Dynamic simulation method of rainwater runoff based on BIM-GIS coupling

Abstract

The application relates to the technical field of rainwater runoff simulation, and discloses a rainwater runoff dynamic simulation method based on BIM-GIS coupling. The method plans a simulation process based on input parameters and environmental conditions, determines required BIM data (containing building geometric data) and GIS data (containing terrain elevation data), and determines different scene types. Then, a dynamic simulation process set is generated, the execution sequence is determined, the coupling parameter setting list and the data fusion scheme list of each scene are generated, and finally, the overall scheme is generated by associating each element. In addition, the method also relates to dynamic simulation boundary determination, data subset extraction, scene type subdivision, simulation process updating, data fusion rules, database establishment and maintenance and the like. The method combines the advantages of BIM and GIS technologies, can adapt to different scenes, realizes dynamic data updating and efficient processing, improves the accuracy and reliability of rainwater runoff dynamic simulation, and can be used in the field of urban rainwater management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rainwater runoff simulation, in particular to a rainwater runoff dynamic simulation method based on BIM-GIS coupling. BACKGROUND

[0002] With the rapid advancement of urbanization, the urban underlying surface conditions have changed significantly, and the impervious area is increasing, leading to increasingly prominent urban rainwater runoff problems, such as frequent waterlogging disasters and aggravated non-point source pollution. Accurate simulation of the dynamic process of rainwater runoff has important guiding significance for urban drainage system planning, waterlogging prevention, and water environment governance.

[0003] Traditional rainwater runoff simulation methods have many limitations. On the one hand, although single BIM (Building Information Modeling) technology can finely describe the geometric characteristics and spatial relationships of buildings and their surrounding facilities, it has limitations in handling large-scale topography, hydrology, and meteorology, and cannot accurately reflect regional-scale rainwater runoff characteristics. On the other hand, although GIS (Geographic Information System) technology is good at handling spatial geographic data and macro topography information, it has limited ability to express the internal structure of buildings and surrounding micro-facilities, and cannot meet the requirements of rainwater runoff simulation in complex urban environments.

[0004] Existing simulation methods often cannot effectively deal with complex scenarios under different rainfall conditions and environmental changes. For example, in high-flow rainfall scenarios, traditional methods lack accuracy in simulating runoff peak values and confluence processes; and in different permeability surface scenarios, the calculation of runoff rate and infiltration volume lacks pertinence, resulting in a large deviation between simulation results and actual conditions. At the same time, existing technologies have defects in data fusion and dynamic updating, and cannot integrate multi-source data in real time and adjust the simulation process according to actual monitoring results, affecting the timeliness and accuracy of the simulation results.

[0005] In terms of data processing, traditional methods lack effective data boundary division and dynamic subset extraction mechanisms, resulting in low efficiency when processing large-scale data, and cannot accurately focus on key simulation areas. In addition, the comprehensive utilization of meteorological change data and surface coverage data is insufficient, and various factors affecting rainwater runoff cannot be fully considered, limiting the applicability and reliability of the simulation method. SUMMARY

[0006] The purpose of the present application is to provide a rainwater runoff dynamic simulation method based on BIM-GIS coupling to solve the problems raised in the background.

[0007] To achieve the above purpose, the present application provides the following technical solution: a rainwater runoff dynamic simulation method based on BIM-GIS coupling, the method comprising:

[0008] determining BIM data content and GIS data content required for simulation based on the input parameters and environmental conditions of the rainwater runoff, the BIM data content including building geometry data, and the GIS data content including terrain elevation data; determining different scenario types of the rainwater runoff according to the input parameters and environmental conditions of the rainwater runoff; generating a rainwater runoff dynamic simulation process set according to the different scenario types of the rainwater runoff, the BIM data content, and the GIS data content, and determining an execution order of the dynamic simulation process set based on an optimal coupling principle; generating a coupling parameter setting list of each scenario according to a preset simulation period, the different scenario types of the rainwater runoff, and rainwater runoff performance index requirements in each scenario, and generating a data fusion scheme list of each scenario according to a preset fusion rule, the different scenario types of the rainwater runoff, and rainwater runoff performance index requirements in each scenario; associating the coupling parameter setting list of each scenario, the data fusion scheme list, and rainwater runoff dynamic simulation content based on the execution order of the dynamic simulation process set, and generating an overall scheme of the rainwater runoff dynamic simulation.

[0009] Preferably, the dynamic simulation process of the rainwater runoff is planned based on the input parameters and environmental conditions of the rainwater runoff, and the BIM data content and GIS data content required for simulation are determined, including: for a current scenario, determining a dynamic simulation boundary of the current scenario according to historical simulation data of the current scenario, historical simulation data of adjacent scenarios, and a preset scenario division threshold; extracting a dynamic simulation data subset of the current scenario according to the dynamic simulation boundary and a preset time window; and determining the content of the BIM data content and the GIS data content within the dynamic simulation data subset as rainwater runoff dynamic simulation content of the current scenario.

[0010] Preferably, the different scenario types of the rainwater runoff are determined according to the input parameters and environmental conditions of the rainwater runoff, including: for a current scenario, in a case where a rainfall intensity of the current scenario is greater than a preset intensity threshold, determining the current scenario as a high-flow scenario; in a case where the rainfall intensity of the current scenario is not greater than the preset intensity threshold, classifying runoff rate and infiltration data of the current scenario based on a first preset classification algorithm to obtain each performance category data and determine a center feature of each performance category data; determining a simulation potential level of the current scenario according to environmental conditions of the current scenario; in a case where a matching degree of the center feature of each performance category data and a preset performance category is greater than a preset matching threshold, determining the current scenario as a specific performance scenario; and in a case where the matching degree is not greater than the preset matching threshold, determining the current scenario as a common performance scenario.

[0011] Preferably, the rainwater runoff dynamic simulation content further comprises surface cover data and meteorological change data, different scenario types of the rainwater runoff comprise a high permeability scenario, a low permeability scenario and a critical scenario, and a rainwater runoff dynamic simulation process set is generated according to different scenario types of the rainwater runoff, BIM data content and GIS data content, comprising: for the current scenario, classifying meteorological change data of the current scenario based on a second preset classification algorithm to obtain meteorological category data; for the current meteorological category data, filtering the current meteorological category data based on an influence degree and a change frequency of each meteorological factor in the current meteorological category data to obtain a key meteorological subset of the current meteorological category data, and determining an influence label of the current meteorological category data according to an influence feature of the key meteorological subset and a mean value of rainwater runoff performance indexes; and determining whether the scenario type of the current scenario is the high permeability scenario.

