River basin cascade water resource decision key information query system for mobile terminal real-time query

Abstract

The invention discloses a drainage basin cascade water resource decision key information query system for mobile terminal real-time query. The system architecture comprises a mobile terminal application layer, a data processing layer and a background management subsystem; the mobile terminal application layer is based on an entrance of a company mobile office APP and comprises a cascade power station operation monitoring module, a power generation plan and weather forecast module, a water regimen and rain regimen thermodynamic diagram module and an AR enhancement visualization module. The data processing layer comprises a real-time data acquisition engine, a historical database and a prediction model library; the background management subsystem comprises a user authority analysis management tool, a data abnormity alarm configuration tool and an AR scene editing tool; the watershed cascade water resource management information can be inquired by using the mobile terminal equipment, corresponding data analysis and prediction information is obtained, information support is provided for decision making, and the problems that data query in the prior art is scattered in data, tedious in operation, mutually isolated in databases and poor in data query efficiency are solved. And data analysis and prediction cannot be provided.

Description

Technical Field

[0001] The present invention relates to the field of virtual reality technology, and in particular to a key information query system for basin cascade water resources decision-making for real-time query on a mobile terminal. Background Art

[0002] With the improvement of the digital capabilities of hydropower stations, 3D technology, as a carrier platform for data analysis and display applications in power production management, is the foundation of smart enterprises and smart power station applications. Currently, the demand for 3D applications is constantly emerging in various business professional aspects, such as operation and maintenance simulation training, virtual power station operation, watershed maintenance support, dam area geological management, safe production, technical training, etc. In the construction of technical systems in different professional fields, there is a demand for the construction of 3D technology functions and the use of 3D models.

[0003] At present, the company has many platforms, systems, and functional modules in the field of watershed scheduling information display. The query method is mainly active search query. When conducting information query based on three-dimensional models, cumbersome operation steps are generally required, and the intuitiveness and real-time performance are poor. Users can only obtain information in a one-way manner or search for key information by inputting queries. The interactivity is poor and the potential of the constructed three-dimensional model library cannot be fully utilized. In addition, the data is scattered in multiple isolated systems and lacks a unified mobile terminal entrance. The ability to compare and analyze real-time data with history is insufficient, and the coordinated display of weather forecasts, power generation plans and reservoir scheduling is missing. In specific operations, the traditional PC system cannot meet the needs of outdoor inspections and emergency command scenarios, and cannot be queried anytime and anywhere.

[0004] Therefore, it is necessary to design a key information query system for basin cascade water resources decision-making for real-time query on mobile terminals, based on virtual reality technology, to realize convenient hydropower station scene recognition and information interaction on mobile terminals. Summary of the Invention

[0005] The purpose of the present invention is to provide a key information query system for river basin cascade water resources decision-making for real-time query on mobile terminals. Mobile terminal devices can be used to query river basin cascade water resources management information and obtain corresponding data analysis and prediction information, thereby providing information support for decision-making. It aims to solve the problems of data query in the existing technology, such as data dispersion, cumbersome operations, isolation between databases, and inability to provide data analysis and prediction.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: The key information query system for river basin cascade water resources decision-making, used for real-time mobile queries, includes a three-level system architecture consisting of a mobile application layer, a data processing layer, and a background management subsystem; The mobile application layer is based on the company's mobile office app entrance, including cascade power station operation monitoring module, power generation plan and weather forecast module, water and rainfall heat map module, and AR enhanced visualization module; The data processing layer includes a real-time data acquisition engine, a historical database, and a prediction model library; The backend management subsystem includes user authority analysis and management, data anomaly alarm configuration, and AR scene editing tools.

[0007] Preferably, the cascade power station operation monitoring module is connected to the historical database of the data processing layer to perform historical comparison and trend analysis functions, including: Extract historical data of power plant electricity and storage capacity from the historical database; Generate visual charts such as year-on-year bar charts and trend lines on mobile devices; Supports users to select time range for dynamic comparative analysis.

[0008] Preferably, the AR enhanced visualization module performs AR enhanced decision support functions, including: Identify the physical or geographic coordinates of the power station through the mobile device camera; Obtain real-time water level and unit status data from the data processing layer; Real-time data, historical extreme value marks and flood control point information are superimposed and displayed in the AR scene.

