model. The elements in the fused model remain associated with the original data in the BIM 3D model, and the detailed information of the BIM 3D model can be traced back in subsequent analysis and operations.

[0082] S4, obtaining hydraulic parameters of the pumped storage project, and introducing the hydraulic parameters into the fusion model of the pumped storage project;

[0083] Hydraulic parameters include: pumping flow, surface runoff, rainfall, production flow, flood discharge, infiltration and evaporation;

[0084] Hydraulic parameters are used to predict the reservoir capacity of pumped storage projects in a specific period of time in the future. The specific calculation method is expressed by the following formula:

[0085] ;

[0086] In the formula, is the storage capacity of the reservoir in a specific period of time, is the pumping flow rate in a specific period of time, is the runoff volume in a specific period of time, is the permeation volume in a specific period of time, is the production flow in a specific period of time, is the evaporation in a specific period of time, is the flood discharge in a specific period of time. Figure 5 shown.

[0087] S5. Obtain the storage capacity of the reservoir within a specific period of time;

[0088] The BIM dynamic model of the pumped storage project is constructed through the above steps.

[0089] In the preferred solution, in step S2, the GIS data includes: digital elevation data, digital orthophoto data, vector data and oblique photography data, and the data from different data sources are converted into different formats, converted into coordinate systems and aligned to construct a unified coordinate system, and a GIS three-dimensional model is established under the unified coordinate system, such as Figure 2 In the process of building GIS 3D models, digital elevation data is used.

[0090] In the preferred solution, the fusion model of the pumped storage project is lightweighted, including the following parallel steps:

[0091] Filter and delete redundant data;

[0092] The specific steps include:

[0093] S001. Use the search function to find duplicate vertices and delete them;

[0094] S002. Clean up isolated entities, find isolated entities that are not connected to large volume entities, identify them through connected parts, and delete the isolated entities;

[0095] S003. Optimize the facets and reduce the number of facets. Delete the internal faces below the threshold;

[0096] S004. Use mesh analysis tools to detect and repair overlapping or intersecting patches;

[0097] Reconstruct some fused models in a topological way to reduce data surfaces. This includes deleting unnecessary surfaces: for example, selectively deleting surfaces that do not affect the overall shape. For flat areas, delete redundant surfaces and keep only necessary edges. Re-mesh: Use loop cutting and sliding tools to create a more reasonable edge flow. Ensure that the topology around key features (such as hydraulic tunnels and turbine groups) is tighter and more accurate. Use quadrilateralization: Convert polygons to quadrilaterals as much as possible. Use subdivision surface modifiers to increase the smoothness of the model while keeping the base face count low.

[0098] Through the above steps, more than 30% to 90% of redundant data can be eliminated.

[0099] Reduce redundant point, line and surface data by clustering;

[0100] Search in a clustering manner, for example: replace entities, replace some entities that are too small, such as replacing rods below the threshold with flat parts, saving the structure of bolts and omitting the details of the threads;

[0101] Accordingly, duplicate vertices, isolated entities, overlapping or intersecting faces are searched in a clustering manner and then processed with the same measures to greatly improve processing efficiency.

[0102] Reduce the dimension of complex curves by fitting;

[0103] Fitting graphics are used to fit complex patches, and complex surfaces and curves are converted into geometric graphics, which reduces the amount of calculation and improves the calculation efficiency of the model.

[0104] Simplify the geometry and attribute information of the fusion model; in some global models, do not import the geometry and attribute information of specific devices to improve the calculation efficiency of the global model. The simplified model data in this example only involves the devices that are not involved in the modification.

[0105] The lightweight fusion model is used to generate a BIM dynamic model, and the BIM dynamic model is used as the basis for the subsequent generation of a digital twin model.

[0106] In the preferred solution, the runoff volume within a specific period of time The calculation formula is:

[0107] ;

[0108] is the surface runoff flow in a specific period of time, The amount of rainfall during a specific period of time. The data of the AI ​​​​data is used to predict the current surface runoff flow based on many years of historical data. It is mainly based on the development trend of surface runoff flow in recent years combined with historical data, and then corrected based on this combined with recent meteorological data. For example, if the meteorological data is rainier than the historical data, the surface runoff flow tends to increase, otherwise it will decrease. The data are rainfall data in a relatively short period of time compared to the meteorological data in surface runoff. This is because it is necessary to consider that there is a time delay for rainfall to develop into surface runoff and it is affected by surface seepage. Only rainfall within the reservoir projection area is considered.