[0012] Preferably, after determining whether the scenario type of the current scenario is the high permeability scenario, the method further comprises: if the scenario type of the current scenario is the high permeability scenario, matching the influence label of each meteorological category data with rainwater runoff performance indexes to generate a targeted optimization strategy; if the scenario type of the current scenario is the low permeability scenario, generating an improvement optimization strategy based on a preset improvement rule and a performance improvement target; and if the scenario type of the current scenario is the critical scenario, generating a performance stability strategy according to historical simulation fluctuations and performance stability rules.

[0013] Preferably, after generating the rainwater runoff dynamic simulation process set, the method further comprises: updating rainwater runoff dynamic simulation content of each scenario based on real-time monitoring data of rainwater runoff; recalculating scenario types of the scenarios according to the updated rainwater runoff dynamic simulation content; and dynamically adjusting an execution order of the dynamic simulation process set according to the recalculated scenario types.

[0014] Preferably, the preset fusion rule comprises: dynamically adjusting a sample size and a repetition number of data fusion based on a confidence interval of rainwater runoff historical fusion data; and automatically triggering a supplementary fusion module when a deviation between measured runoff data and coupled simulation results exceeds a preset error threshold.

[0015] Preferably, the preset fusion rule further comprises: establishing a data acquisition device calibration compensation mechanism, and compensating and calibrating data acquisition devices based on reference values of standard rain gauges and flowmeters before each fusion.

[0016] Preferably, after generating the overall scheme of the rainwater runoff dynamic simulation, the method further comprises: establishing a rainwater runoff dynamic simulation database, storing the coupling parameter setting list, data fusion scheme list, rainwater runoff dynamic simulation content and scene type data in the overall scheme into the database; regularly cleaning and maintaining the data in the database to remove invalid data and duplicate data; based on the database, long-term tracking and comparative analysis of the rainwater runoff dynamic simulation results are performed to evaluate the trend of the rainwater runoff performance over time.

[0017] Preferably, the second preset classification algorithm is a K-means clustering algorithm, and the specific steps of classifying the meteorological change data of the current scene comprise: calculating the distance between each sample in the meteorological change data; assigning the samples to different cluster centers according to the distance to form initial categories; according to the preset number of clusters or the number of cluster iterations, optimizing the cluster centers and updating the sample assignment to obtain the meteorological category data.

[0018] Compared with the prior art, the method has the following beneficial effects:

[0019] In terms of data processing and simulation planning, the method plans the simulation process based on the input parameters and environmental conditions of the rainwater runoff, can determine the dynamic simulation boundary according to the historical simulation data of the current scene, the historical simulation data of the adjacent scene and the preset scene division threshold, and then extract the dynamic simulation data subset to accurately determine the BIM data content and GIS data content required for simulation. This way effectively improves the efficiency and pertinence of data processing, avoids the blindness of large-scale data processing, and enables the simulation to focus on key areas and key data, laying a foundation for subsequent accurate simulation.

[0020] In terms of scene classification and processing, the method can accurately determine different scene types of rainwater runoff according to rainfall intensity, runoff rate, infiltration data and environmental conditions, including high-flow scenarios, specific performance scenarios, ordinary performance scenarios, high-permeability scenarios, low-permeability scenarios and critical scenarios. For different scene types, corresponding targeted optimization strategies, improved optimization strategies and performance stability strategies are generated, so that the simulation can better adapt to various complex actual situations, improve the flexibility and adaptability of the simulation, and ensure that more accurate simulation results can be obtained in different scenarios.

[0021] Regarding the simulation process and parameter settings, the generated dynamic simulation process set for stormwater runoff determines the execution order based on the optimal coupling principle. It also generates a list of coupling parameter settings and data fusion schemes for each scenario, combined with preset simulation cycles and performance requirements. This organically links the simulation content, parameter settings, and data fusion schemes to form a holistic solution. This systematic design makes the simulation process more orderly and scientific, comprehensively considering the influence of various factors and improving the systematic nature and accuracy of the simulation.

[0022] Regarding data fusion and dynamic updates, the preset fusion rules not only consider the confidence intervals of historical fusion data and dynamically adjust the sample size and repetition frequency of data fusion, but also establish a data acquisition equipment calibration and compensation mechanism to ensure data accuracy. Simultaneously, it can update simulation content based on real-time rainwater runoff monitoring data, recalculate scenario types, and dynamically adjust the execution order of the simulation process, achieving dynamic updates to the simulation. This ensures that simulation results can promptly reflect changes in actual conditions, improving the timeliness and reliability of the simulation.

[0023] In terms of database construction and application, a dynamic simulation database for stormwater runoff is established to store various types of data and is regularly cleaned and maintained. Simultaneously, the simulation results are tracked and compared over a long period based on the database to assess the temporal trends in stormwater runoff performance. This provides rich data support for long-term research and management of stormwater runoff, helps identify potential problems and patterns, and provides a more scientific basis for urban stormwater management decisions. Attached Figure Description

[0024] Figure 1 This is a schematic diagram illustrating the working principle of the dynamic simulation method for rainwater runoff based on BIM-GIS coupling described in this invention.

[0025] Figure 2 Design drawings for dynamic simulation boundary determination and data extraction;

[0026] Figure 3 Design drawings for determining the type of stormwater runoff scenario;

[0027] Figure 4 Design drawings generated for a dynamic simulation process set of rainwater runoff;

[0028] Figure 5 This is a design diagram for data fusion rules and calibration mechanisms. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Please see Figures 1-5 The present invention relates to a method for dynamic simulation of stormwater runoff based on BIM-GIS coupling, the specific implementation steps of which are as follows:

[0031] The dynamic simulation process of rainwater runoff is planned based on the input parameters and environmental conditions of rainwater runoff, and the BIM data content and GIS data content required for the simulation are determined. The BIM data content includes building geometry data, and the GIS data content includes terrain elevation data.

[0032] Different stormwater runoff scenario types are determined based on the input parameters and environmental conditions of stormwater runoff.

[0033] Based on different stormwater runoff scenario types, BIM data content, and GIS data content, a dynamic simulation process set for stormwater runoff is generated, and the execution order of the dynamic simulation process set is determined based on the optimal coupling principle.