[0009] Preferably, the cascade power station operation monitoring module performs a real-time aggregation function of cascade power station operation information, including: Obtain power plant operation data, including water level, flow rate, and unit status, from the real-time data acquisition engine in the data processing layer; Display real-time information on mobile devices in the form of dynamic charts, color coding, etc. Supports users to click through to view detailed data of a single power station or a single unit.

[0010] Preferably, the power generation plan and weather forecast module performs a multi-time-scale power generation plan and weather fusion display function, including: Obtain power generation plan and flow forecast data from the prediction model library of the data processing layer; Get short-term weather forecast and rainfall forecast data from the Meteorological Bureau API; Multi-timescale data is displayed on mobile devices in the form of dual Y-axis charts or heat maps, including next-day power generation plans, 5-day flow forecasts, and 36-hour rainfall heat maps.

[0011] Preferably, the data processing layer provides real-time data, historical data and forecast data support for the mobile terminal. The background management subsystem is used for configuration management, permission control and AR scene editing, indirectly supporting the implementation of core functions. The mobile application layer is the main carrier of the four core functional modules, responsible for data visualization and user interaction.

[0012] Preferably, the real-time data acquisition engine is connected to the SCADA, meteorological bureau and hydrological station API, and the historical database stores more than 5 years of power station operation data; the prediction model library has built-in trained flow forecast and power generation prediction models for flow and power generation prediction.

[0013] Preferably, the real-time data collection engine executes a dynamic data aggregation algorithm, specifically: (1) Set input and output: Input: Raw high-frequency data streams from SCADA systems, hydrological stations, and meteorological APIs; Output: Compressed data packets aggregated by time window for efficient transmission and display on mobile devices; (2) Define the time window and sliding mechanism: Time window: set a fixed time window length T; Sliding step size: window sliding interval S; Data point interval: raw data sampling interval ; (3) Data aggregation within the window: For each original data in time window T Aggregate according to the following rules: ; Where, n Indicates the number of original data points in the window, n=T / =60; Indicates the average value of the data in the window, which is used to represent the overall trend; Indicates the extreme value within the window, used to retain key fluctuation characteristics; (4) Sliding window update: After each sliding step S, the data of the previous S period of the old window is discarded, the data of the new S period is added, and the aggregate value is recalculated: Initial window: ; Sliding rear window: ; Data compression and transmission: The aggregated data packets Encoded in JSON format, it contains only the following fields: ; Where, are the window start and end timestamps, Indicates the data type identifier.

[0014] Preferably, the real-time data acquisition engine executes a multi-source heterogeneous data fusion algorithm, specifically the following method: (1) Set input and output: Input: Multi-source heterogeneous data, including weather forecasts (JSON), water regime data (Modbus), and power generation plans (XML); Output: unified standard time series data format, including timestamp, data type, value and unit; (2) Data extraction and analysis: extracting raw data from different data sources according to protocols, including: Weather data JSON: parsed into key-value pairs; Water condition data Modbus: read register values and convert them into physical quantities; Power generation plan XML: parsing tag content; The formula is as follows: For any data source S, the analytical function P( S ) to convert it into intermediate data: ; Where, Indicates the data timestamp, Indicates that the original values include water level, flow rate, and power generation. represents the original unit; (3) Data conversion and unit standardization, converting data in different units into system standard units: ; Where, express Unit conversion factors; (4) Unified data model mapping: Defining a standard time series data model , contains the following fields: ； Where, t std indicates the aligned standard timestamp, which is aligned to the whole minute or hour; type indicates the data type, including water level, flow, and power generation; source_id indicates the data source identifier; (5) Time alignment and interpolation: For non-uniformly sampled data, linear interpolation is used to generate uniform timestamp data: ; (6) Data association and storage: associate standardized data by timestamp and data type and store them in a historical database: .