[0109] In the preferred solution, in order to calculate the evaporation of water in the reservoir within a specific period of time, the reference evaporation is first calculated, and the calculation formula of the reference evaporation is:

[0110] ;

[0111] Where, E is the reference evaporation, is the net radiation (MJ / m² / day), G is the soil heat flux (MJ / m² / day), γ is the humidity calculation constant (kPa / °C), T is the mean temperature (°C), is the wind speed at 2 meters height (m / s), is the saturated water vapor pressure (kPa), is the actual water vapor pressure (kPa), Δ is the slope of the saturated water vapor pressure curve (kPa / °C); in the formula, the average temperature T directly affects the calculation of evaporation by affecting the saturated water vapor pressure and humidity calculation constant γ.

[0112] The product of the reference evaporation amount and the specific period ΔT is the evaporation amount in the specific period.

[0113] In the preferred solution, the method for calculating the flood discharge depends on the type of flood discharge facility:

[0114] Spillway: When there is no gate control or the gate is fully open, the discharge flow of the spillway can be calculated according to the weir flow formula:

[0115] ;

[0116] in, is the discharge flow of the spillway (m³ / s), C is the discharge coefficient, B is the clear width of the weir top (m), and H is the water head above the weir (m);

[0117] Flood discharge tunnel: The discharge flow of the flood discharge tunnel can be calculated according to the pressurized pipe flow:

[0118] ;

[0119] in, is the discharge flow of the flood discharge tunnel (m³ / s), is the cross-sectional area of ​​the spillway (m²), and H is the calculated water head (m).

[0120] In the preferred solution, the calculation method of the water infiltration in the reservoir within a specific period of time is:

[0121] ;

[0122] In the formula, is the infiltration volume in a specific period of time, K is the infiltration parameter, A is the water flow section, H is the total head loss, L is the length of the seepage path, and ΔT is the specific period of time.

[0123] In a preferred solution, the permeability parameter K is obtained by a field test method, and the field test specifically includes:

[0124] Water injection test: by continuously injecting water into the borehole to maintain a certain water level, the permeability coefficient of the aquifer is calculated based on the relationship between the amount of water injected and the water level rise;

[0125] Pumping test: Calculate the aquifer permeability parameter K using the final stable water level value of the pumping well;

[0126] Water pressure test: By injecting water into the borehole and applying pressure, the permeability parameter K is calculated based on the changes in water pressure and water flow.

[0127] In the preferred solution, the specific steps of the pumping test are:

[0128] S01. Arrange a main pumping hole and multiple observation holes;

[0129] S02. At the beginning, keep the water volume stable and make the water level change continuously or keep the water level stable and make the water volume change continuously, and detect the change of water level at a certain time interval or the change of water volume over time;

[0130] S03. After the pumping is completed, observe the water level or water volume recovery until the fluctuation stabilizes;

[0131] S04, collecting multiple water level or water volume change data of the pumping holes and observation holes, and calculating the permeability parameter K by fitting and superposition;

[0132] The water level or water volume data h of each observation well at a specific time point is calculated separately using the Jacobs approximation formula:

[0133] ;

[0134] In the formula: pumping rate Q, distance from observation hole to pumping hole r, water level or water volume data at a specific time point h, aquifer thickness B, constant term C; determine the hydraulic conductivity T and calculate the permeability parameter K:

[0135] .

[0136] In the preferred solution, the specific steps of the water pressure test are:

[0137] S11. Select a representative test section, isolate the test section from other parts, and maintain the closedness of the test section;

[0138] S12. Inject high-pressure water into the borehole, gradually increase the pressure, record the water flow rate under different pressures, and calculate the permeability parameter K:

[0139] ;

[0140] Where: the amount of water passing through per unit time Q, the length of the test section L, the cross-sectional area of ​​the test section A, the applied fluid pressure P, and the atmospheric pressure .

[0141] Preferably, the fusion model of the pumped storage project is gridded and graded;

[0142] Grid division refers to dividing the fusion model of the pumped storage project into multiple units, each unit using an irregular grid to improve the stability of the model during operation and display;

[0143] Hierarchical processing refers to dividing the fusion model of the pumped storage project into multiple levels so that the fusion model of the pumped storage project presents different levels of detail at different distances or detail levels, that is, setting different grid sizes for the fusion models of pumped storage projects at different levels to optimize rendering performance and maintain visual quality.