[0034] Based on the preset simulation period, different scenarios of rainwater runoff, and rainwater runoff performance index requirements, a list of coupling parameter settings for each scenario is generated. Based on the preset fusion rules, different scenarios of rainwater runoff, and rainwater runoff performance index requirements, a list of data fusion schemes for each scenario is generated.

[0035] Based on the execution order of the dynamic simulation process set, the coupling parameter setting list, data fusion scheme list, and stormwater runoff dynamic simulation content of each scenario are linked together to generate an overall scheme for stormwater runoff dynamic simulation.

[0036] Example 1

[0037] When planning the dynamic simulation of stormwater runoff, for the current scenario, the dynamic simulation boundary needs to be determined based on historical simulation data of the current scenario, historical simulation data of adjacent scenarios, and a preset scenario division threshold. Historical simulation data includes data generated from previous dynamic simulations of stormwater runoff under the same or similar scenarios, such as runoff data under different rainfall conditions and the application of building geometry data. Historical simulation data of adjacent scenarios refers to simulation data from scenarios spatially or temporally close to the current scenario. This data provides a reference for determining the boundary of the current scenario, such as the impact of changes in terrain elevation in adjacent areas on stormwater runoff in the current area. The preset scenario division threshold is a standard value pre-set based on a large amount of historical data and practical experience to define the scenario boundary. The setting of this threshold needs to comprehensively consider various factors, such as the differences in stormwater runoff characteristics under different terrains and the degree of influence of building distribution on runoff. For example, in areas with significant terrain undulations, the preset scenario division threshold may be set smaller to more finely define the scenario boundary; while in areas with relatively flat terrain, the threshold may be correspondingly increased.

[0038] After determining the dynamic simulation boundary, a subset of dynamic simulation data for the current scenario is extracted based on this boundary and a preset time window. The preset time window is a time range set according to the simulation's needs and objectives, such as data extraction within a specific time period of a rainfall event. The setting of this time window needs to be combined with the specific simulation objective. For example, if the goal is to analyze the impact of short-duration heavy rainfall on a certain area, the preset time window might be set to a period from several hours before the rainfall occurs to the end of the rainfall; if the goal is to study the impact of long-term rainfall on regional runoff, the time window might be extended to several days or even months. The dynamic simulation data subset includes various data related to rainwater runoff simulation within this boundary and time window, such as rainfall data within a specific time period, including changes in rainfall amount and intensity over time; and changes in building geometry data within this time period, such as whether new buildings have been constructed or the structure of existing buildings has been altered. These changes will affect the runoff path and collection of rainwater on building surfaces.

[0039] The content of BIM and GIS data within the dynamic simulation data subset is defined as the dynamic simulation content of rainwater runoff in the current scenario. This means selecting building geometry data from the BIM data that falls within the dynamic simulation data subset, such as the building's shape, dimensions, height, and roof slope. This data accurately describes the building's three-dimensional structure and is crucial for simulating rainwater flow and collection on the building surface, as well as the design of the drainage system. For example, the roof slope determines the flow velocity and direction of rainwater on the roof, thus affecting the load on rainwater pipes. The GIS data subset includes topographic elevation data, such as the terrain undulations and contour line distribution of the area. Topographic elevation data is the foundation for simulating surface runoff, determining the flow direction, confluence velocity, and potential waterlogging areas. Additionally, it may include data on the area's water system distribution and soil type. Soil type affects the rainwater infiltration rate, thus influencing the distribution ratio of surface runoff and groundwater runoff.

[0040] In practice, obtaining historical simulation data requires the use of a database system. This system stores all relevant past simulation data. By setting query conditions, such as the scene's geographical location, time range, and rainfall conditions, the required historical simulation data for the current scene and adjacent scenes can be quickly retrieved. The preset scene division thresholds need to be comprehensively analyzed and set by professional engineers based on historical meteorological data, terrain data, and building data for the region. These thresholds are then adjusted and optimized based on the accuracy of the simulation results during actual application.

[0041] When extracting a subset of dynamic simulation data, data processing techniques are required to screen and filter the large amount of data acquired, ensuring that the extracted data accurately falls within the dynamic simulation boundaries and the preset time window. This process may involve processing spatial data and time series data, such as using Geographic Information System (GIS) software to clip spatial data boundaries and using time series analysis software to filter temporal data.

[0042] When matching BIM and GIS data with a subset of dynamic simulation data, it is necessary to establish a data mapping relationship to ensure that building geometry data in the BIM data and topographic elevation data in the GIS data accurately correspond to the spatial and temporal range defined by the dynamic simulation data subset. This may require the use of specialized BIM-GIS data integration software, which can convert and merge the two data formats, ensuring data consistency and accuracy.

[0043] For example, in a specific current scenario, historical simulation data from the past five years is retrieved by querying a database. This includes runoff data under different rainfall intensities and records of changes in building geometry. Simultaneously, historical simulation data from four adjacent scenarios within the same time period is also retrieved. Based on the terrain and building characteristics of the area, a preset scenario division threshold is set to a 500-meter radius centered on the current scenario. The preset time window is set to an upcoming 24-hour rainfall event, from one hour before the rainfall begins to two hours after it ends. Then, using data processing techniques, all relevant data within the 500-meter radius and the 25-hour time window are extracted from the database to form a dynamic simulation data subset. Next, building geometry data from this subset is selected from the BIM model, such as the shape, size, and roof slope of all buildings within the range; topographic elevation data, water system distribution, and soil type data from the GIS system are also selected. These data collectively constitute the dynamic simulation content of rainwater runoff for the current scenario, providing accurate basic data for subsequent simulation analysis.

[0044] In this process, every step must strictly adhere to data processing standards and procedures to ensure data accuracy and reliability. For example, when acquiring historical simulation data, the source and quality of the data need to be checked to ensure that the data has not been interfered with or incorrectly recorded; when setting preset scenario division thresholds and preset time windows, the simulation objectives and actual needs must be fully considered to avoid inaccurate simulation results due to unreasonable threshold and window settings; when extracting and matching data, validated software and algorithms must be used to ensure the accuracy and efficiency of data processing.

[0045] Example 2

[0046] When determining different scenarios for stormwater runoff, for the current scenario, it is necessary to determine whether the rainfall intensity exceeds a preset intensity threshold. The preset intensity threshold is a critical value determined based on long-term rainfall data observed by the local meteorological department, combined with factors such as the region's drainage system design standards and topographic features. For example, in a certain urban area, by analyzing rainfall data from the past 30 years, it is statistically determined that the rainfall intensity that occurs less frequently but could cause severe flooding in this area is 50 mm / h. Considering the design drainage capacity of the area's drainage network, the preset intensity threshold is set at 45 mm / h, serving as a standard to distinguish high-flow scenarios from other scenarios.