[0015] Preferably, after the data processing layer performs dynamic data aggregation and multi-source heterogeneous data fusion, the water and rainfall heat map module executes the heat map generation optimization algorithm, and the specific method is as follows: (1) Set input and output: Input: Basin rainfall data in 6-hour units over the past 36 hours, including time, location, and rainfall amount; Output: interactive heatmap with a frame rate of ≥30fps, supporting zooming and timeline dragging; (2) Spatiotemporal data aggregation: Time block: Divide the 36-hour data into 6-hour blocks, each containing data from multiple observation points: ; Where, ( xi , yi ) represents the latitude and longitude coordinates of the observation point, normalized to the interval [0,1]; ri Indicates the accumulated rainfall within 6 hours; k Indicates the time block index, a total of 6 blocks; Spatial grid division: Divide the watershed map into N × N , calculate each grid cell ( m , n ) density value dm , n : ; Where, Sm , n Indicates that it falls into the grid ( m , n ) observation point set, μx , μy represents the grid center coordinates, σ Indicates the Gaussian kernel radius, controls the smoothness, default σ =0.05; (3) Color mapping and texture generation: Color coding, density values dm , n Mapping to RGBA color space uses a piecewise linear function: ; Where, express Density range dynamic threshold; α Indicates transparency, which is positively correlated with density; Texture Compression: Will N × N The color matrix is encoded as a WebGL texture, using the ASTC compression format to reduce video memory usage; (4) WebGL rendering optimization: Shader Programming: Executing Vertex Shaders and Fragment Shaders in the GPU: The vertex shader converts the mesh coordinates into screen positions, and the fragment shader mixes colors based on the texture sampling results and overlays the timeline markers; Then use gl.drawArrays() to submit all mesh vertex data at once, reducing CPU-GPU communication overhead The beneficial effects of the key information query system for basin cascade water resources decision-making for real-time mobile query provided by the present invention are: 1. This invention achieves deep integration and efficient collaboration of multi-dimensional water resource decision-making data, significantly improving decision-making accuracy and response speed. Traditional river basin management systems struggle to support complex hydropower coordinated scheduling and flood control decisions due to issues such as data dispersion, heterogeneous formats, and delayed updates. This invention addresses this core pain point through the following innovations: Multi-source heterogeneous data fusion technology: This technology converts multiple types of heterogeneous data, such as meteorological, hydrological, and power generation planning, into a standard time series format, eliminating data silos and improving the efficiency of cross-system data correlation analysis. Dynamic data aggregation algorithm: A sliding time window is used to compress and transmit the mean and extreme values of high-frequency raw data, which reduces mobile data consumption while retaining key trend features.

[0016] 2. This invention innovates mobile AR enhancement and high-performance visualization technology, breaking through the bottleneck of information interaction in complex outdoor environments. To address the problem that traditional PC-based systems cannot meet the needs of outdoor scenarios such as power station inspections and emergency command, this invention achieves efficient human-computer interaction on mobile devices through the following technologies: AR augmented reality overlay technology: Through geographic coordinate recognition and camera real-time scene capture, real-time water levels, historical extreme value marks and flood control points are superimposed on mobile devices; WebGL accelerated heat map rendering: GPU shader programming is used to improve the rendering frame rate of basin rainfall data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 Schematic diagram of the system structure of the present invention; Figure 2 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0018] Example 1: like Figure 1 As shown in the figure, the key information query system for river basin cascade water resources decision-making used for real-time query on mobile terminals includes a three-level system architecture consisting of a mobile terminal application layer, a data processing layer, and a background management subsystem; The mobile application layer is based on the company's mobile office app entrance, including cascade power station operation monitoring module, power generation plan and weather forecast module, water and rainfall heat map module, and AR enhanced visualization module; The data processing layer includes a real-time data acquisition engine, a historical database, and a prediction model library; The backend management subsystem includes user authority analysis and management, data anomaly alarm configuration, and AR scene editing tools.

[0019] Preferably, the cascade power station operation monitoring module is connected to the historical database of the data processing layer to perform historical comparison and trend analysis functions, including: Extract historical data of power plant electricity and storage capacity from the historical database; Generate visual charts such as year-on-year bar charts and trend lines on mobile devices; Supports users to select time range for dynamic comparative analysis.

[0020] Preferably, the AR enhanced visualization module performs AR enhanced decision support functions, including: Identify the physical or geographic coordinates of the power station through the mobile device camera; Obtain real-time water level and unit status data from the data processing layer; Real-time data, historical extreme value marks and flood control point information are superimposed and displayed in the AR scene.

[0021] Preferably, the cascade power station operation monitoring module performs a real-time aggregation function of cascade power station operation information, including: Obtain power plant operation data, including water level, flow rate, and unit status, from the real-time data acquisition engine in the data processing layer; Display real-time information on mobile devices in the form of dynamic charts, color coding, etc. Supports users to click through to view detailed data of a single power station or a single unit.

[0022] Preferably, the power generation plan and weather forecast module performs a multi-time-scale power generation plan and weather fusion display function, including: Obtain power generation plan and flow forecast data from the prediction model library of the data processing layer; Get short-term weather forecast and rainfall forecast data from the Meteorological Bureau API; Multi-timescale data is displayed on mobile devices in the form of dual Y-axis charts or heat maps, including next-day power generation plans, 5-day flow forecasts, and 36-hour rainfall heat maps.