[0144] Preferably, an associated unique identification code is created, and the identification code is matched one-to-one with the unit components and attribute information of the fusion model of the pumped storage project to ensure that each unit component has a unique identification code to achieve integrated dynamic management of operation and maintenance data.

[0145] Preferably, according to the measured hydraulic parameters, combined with the field test permeability parameter K, an array of permeability arrays is obtained through an artificial intelligence algorithm;

[0146] The infiltration array is a multi-level data structure used to characterize the infiltration at different locations under different reservoir heights;

[0147] The permeability array is divided into three dimensions. The first dimension represents different elevation layers, the second dimension represents different positions in each layer, and the third dimension represents the amount of change of the corresponding parameter over time.

[0148] Preferably, the specific steps of constructing the three-dimensional seepage data model are:

[0149] S21, solving the infiltration parameter array under different reservoir capacity elevations to obtain the infiltration parameter distribution characteristics;

[0150] The physical information neural network PINNs algorithm is used to solve the seepage velocity. PINNs is used to estimate the permeability parameter K and invert the permeability parameter array under different storage capacity conditions.

[0151] S22. Define the physical model: Use Darcy's law to express the relationship between the seepage velocity v and the pressure gradient. It is known that K is the permeability parameter and p is the fluid pressure. is the gradient operator;

[0152]

[0153] Construct a neural network architecture: Construct a multi-layer perceptron whose input includes spatial coordinate data and time t, and whose output is fluid pressure P and permeability parameter K;

[0154] Define the loss function :

[0155] ,

[0156] Among them, the loss function contains the physical residual term , which is the residual of the above physical model, is the boundary condition term, that is, the deviation of the output on the boundary, is the initial condition term, is the data fitting term;

[0157] Training model: training model data through optimization algorithm to minimize loss function;

[0158] Parameter inversion: After training is completed, the fluid pressure P output by the neural network is used to invert the permeability parameter K under different storage capacity conditions.

[0159] Based on the permeability parameter array, the permeability parameter K of each part is obtained by using the arbitrary section interpolation grid method, and the permeability distribution characteristics under various operating conditions are analyzed.

[0160] S23, establishing a three-dimensional seepage data model based on the initial three-dimensional model and the permeability parameter K;

[0161] The cross-section node control method is used to form a three-dimensional unit grid;

[0162] The flow unit types are divided by FZI method, R35 method, multi-parameter cluster analysis method, stepwise discriminant function method or variance analysis method. The finer the flow unit is divided, the higher the accuracy of the permeability parameter K.

[0163] S24, integrating the permeability array, calculating the known water head boundary, seepage boundary and impermeable boundary, and obtaining a three-dimensional seepage data model.

[0164] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A BIM dynamic model construction method for pumped storage engineering, characterized by: The following steps are involved: S1. Obtain BIM model data of the pumped storage project and construct a BIM three-dimensional model; S2, obtain GIS data and construct a GIS three-dimensional model with digital elevation data; S3, integrating the BIM 3D model with the GIS 3D model to obtain a fusion model; Among them, a reference point is set in the BIM three-dimensional model, and the coordinates of the reference point are converted into coordinates in the unified coordinate system of the GIS three-dimensional model; During the fusion process of BIM 3D model and GIS 3D model, addition and intersection operations are performed. Addition operations are performed at positions where the BIM 3D model and GIS 3D model do not interfere with each other, and intersection operations are performed at positions where they interfere with each other, and the data corresponding to the BIM 3D model at the interfering positions is retained; S4, obtaining hydraulic parameters of the pumped storage project and inputting them into the fusion model of the pumped storage project; Hydraulic parameters include: pumping flow, surface runoff, rainfall, production flow, flood discharge, infiltration and evaporation; The hydraulic parameters are used to predict the reservoir capacity of the pumped storage project in a specific period of time in the future. The specific calculation formula is: ； In the formula, is the storage capacity of the reservoir in a specific period of time, is the pumping flow rate in a specific period of time, is the runoff volume in a specific period of time, is the permeation volume in a specific period of time, is the production flow in a specific period of time, is the evaporation in a specific period of time, is the flood discharge in a specific period of time; The calculation method for the infiltration of water in the reservoir during a specific period of time is: ； In the formula, is the infiltration volume in a specific period of time, K is the infiltration parameter, A is the water flow section, H is the total head loss, L is the length of the seepage path, and ΔT is the specific period of time; The permeability parameter K is obtained by field test, which specifically includes: Water injection test: by continuously injecting water into the borehole to maintain a certain water level, the permeability coefficient of the aquifer is calculated based on the relationship between the amount of water injected and the water level rise; Pumping test: Calculate the aquifer permeability parameter K using the final stable water level value of the pumping well; Water pressure test: By injecting water into the borehole and applying pressure, the permeability parameter K is calculated based on the changes in water pressure and water flow; The specific steps of the pumping test are: S01. Arrange a main pumping hole and several observation holes; S02. At the beginning, keep the water volume stable and make the water level change continuously or keep the water level stable and make the water volume change continuously, and detect the change of water level at a certain time interval or the change of water volume over time; S03. After the pumping is completed, observe the water level or water volume recovery until the fluctuation stabilizes; S04. Collect the water level or water volume change data of the pumping holes and observation holes, and calculate the permeability parameter K by fitting and superposition; The water level or water volume data h of each observation hole at a specific time point is calculated separately using the Jacobs approximation formula: ； In the formula: pumping rate Q, distance from observation hole to pumping hole r, water level or water volume data at a specific time point h, aquifer thickness B, constant term C; determine the hydraulic conductivity T and calculate the permeability parameter K: ； S5. Obtain the storage capacity of the reservoir within a specific period of time; The BIM dynamic model is constructed through the above steps S1 to S5.