[0047] If the rainfall intensity in the current scenario exceeds the preset intensity threshold, the runoff will increase significantly, potentially exceeding the normal drainage capacity. Therefore, the current scenario is directly identified as a high-flow scenario. In this case, the flow speed of rainwater on the surface increases, the confluence time is shortened, and water accumulation is easily formed in low-lying areas, putting significant pressure on the urban drainage system. It is necessary to focus on the rapid discharge of runoff and the prediction of water accumulation areas.

[0048] If the rainfall intensity in the current scenario is not greater than a preset intensity threshold, then the runoff rate and infiltration data for the current scenario need to be classified based on a first preset classification algorithm. The first preset classification algorithm can be a supervised learning algorithm such as Support Vector Machine (SVM) or Random Forest, or an unsupervised learning algorithm such as DBSCAN; the specific choice depends on the data characteristics and simulation requirements. Before applying this algorithm, the runoff rate and infiltration data need to be preprocessed, including data cleaning and standardization, to remove outliers and noisy data, ensuring the accuracy and consistency of the data.

[0049] Taking the random forest algorithm as an example, the first step is to prepare a training dataset. This dataset contains runoff rate and infiltration data for historically similar areas under different rainfall conditions, along with corresponding performance category labels. During training, the algorithm constructs multiple decision trees to extract features and classify the input runoff rate and infiltration data, ultimately obtaining data for each performance category. These performance category data may include different categories such as high infiltration and low runoff, and low infiltration and high runoff, each reflecting specific rainwater runoff and infiltration characteristics.

[0050] After obtaining the data for each performance category, it is necessary to determine the central characteristic of the data for each performance category. The central characteristic can be determined by calculating statistics such as the mean, median, and variance of each category of data, or by using dimensionality reduction methods such as principal component analysis (PCA) to extract the main features. For example, for high-permeability, low-runoff data, the central characteristic might be a higher mean permeability rate, a lower mean runoff rate, and a relatively concentrated data distribution; while for low-permeability, high-runoff data, the central characteristic might be a lower mean permeability rate, a higher mean runoff rate, and greater data dispersion.

[0051] The simulation potential level of the current scenario is determined based on its environmental conditions. These conditions encompass multiple aspects, including topography, land cover type, vegetation cover, and soil type. Regarding topography, areas with steep slopes experience faster rainwater runoff and shorter infiltration times, requiring close monitoring of runoff paths and erosion during simulation. Conversely, plains are prone to waterlogging, necessitating attention to infiltration and drainage capacity. Among land cover types, hardened surfaces (such as concrete and asphalt) have poor infiltration capacity, increasing surface runoff; green spaces and vegetated areas have high infiltration capacity, effectively reducing runoff and increasing the groundwater level. Soil type has a more direct impact on infiltration rate, with sandy soils exhibiting high infiltration rates and clay soils low infiltration rates.

[0052] The simulation potential level can be determined using multi-index comprehensive evaluation methods such as the Analytic Hierarchy Process (AHP). Weights are assigned to each environmental condition indicator, and a comprehensive score is obtained through weighted calculation. The potential level is then divided according to the score range, such as high, medium, and low. For example, in a current scenario with gentle slopes, predominantly green land cover, sandy soil, and high vegetation cover, the simulation potential level can be determined as "high" through comprehensive evaluation calculation. This indicates that the rainwater runoff and infiltration characteristics of this scenario are complex, requiring more refined simulation analysis.

[0053] The system determines whether the central feature of each performance category matches a preset performance category's feature vector more than a preset matching threshold. Preset performance categories are typical performance types defined in advance based on historical experience and actual needs, such as "high-efficiency permeability" and "easy-to-accumulate water." Each preset performance category has its specific feature description and index range. The matching degree can be calculated using methods such as cosine similarity and Euclidean distance to measure the similarity between the central feature and the feature vector of the preset performance category.

[0054] The setting of the preset matching threshold needs to consider the accuracy requirements of the simulation and the needs of actual applications, and is usually determined through multiple experiments and debugging. If the matching degree is greater than the preset matching threshold, it means that the performance characteristics of the current scenario are highly similar to the preset performance category, and the current scenario can be identified as a specific performance scenario; if the matching degree is not greater than the preset matching threshold, it means that the performance characteristics of the current scenario are relatively ordinary, and it is identified as an ordinary performance scenario.

[0055] For example, in a given scenario, the rainfall intensity is 30 mm / h, which is less than the preset intensity threshold of 45 mm / h. After obtaining the runoff rate and infiltration data for this scenario, a random forest algorithm is used for classification, resulting in three performance categories. The central features of each performance category are calculated. One category's central features are a mean infiltration rate of 2.5 mm / min and a mean runoff rate of 1.2 m / s. The matching degree is calculated with the features of the preset performance category "high-efficiency infiltration type" (infiltration rate ≥ 2.0 mm / min, runoff rate ≤ 1.5 m / s). Using cosine similarity, the matching degree is 0.85, which is greater than the preset matching threshold of 0.7. Therefore, this current scenario is determined to be a "high-efficiency infiltration type" scenario within the specific performance scenario.

[0056] Throughout the entire determination process, each step requires rigorous data processing and the application of scientific methods. Acquiring rainfall intensity data relies on high-precision rainfall monitoring equipment to ensure the real-time nature and accuracy of the data; collecting runoff rate and infiltration data requires setting up monitoring points at different locations and using specialized equipment such as flow meters and permeameters for measurement; collecting environmental condition data requires the use of GIS systems and field surveys to ensure the completeness and accuracy of data on topography, land cover, soil, etc.

[0057] In addition, the selection and parameter adjustment of the first preset classification algorithm, the definition and update of the preset performance category, and the selection of the matching degree calculation method all need to be optimized and improved according to the actual situation in order to improve the accuracy and reliability of scene type determination.

[0058] Example 3

[0059] When generating the dynamic simulation workflow set for rainwater runoff, since the dynamic simulation content of rainwater runoff also includes land cover data and meteorological change data, and the different scenario types of rainwater runoff include high infiltration scenario, low infiltration scenario and critical scenario, for the current scenario, it is necessary to first classify the meteorological change data based on the second preset classification algorithm, where the second preset classification algorithm is the K-means clustering algorithm.