[0023] Preferably, the data processing layer provides real-time data, historical data and forecast data support for the mobile terminal. The background management subsystem is used for configuration management, permission control and AR scene editing, indirectly supporting the implementation of core functions. The mobile application layer is the main carrier of the four core functional modules, responsible for data visualization and user interaction.

[0024] Preferably, the real-time data acquisition engine is connected to the SCADA, meteorological bureau and hydrological station API, and the historical database stores more than 5 years of power station operation data; the prediction model library has built-in trained flow forecast and power generation prediction models for flow and power generation prediction.

[0025] like Figure 2 As shown, preferably, the real-time data collection engine executes a dynamic data aggregation algorithm, specifically: (1) Set input and output: Input: Raw high-frequency data streams from SCADA systems, hydrological stations, and meteorological APIs; Output: Compressed data packets aggregated by time window for efficient transmission and display on mobile devices; (2) Define the time window and sliding mechanism: Time window: set a fixed time window length T; Sliding step size: window sliding interval S; Data point interval: raw data sampling interval ; (3) Data aggregation within the window: For each original data in time window T Aggregate according to the following rules: ; Where, n Indicates the number of original data points in the window, n=T / =60; Indicates the average value of the data in the window, which is used to represent the overall trend; Indicates the extreme value within the window, used to retain key fluctuation characteristics; (4) Sliding window update: After each sliding step S, the data of the previous S period of the old window is discarded, the data of the new S period is added, and the aggregate value is recalculated: Initial window: ; Sliding rear window: ; Data compression and transmission: Encode the aggregated data packets into JSON format, which only contains the following fields: ; wherein, are the start and end timestamps of the window, represents the data type identifier.

[0026] Example 2: Taking the inflow of the Three Gorges Power Station as an example, the process of aggregating flow data is as follows: The original data is the sampling value per minute (unit: m 3 / s), then: ; The size of the compressed data packet received by the mobile device is reduced from the original 60 pieces of data (about 3KB) to 1 piece of aggregated data (about 0.2KB), and the traffic is reduced by 93%.

[0027] Example 3: The real-time data acquisition engine executes a multi-source heterogeneous data fusion algorithm. The specific method is as follows: (1) Set the input and output: Input: Multi-source heterogeneous data, including weather forecast - JSON, water regime data - Modbus, and power generation plan - XML; Output: Unified standard time series data format, including timestamp, data type, value, and unit; (2) Data extraction and parsing. Extract the original data from different data sources according to the protocol, including: Weather data JSON: Parse it into key-value pairs; such as {"time":"2023-10-01T08:00","rainfall":12.5,"unit":"mm"}; Water regime data Modbus: Read the register value and convert it into a physical quantity; such as the register address 0x1001 corresponding to the water level value H H (unit: meter); Power generation plan XML: Parse the label content; such as: <stationname="Three Gorges"> <power> 15000< / power> <unit> MW< / unit> ; The formula is expressed as follows: For any data source S, the parsing function P( S ) converts it into intermediate data: ; wherein, represents the data timestamp, Indicates that the original values include water level, flow rate, and power generation Indicates the original unit; (3) Data conversion and unit standardization, converting data in different units to the system standard unit, such as unifying the flow rate to m³ / s and the power generation to MW): ; In the formula, express Unit conversion coefficient; (for example, 1 foot = 0.3048 meters, then = 0.3048); (4) Unified data model mapping: Define the standard time series data model , including the following fields: ； In the formula, t std represents the aligned standard timestamp, aligned by the whole minute or the whole hour; type represents the data type including water level, flow rate, and power generation; source_id represents the data source identifier (such as "Three Gorges Power Station", "Jinsha River Meteorological Station"); (5) Time alignment and interpolation, for non-uniformly sampled data, linearly interpolate to generate unified timestamp data: ; (6) Data association and storage, associate the standardized data according to the timestamp and data type, and store it in the historical database: .

[0028] Formula example: Integration of meteorology and power generation plan: Meteorological data (JSON): Original data: {"time":"2023-10-01T08:00","rainfall":10.2,"unit":"mm"}; After conversion: .

[0029] Power generation plan (XML): Original data: <stationname="Three Gorges"> <power> 15000< / power> <unit> MW< / unit> ; After conversion: .