2. The BIM dynamic model construction method for pumped storage engineering according to claim 1 is characterized by: The GIS data in step S2 includes: digital elevation data, digital orthophoto data, vector data and oblique photography data; Constructing a GIS three-dimensional model also includes: converting the format of GIS data, converting the coordinate system and aligning the data, constructing a unified coordinate system, and establishing a GIS three-dimensional model under the unified coordinate system.

3. The BIM dynamic model construction method for pumped storage engineering according to claim 1 is characterized by: The fusion model in step S3 is lightweighted, including: Filter and delete redundant data; Reconstruct some fusion models in a topological manner to reduce the data surface; Reduce redundant point, line and surface data by clustering; Reduce the dimension of complex curves by fitting; Simplify the geometric information and attribute information of the fusion model; The lightweight fusion model is used to generate a BIM dynamic model, and the BIM dynamic model is used as the basis for the subsequent generation of a digital twin model.

4. The BIM dynamic model construction method for pumped storage engineering according to claim 1 is characterized by: The amount of runoff during a specific period of time The calculation formula is: ； is the surface runoff in a specific period of time, The amount of rainfall during a specific period of time.

5. The BIM dynamic model construction method for pumped storage engineering according to claim 1 is characterized by: In order to calculate the evaporation of water in the reservoir in a specific period of time, the reference evaporation is first calculated. The product of the reference evaporation and the specific period ΔT is the evaporation in the specific period. The calculation formula of the reference evaporation is: ； Where, E is the reference evaporation, R n is the net radiation, G is the soil heat flux, γ is the humidity calculation constant, T is the mean temperature, is the wind speed at a height of 2 meters, is the saturated vapor pressure, is the actual water vapor pressure, and Δ is the slope of the saturated water vapor pressure curve.

6. The method for constructing a BIM dynamic model for a pumped storage project according to claim 1, characterized in that: The method for calculating the flood discharge depends on the type of flood discharge facility, specifically: Spillway: When there is no gate control or the gate is fully open, the spillway discharge formula is: ； in, is the discharge flow of the spillway, unit: m³ / s; C is the discharge coefficient; B is the net width of the weir top, unit: m; H is the water head on the weir, unit: m; Flood discharge tunnel: The discharge flow formula of the flood discharge tunnel is: ； in, is the discharge flow of the spillway tunnel, ω is the cross-sectional area of ​​the spillway tunnel, and H is the calculated water head.

7. The method for constructing a BIM dynamic model for a pumped storage project according to claim 1, characterized in that: The specific steps of the water pressure test are: S11. Select a representative test section, isolate the test section from other parts, and maintain the closedness of the test section; S12. Inject high-pressure water into the borehole, gradually increase the pressure, record the water flow rate under different pressures, and calculate the permeability parameter K: ； Where: the amount of water passing through per unit time Q, the length of the test section L, the cross-sectional area of ​​the test section A, the applied fluid pressure P, and the atmospheric pressure .

Images