[0060] Meteorological change data encompasses various meteorological factors related to rainwater runoff, such as time-series data on rainfall, rainfall intensity, temperature, wind speed, and humidity. When applying the K-means clustering algorithm, these data require preprocessing, including missing value imputation, outlier removal, and data standardization. For example, for rainfall intensity data over a certain period, if there are abnormally high values ​​at individual moments, these need to be identified and removed using statistical methods (such as the 3σ principle) to avoid interfering with the clustering results. Simultaneously, meteorological factors with different dimensions (such as temperature in degrees Celsius and rainfall intensity in millimeters per hour) are normalized to a uniform scale to ensure the accuracy of the clustering algorithm.

[0061] When calculating the distance between samples in meteorological change data, the Euclidean distance formula is typically used. Taking two-dimensional data as an example, the Euclidean distance between sample point A (x1, y1) and sample point B (x2, y2) is: This distance metric effectively reflects the geometrical differences of samples in multidimensional space. For n-dimensional meteorological data, it can be expanded to the square root of the sum of squares of the differences in each dimension. After the distance calculation is completed, the samples are assigned to different categories according to the set initial cluster centers, forming initial categories. The selection of initial cluster centers has a certain impact on the clustering results. Random selection can be used, or the representativeness of the initial centers can be improved and the number of clustering iterations can be reduced through optimization algorithms (such as k-means++).

[0062] The cluster centers are optimized and sample assignments are updated based on a preset number of clusters or the number of clustering iterations. The preset number of clusters needs to be determined based on the characteristics of the meteorological data and the simulation requirements. For example, if the focus is on the impact of rainfall on runoff, the meteorological data can be divided into 3-5 categories such as "heavy rain," "moderate rain," and "light rain." If the coupling effect between temperature and rainfall needs to be considered, the number of clusters can be increased appropriately. When the preset number of iterations (e.g., 50 times) is reached or the change in cluster centers is less than a set threshold (e.g., 0.01), the clustering process terminates, and the data for each meteorological category are obtained. For example, the meteorological data of a certain region can be clustered into three categories: "high temperature and low rainfall," "low temperature and high rainfall," and "mild and humid," with each category corresponding to a specific combination of meteorological characteristics.

[0063] For current meteorological data, a key meteorological subset needs to be selected based on the degree of influence and frequency of change of each meteorological factor. The degree of influence can be determined by calculating the correlation coefficient between each factor and rainwater runoff performance indicators (such as the Pearson correlation coefficient); the higher the correlation, the greater the influence. The frequency of change is measured by counting the number of times each factor exceeds a certain threshold in historical data, such as the frequency of rainfall intensity greater than 20 mm / h. By setting an influence threshold (such as an absolute value of the correlation coefficient ≥ 0.5) and a frequency of change threshold (such as the number of occurrences per year ≥ 10), meteorological factors that significantly affect runoff and change frequently are selected to form the key meteorological subset. For example, rainfall intensity, temperature, and humidity are selected as key meteorological subsets because they have a high correlation with runoff rate and infiltration and change frequently in real-world scenarios.

[0064] The impact labels for the current meteorological category data are determined based on the impact characteristics of key meteorological subsets and the mean values ​​of stormwater runoff performance indicators. Impact characteristics include the changing trends of key meteorological factors (e.g., continuously increasing rainfall intensity) and the occurrence of extreme values ​​(e.g., historical maximum rainfall intensity). The mean values ​​of stormwater runoff performance indicators (e.g., total runoff and permeability coefficient) reflect the typical runoff characteristics under this meteorological category. Impact labels should concisely and accurately describe the nature of the meteorological factors' impact on runoff, such as "heavy rainfall leads to high runoff risk" or "high temperature and drought reduce soil infiltration."

[0065] When determining whether a current scene is a high-permeability scene, it is necessary to combine surface cover data and soil type data for judgment. High-permeability scenes typically have large areas of green space, permeable paving, and other surface covers, and the soil is mainly sandy soil with a high permeability coefficient (e.g., greater than 100%). Low-permeability scenarios are mostly hardened roads and clay soils, with low permeability coefficients (e.g., less than 100%). In critical scenarios, permeability falls between these two extremes, or there may be a mixture of various surface covers. For example, in a current scenario, green space accounts for 60% of the surface cover, the soil type is sandy loam, and the measured permeability coefficient is [missing value]. This can be identified as a high-penetration scenario.

[0066] If the current scenario is a high-permeability scenario, the impact labels of each meteorological category data need to be matched with rainwater runoff performance indicators to generate targeted optimization strategies. During the matching process, the specific impact of the meteorological conditions corresponding to the impact labels on the runoff characteristics of the high-permeability scenario is analyzed. For example, under the label "high runoff risk due to heavy rainfall," although the permeability of the high-permeability scenario is relatively strong, short-term heavy rainfall may exceed the soil permeability limit, leading to an increase in surface runoff. In this case, the optimization strategy can focus on increasing temporary water storage areas and optimizing the porosity of permeable pavement to improve rainwater storage capacity.

[0067] For low-permeability scenarios, improvement and optimization strategies are generated based on preset improvement rules and performance enhancement goals. Preset improvement rules may include engineering measures such as increasing the area of ​​permeable pavement and optimizing the slope of drainage pipe networks. Performance enhancement goals (such as reducing the runoff coefficient by 15%) need to be set in conjunction with regional drainage standards and actual needs. For example, to address the issue of a high proportion of hardened pavement in low-permeability scenarios, the improvement strategy could be set to add permeable brick pavement on both sides of the main road and combine it with the design of bioretention ponds, with the goal of reducing the runoff coefficient during heavy rain from 0.8 to 0.65.

[0068] For critical scenarios, performance stabilization strategies are generated based on historical simulation fluctuations and performance stabilization rules. Historical simulation fluctuations reflect the performance variations of the critical scenario under different meteorological conditions, while performance stabilization rules are based on the principle of avoiding drastic fluctuations in runoff performance, such as maintaining soil moisture within a certain range to stabilize infiltration capacity. For example, in a certain critical scenario, sudden temperature changes often cause soil shrinkage in historical simulations, thus affecting infiltration performance. Stabilization strategies could include planting vegetation with well-developed root systems in the area to enhance soil structural stability and reduce infiltration performance fluctuations caused by meteorological changes.