[0030] Integration result: At the timestamp 2023-10-01T08:00, associate the rainfall and power generation plan data to analyze the impact of weather on power generation.

[0031] Example 4: Preferably, after the data processing layer performs dynamic data aggregation and multi-source heterogeneous data fusion, the water and rainfall heat map module executes the heat map generation optimization algorithm, and the specific method is as follows: (1) Set input and output: Input: Basin rainfall data in 6-hour units over the past 36 hours, including time, location, and rainfall amount; Output: interactive heatmap with a frame rate of ≥30fps, supporting zooming and timeline dragging; (2) Spatiotemporal data aggregation: Time block: Divide the 36-hour data into 6-hour blocks, each containing data from multiple observation points: ; Where, ( xi , yi ) represents the latitude and longitude coordinates of the observation point, normalized to the interval [0,1]; ri Indicates the accumulated rainfall within 6 hours; k Indicates the time block index, a total of 6 blocks; Spatial grid division: Divide the watershed map into N × N , calculate each grid cell ( m , n ) density value dm , n : ; Where, Sm , n Indicates that it falls into the grid ( m , n ) observation point set, μx , μy represents the grid center coordinates, σ Indicates the Gaussian kernel radius, controls the smoothness, default σ =0.05; (3) Color mapping and texture generation: Color coding, density values dm , n Mapping to RGBA color space uses a piecewise linear function: ; Where, express Density range dynamic threshold; α Indicates transparency, which is positively correlated with density; Texture Compression: Will N ×N The color matrix is encoded as a WebGL texture, using the ASTC compression format to reduce video memory usage; (4) WebGL rendering optimization: Shader Programming: Executing Vertex Shaders and Fragment Shaders in the GPU: The vertex shader converts the mesh coordinates into screen positions, and the fragment shader mixes colors based on the texture sampling results and overlays the timeline markers; Then use gl.drawArrays() to submit all mesh vertex data at once, reducing CPU-GPU communication overhead Formula example: density calculation and rendering: Assume that a 6-hour time block contains 3 observation points: Observation point 1: ( x 1=0.2, y 1=0.5, r 1 = 15mm); Observation point 2: ( x 2=0.3, y 2=0.6, r 2 = 20 mm); Observation point 3: ( x 3=0.25, y 3=0.55, r 3 = 10mm); For the grid ( m =5, n =5) with the center coordinates x=0.25, y=0.55, μy =0.55Calculate the density: ; The final color is red ( Exceeding the threshold d max=30).

Claims

1. A key information query system for river basin cascade water resources decision-making for real-time mobile query, featuring: A three-level system architecture consisting of mobile application layer, data processing layer and background management subsystem; The mobile application layer is based on the company's mobile office app entrance, including cascade power station operation monitoring module, power generation plan and weather forecast module, water and rainfall heat map module, and AR enhanced visualization module; The data processing layer includes a real-time data acquisition engine, a historical database, and a prediction model library; The backend management subsystem includes user authority analysis and management, data anomaly alarm configuration, and AR scene editing tools.

2. The basin-level water resources decision-making key information query system for real-time mobile query according to claim 1 is characterized by: The cascade power station operation monitoring module communicates with the historical database of the data processing layer to perform historical comparison and trend analysis functions, including: Extract historical data of power plant electricity and storage capacity from the historical database; Generate visual charts such as year-on-year bar charts and trend lines on mobile devices; Supports users to select time range for dynamic comparative analysis.

3. The key information query system for river basin cascade water resources decision-making for real-time mobile query according to claim 1 is characterized by: The AR enhanced visualization module performs AR enhanced decision support functions, including: Identify the physical or geographic coordinates of the power station through the mobile device camera; Obtain real-time water level and unit status data from the data processing layer; Real-time data, historical extreme value marks and flood control point information are superimposed and displayed in the AR scene.

4. The key information query system for river basin cascade water resources decision-making for real-time mobile terminal query according to claim 1 is characterized by: The cascade power station operation monitoring module performs real-time aggregation of cascade power station operation information, including: Obtain power plant operation data, including water level, flow rate, and unit status, from the real-time data acquisition engine in the data processing layer; Display real-time information on mobile devices in the form of dynamic charts, color coding, etc. Supports users to click through to view detailed data of a single power station or a single unit.