[0069] Throughout the implementation process, land cover data acquisition requires a combination of remote sensing image interpretation and field surveys to ensure the accuracy of the distribution of green spaces, paved roads, etc. Soil type data can be obtained by drilling and sampling to determine parameters such as permeability coefficients. Meteorological data collection relies on meteorological stations distributed throughout the region to ensure real-time performance and accuracy. The parameters of the K-means clustering algorithm (such as the number of clusters and the number of iterations) need to be adjusted according to the meteorological characteristics of different regions. For example, the clustering methods for meteorological data in humid and arid regions differ and need to be optimized separately. In addition, the generation of optimization strategies needs to be combined with engineering practice experience to ensure their feasibility and effectiveness. For example, the selection of permeable pavement materials in targeted optimization strategies needs to consider local climate conditions and material durability to avoid pavement failure due to freeze-thaw cycles.

[0070] Example 4

[0071] After generating the dynamic simulation workflow set for stormwater runoff, the dynamic simulation content for each scenario needs to be updated based on real-time stormwater runoff monitoring data. Acquiring real-time stormwater runoff monitoring data relies on various sensor devices deployed within the monitoring area, including rain gauges, flow meters, and water level sensors. Rain gauges are used to collect data such as rainfall intensity and amount in real time, typically employing tipping bucket or siphon rain gauges, installed in open, unobstructed locations to ensure measurement accuracy. Flow meters are used to monitor runoff flow in rivers and drainage pipes; depending on the monitoring scenario, ultrasonic flow meters, electromagnetic flow meters, or Doppler flow meters can be selected. For example, in urban drainage networks, electromagnetic flow meters are often used to obtain real-time water velocity and flow rate within the pipes. Water level sensors are used to monitor water level changes in waterlogged areas; common types include pressure-type and radar-type water level sensors, with the appropriate type selected based on the monitoring environment.

[0072] These sensor devices transmit real-time collected data to data centers or servers via data transmission networks (such as 4G, 5G, NB-IoT, etc.) to form a real-time stormwater runoff monitoring database. When updating the dynamic simulation of stormwater runoff, it is necessary to extract real-time data related to each scenario from this database. For example, for a high-flow scenario, it is necessary to extract real-time rainfall intensity data monitored by rain gauges and runoff flow data monitored by flow meters within and around that scenario. After data extraction, the data needs to be preprocessed, including data cleaning and format conversion, to remove noise and outliers and ensure data accuracy and consistency. For example, by setting data thresholds, obviously unreasonable flow data can be identified and eliminated. For instance, when the flow value monitored by the flow meter exceeds twice the maximum design flow of the pipeline, it can be judged as abnormal data and marked or eliminated.

[0073] The scenario type for each scene is recalculated based on the updated dynamic simulation of stormwater runoff. The process of recalculating the scenario type is largely the same as the initial determination, but it needs to be based on updated real-time data. For example, if the current scenario is initially classified as a normal performance scenario, but real-time monitoring data reveals that the rainfall intensity in this scenario has increased sharply in a short period, exceeding a preset intensity threshold, then the scenario type needs to be reassessed based on the updated rainfall intensity data. If it does indeed exceed the threshold, the scenario is reclassified as a high-flow scenario.

[0074] When recalculating the scene type, the determination of rainfall intensity requires real-time acquisition of the latest rainfall data and calculation of rainfall intensity within a time interval. For example, with a 10-minute time interval, it is necessary to calculate whether the rainfall intensity within that time period exceeds a preset intensity threshold. For the classification of runoff rate and infiltration data, the same first preset classification algorithm as the initial classification is used based on the real-time monitored runoff rate and infiltration data to obtain new performance category data. The central features of each performance category data are then redefined to determine the degree of matching with the preset performance category, thereby determining whether the scene type has changed.

[0075] The execution order of the dynamic simulation process set is dynamically adjusted based on the recalculated scenario type. Different scenario types correspond to different simulation focuses and process priorities. Therefore, it is necessary to adjust the execution order in a timely manner according to changes in scenario type to ensure the accuracy and timeliness of simulation results. For example, when a scenario is reclassified from a normal performance scenario to a high-flow scenario, simulation processes for high-flow scenarios that originally had lower priority in the dynamic simulation process set (such as rapid runoff path analysis and drainage system load assessment) need to be adjusted to higher priority and executed first to promptly analyze the potential water accumulation risk and drainage system pressure under high flow conditions.

[0076] The process of dynamically adjusting the execution order needs to consider the logical relationships and data dependencies between the various simulation processes. For example, before conducting a drainage system load assessment, runoff path analysis needs to be completed to obtain accurate runoff flow and direction data. Therefore, there is a sequential dependency between these two processes, and this dependency must not be disrupted when adjusting the execution order. Simultaneously, the execution time and resource consumption of the simulation processes also need to be considered, and computational resources should be allocated rationally to ensure that high-priority processes are executed in a timely manner.

[0077] In practical applications, for example, if a city area is initially classified as a low-permeability scenario in the simulation, the execution order of the dynamic simulation process set mainly revolves around the permeability performance analysis and drainage capacity assessment under the low-permeability scenario. However, during the simulation, real-time monitoring data reveals that the area experienced a short-term heavy rainfall, with the rainfall intensity exceeding the preset intensity threshold, and the runoff flow monitored by the flowmeter also increased significantly. At this point, it is necessary to recalculate the scenario type based on the updated rainfall intensity and runoff flow data, determining that the scenario has transformed into a high-flow scenario. Subsequently, the execution order of the dynamic simulation process set is dynamically adjusted, reducing the priority of the permeability performance analysis process originally designed for the low-permeability scenario, and adjusting processes such as rapid runoff simulation and waterlogging risk warning under the high-flow scenario to high priority, prioritizing the execution of these processes to provide timely decision support for the city's flood control department.

[0078] To achieve efficient processing of real-time monitoring data and rapid scene type determination, an automated data processing and analysis system needs to be established. This system should possess the following functions: real-time data acquisition and transmission to ensure timely and accurate transmission of sensor data to the server; data preprocessing and anomaly detection to clean the acquired data and handle outliers; automatic scene type determination to automatically recalculate the scene type based on updated data; and dynamic execution order adjustment to automatically adjust the execution order of the dynamic simulation process set according to changes in scene type and issue corresponding instructions.

[0079] In addition, a data backup and recovery mechanism needs to be established to prevent the loss of real-time monitoring data due to data transmission failures or server malfunctions, which could affect the updating of scene types and the adjustment of execution order. At the same time, sensor equipment should be calibrated and maintained regularly to ensure the accuracy of real-time monitoring data; for example, rain gauges should be calibrated annually, and flow meters should be inspected and maintained monthly.

[0080] Example 5

[0081] After generating the overall scheme for dynamic simulation of stormwater runoff, a dynamic simulation database for stormwater runoff needs to be established to store data such as the list of coupling parameter settings, the list of data fusion schemes, the dynamic simulation content of stormwater runoff, and data for various scenario types. The database architecture design needs to consider the diversity of data types, covering structured data (such as numerical values ​​of coupling parameters and classification labels for scenario types) and unstructured data (such as BIM model files and GIS raster data). For example, the list of coupling parameter settings can be stored using a relational database table structure, containing fields such as scenario ID, parameter name, parameter value, and unit; BIM model files can be stored in the database in the form of binary large objects (BLOBs), or through a storage path associated with the database via a file system. The database management system can be open-source software such as MySQL or PostgreSQL, or commercial software such as Oracle, depending on the data volume, concurrent access requirements, and budget. A backup strategy, such as daily full backups and hourly incremental backups, should also be configured to ensure data security.

[0082] Regularly clean and maintain the database to remove invalid and duplicate data. The data cleanup frequency can be set monthly or quarterly, depending on the data update frequency. Invalid data identification rules include: historical simulation data exceeding a preset time range (such as atypical scenario data from three years ago), monitoring data with collection errors exceeding a threshold (such as abnormal rainfall data caused by rain gauge malfunctions), and imported data with incorrect formats (such as corrupted BIM model files). Duplicate data detection can be achieved by comparing data content using unique identifiers (such as scenario IDs or simulation task numbers). For example, if multiple identical reports are generated from the same simulation task, the latest version can be retained and the older version deleted. During the cleanup process, the data to be cleaned should be exported and backed up beforehand to avoid accidentally deleting important information. After cleanup, the database index should be updated to improve query efficiency.

[0083] Based on this database, long-term tracking and comparative analysis of stormwater runoff simulation results are conducted to assess the changing trends of stormwater runoff performance over time. Analysis dimensions include: changes in runoff coefficients for the same scenario in different years; for example, comparing the runoff coefficients of a commercial area under the same rainfall conditions in 2022 and 2024. If the coefficient increases from 0.75 to 0.82, it may reflect an increase in hardened area or blockage in the drainage system. Seasonal performance fluctuations, such as the difference in infiltration rates between the rainy summer and dry winter, can be analyzed by statistically analyzing the average infiltration data for each season. Comparison of the effects before and after the implementation of engineering measures, such as comparing the frequency and depth of water accumulation three years after a sponge city renovation project in a certain area. Analysis methods can include time series analysis, plotting trend lines, or using statistical tests (such as t-tests) to determine whether the changes in performance indicators are significant.

[0084] The preset fusion rules include confidence intervals based on historical stormwater runoff fusion data, dynamically adjusting the sample size and number of repetitions for data fusion. Historical fusion data refers to the sample set from past data fusions. For example, a fusion might have used 50 sets of measured data compared to simulated data, resulting in a fusion error confidence interval of [-0.12, 0.15]. When a new fusion task is initiated, the system automatically retrieves historical fusion data similar to the current scenario. If the historical confidence interval width exceeds a preset threshold (e.g., 0.2), the current fusion sample size is increased (e.g., from 50 to 80 sets), and the number of fusion repetitions is increased (e.g., from 3 to 5 times) to reduce fusion error. The adjustment of the sample size must follow statistical principles, such as determining the minimum sample size based on the central limit theorem, to ensure that the sample mean approximately follows a normal distribution.

[0085] When the deviation between the measured runoff data and the coupled simulation results exceeds a preset error threshold, the supplementary fusion module is automatically triggered. The preset error threshold is set according to the simulation accuracy requirements. For example, for runoff flow simulation, the threshold can be set to 15%, meaning that supplementary fusion is triggered when the difference between the measured flow and the simulated flow exceeds 15% of the simulated value. The workflow of the supplementary fusion module is as follows: First, locate the data points with large deviations and analyze their causes (such as sensor failure or incorrect simulation parameter settings). If it is a sensor failure, the data point needs to be marked and re-acquired. If it is an incorrect parameter setting, adjust the relevant coupling parameters (such as soil permeability coefficient and Manning roughness coefficient). Then, recalculate using the original data and the newly added data until the deviation is reduced to within the threshold. For example, in a certain scenario, the measured runoff flow rate was 120 m³ / s, while the simulation result was 95 m³ / s, with a deviation of 26.3%, exceeding the 15% threshold. After triggering the supplementary fusion module, it was found that the soil permeability coefficient was too low. After adjusting it from 0.05 m / s to 0.08 m / s and re-fusioning, the simulation result improved to 112 m³ / s, and the deviation decreased to 7.1%, meeting the requirements.

[0086] The pre-defined fusion rules also include establishing a data acquisition equipment calibration compensation mechanism. Before each fusion, the data acquisition equipment is calibrated for error compensation based on the benchmark values ​​of standard rain gauges and flow meters. Standard equipment needs to be sent to a metrology institution periodically to obtain calibration certificates. For example, standard rain gauges are calibrated annually by the provincial metrology institute to determine their error correction coefficients under different rainfall intensities. The implementation steps of the calibration compensation mechanism are as follows: Before fusion, the equipment to be calibrated and the standard equipment are placed in the same environment for comparative measurement. For example, the field rain gauge and the standard rain gauge are installed side-by-side, and the rainfall within one hour is recorded synchronously. The difference between the field equipment and the standard equipment is calculated to generate a calibration coefficient (e.g., if the field equipment measurement is 52 mm and the standard equipment is 50 mm, the calibration coefficient is 50 / 52≈0.96). Then, all data collected by the field equipment is multiplied by this calibration coefficient for compensation. For flow meter calibration, the standard flow source method can be used. By conveying water of known flow rate through the flow meter, the difference between the measured value and the standard value is recorded, generating a linear calibration equation (e.g., ...). (where x is the measured value and y is the calibrated value).

[0087] For example, before a certain integration process, an ultrasonic flow meter in a certain area is calibrated. The standard flow source outputs 100 m³ / h of water, and the flow meter measures 97.5 m³ / h. The calculated calibration coefficient is 100 / 97.5 ≈ 1.025. All subsequent data collected by this flow meter are then multiplied by 1.025 for calibration. The calibration process must record information such as calibration time, calibration equipment number, and calibration coefficient, and store this information in a database for traceability. If the equipment error still exceeds the allowable range after calibration (e.g., rain gauge error exceeds ±4%, flow meter error exceeds ±2%), the equipment must be replaced or returned to the factory for repair.

[0088] Through the construction and maintenance of the database, long-term tracking and analysis, and the implementation of data fusion rules, systematic management of dynamic simulation data of rainwater runoff can be achieved, ensuring the accuracy and availability of the data and providing data support for urban drainage planning and sponge city construction. For example, by analyzing five years of simulation data in the database, it was found that the runoff coefficient of an industrial park increased year by year. Combined with field investigation, it was determined that this was caused by the reduction of green area within the park. Based on this, it can be recommended to increase rooftop greening or permeable paving to improve rainwater runoff performance.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0090] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dynamic simulation method for stormwater runoff based on BIM-GIS coupling, characterized in that, The method includes: The dynamic simulation process of rainwater runoff is planned based on input parameters and environmental conditions. This involves determining the required BIM and GIS data content for the simulation, including building geometry data and topographic elevation data. Different scenario types for rainwater runoff are determined based on the input parameters and environmental conditions. A dynamic simulation process set for rainwater runoff is generated based on these scenario types, the BIM data content, and the GIS data content, and the execution order of the dynamic simulation process set is determined based on the optimal coupling principle. A coupling parameter setting list for each scenario is generated according to a preset simulation period, the different scenario types of rainwater runoff, and the performance index requirements of rainwater runoff. A data fusion scheme list for each scenario is also generated according to preset fusion rules, the different scenario types of rainwater runoff, and the performance index requirements of rainwater runoff. Finally, based on the execution order of the dynamic simulation process set, the coupling parameter setting list, the data fusion scheme list, and the dynamic simulation content of each scenario are linked together to generate the overall scheme for the dynamic simulation of rainwater runoff. The dynamic simulation process of stormwater runoff is planned based on input parameters and environmental conditions, and the BIM and GIS data content required for the simulation is determined. This includes: for the current scenario, determining the dynamic simulation boundary of the current scenario based on historical simulation data of the current scenario, historical simulation data of adjacent scenarios, and a preset scenario division threshold; extracting a subset of dynamic simulation data of the current scenario based on the dynamic simulation boundary and a preset time window; and determining the content of BIM and GIS data within the subset of dynamic simulation data as the dynamic simulation content of stormwater runoff for the current scenario.

2. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 1, characterized in that, Based on the input parameters and environmental conditions of the stormwater runoff, different scenario types of stormwater runoff are determined, including: for the current scenario, if the rainfall intensity of the current scenario is greater than a preset intensity threshold, the current scenario is determined to be a high-flow scenario; if the rainfall intensity of the current scenario is not greater than the preset intensity threshold, the runoff rate and infiltration data of the current scenario are classified according to a first preset classification algorithm to obtain data for each performance category, and the central feature of each performance category data is determined; the simulation potential level of the current scenario is determined according to the environmental conditions of the current scenario; if the matching degree between the central feature of each performance category data and the preset performance category is greater than a preset matching threshold, the current scenario is determined to be a specific performance scenario; if the matching degree is not greater than the preset matching threshold, the current scenario is determined to be a normal performance scenario.

3. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 2, characterized in that, The dynamic simulation of rainwater runoff also includes land cover data and meteorological change data. The different scenario types of rainwater runoff include high-permeability scenarios, low-permeability scenarios, and critical scenarios. Based on the different scenario types of rainwater runoff, BIM data content, and GIS data content, a dynamic simulation process set for rainwater runoff is generated, including: for the current scenario, classifying the meteorological change data of the current scenario based on a second preset classification algorithm to obtain data for each meteorological category; for the current meteorological category data, filtering the current meteorological category data based on the influence degree and change frequency of each meteorological factor in the current meteorological category data to obtain a key meteorological subset of the current meteorological category data; determining the influence label of the current meteorological category data based on the influence characteristics of the key meteorological subset and the mean of the rainwater runoff performance index; and determining whether the scenario type of the current scenario is a high-permeability scenario.

4. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 3, characterized in that, After determining whether the current scenario is a high-permeability scenario, the method further includes: if the current scenario is a high-permeability scenario, matching the impact labels of each meteorological category data with rainwater runoff performance indicators to generate targeted optimization strategies; if the current scenario is a low-permeability scenario, generating improvement optimization strategies based on preset improvement rules and performance improvement targets; and if the current scenario is a critical scenario, generating performance stabilization strategies based on historical simulation fluctuations and performance stabilization rules.

5. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 4, characterized in that, After generating the dynamic simulation process set for rainwater runoff, the method further includes: updating the dynamic simulation content of rainwater runoff for each scenario based on real-time monitoring data of rainwater runoff; recalculating the scenario type of each scenario based on the updated dynamic simulation content of rainwater runoff; and dynamically adjusting the execution order of the dynamic simulation process set based on the recalculated scenario type.

6. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 1, characterized in that, The preset fusion rules include: dynamically adjusting the sample size and number of repetitions of data fusion based on the confidence interval of historical fusion data of rainwater runoff; and automatically triggering the supplementary fusion module when the deviation between the measured runoff data and the coupled simulation results exceeds the preset error threshold.

7. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 6, characterized in that, The preset fusion rules also include: establishing a data acquisition equipment calibration compensation mechanism, and performing error compensation calibration on the data acquisition equipment based on the reference values ​​of standard rain gauges and flow meters before each fusion.

8. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 1, characterized in that, After generating the overall scheme for the dynamic simulation of stormwater runoff, the method further includes: establishing a dynamic simulation database for stormwater runoff, storing the coupling parameter setting list, data fusion scheme list, dynamic simulation content of stormwater runoff, and various scenario types in the database; regularly cleaning and maintaining the data in the database to remove invalid and duplicate data; and conducting long-term tracking and comparative analysis of the dynamic simulation results of stormwater runoff based on the database to evaluate the changing trend of stormwater runoff performance over time.

9. The method for dynamic simulation of stormwater runoff based on BIM-GIS coupling according to claim 3, characterized in that, The second preset classification algorithm is the K-means clustering algorithm. The specific steps for classifying the meteorological change data of the current scene include: calculating the distance between each sample in the meteorological change data; assigning the samples to different cluster centers according to the distance to form an initial category; optimizing the cluster centers and updating the sample assignment according to the preset number of clusters or the number of cluster iterations to obtain the meteorological category data.

Images