5. The basin-level water resources decision-making key information query system for real-time mobile terminal query according to claim 1 is characterized by: The power generation plan and weather forecast module performs multi-timescale power generation plan and weather fusion display functions, including: Obtain power generation plan and flow forecast data from the prediction model library of the data processing layer; Get short-term weather forecast and rainfall forecast data from the Meteorological Bureau API; Multi-timescale data is displayed on mobile devices in the form of dual Y-axis charts or heat maps, including next-day power generation plans, 5-day flow forecasts, and 36-hour rainfall heat maps.

6. The basin cascade water resources decision-making key information query system for real-time mobile terminal query according to claim 1 is characterized in that: The data processing layer provides real-time data, historical data, and forecast data support for mobile terminals. The background management subsystem is used for configuration management, permission control, and AR scene editing, indirectly supporting the implementation of core functions. The mobile application layer is the main carrier of the four core functional modules, responsible for data visualization and user interaction.

7. The key information query system for river basin cascade water resources decision-making for real-time mobile terminal query according to claim 1 is characterized in that: The real-time data acquisition engine connects to SCADA, meteorological bureau, and hydrological station APIs, and the historical database stores more than five years of power station operation data; the prediction model library has built-in trained flow forecast and power generation prediction models for flow and power generation prediction.

8. A key information query system for river basin cascade water resources decision-making for real-time mobile query, characterized by: The real-time data collection engine executes a dynamic data aggregation algorithm, specifically: (1) Set input and output: Input: Raw high-frequency data streams from SCADA systems, hydrological stations, and meteorological APIs; Output: Compressed data packets aggregated by time window for efficient transmission and display on mobile devices; (2) Define the time window and sliding mechanism: Time window: set a fixed time window length T; Sliding step size: window sliding interval S; Data point interval: raw data sampling interval ; (3) Data aggregation within the window: For each original data in time window T Aggregate according to the following rules: ； Where, n Indicates the number of original data points in the window, n=T / =60; Indicates the average value of the data in the window, which is used to represent the overall trend; Indicates the extreme value within the window, used to retain key fluctuation characteristics; (4) Sliding window update: After each sliding step S, the data of the previous S period of the old window is discarded, the data of the new S period is added, and the aggregate value is recalculated: Initial window: ; Sliding rear window: ; Data compression and transmission: The aggregated data packets Encoded in JSON format, it contains only the following fields: ； Where, are the window start and end timestamps, Indicates the data type identifier.

9. A key information query system for river basin cascade water resources decision-making for real-time mobile query, characterized by: The real-time data acquisition engine executes the multi-source heterogeneous data fusion algorithm. The specific method is as follows: (1) Set input and output: Input: Multi-source heterogeneous data, including weather forecasts (JSON), water regime data (Modbus), and power generation plans (XML); Output: unified standard time series data format, including timestamp, data type, value and unit; (2) Data extraction and analysis: extracting raw data from different data sources according to protocols, including: Weather data JSON: parsed into key-value pairs; Water condition data Modbus: read register values and convert them into physical quantities; Power generation plan XML: parsing tag content; The formula is as follows: For any data source S, the analytical function P( S ) to convert it into intermediate data: ； Where, Indicates the data timestamp, Indicates that the original values include water level, flow rate, and power generation. represents the original unit; (3) Data conversion and unit standardization, converting data in different units into system standard units: ； Where, express Unit conversion factors; (4) Unified data model mapping: Defining a standard time series data model , contains the following fields: ； Where, t std indicates the aligned standard timestamp, which is aligned to the whole minute or hour; type indicates the data type, including water level, flow, and power generation; source_id indicates the data source identifier; (5) Time alignment and interpolation: For non-uniformly sampled data, linear interpolation is used to generate uniform timestamp data: ； (6) Data association and storage: associate standardized data by timestamp and data type and store them in a historical database: 。 10. A key information query system for river basin cascade water resources decision-making for real-time mobile query, characterized by: After the data processing layer performs dynamic data aggregation and multi-source heterogeneous data fusion, the water and rainfall heat map module executes the heat map generation optimization algorithm. The specific method is as follows: (1) Set input and output: Input: Basin rainfall data in 6-hour units over the past 36 hours, including time, location, and rainfall amount; Output: interactive heatmap with a frame rate of ≥30fps, supporting zooming and timeline dragging; (2) Spatiotemporal data aggregation: Time block: Divide the 36-hour data into 6-hour blocks, each containing data from multiple observation points: ； Where, ( xi , yi ) represents the latitude and longitude coordinates of the observation point, normalized to the interval [0,1]; ri Indicates the accumulated rainfall within 6 hours; k Indicates the time block index, a total of 6 blocks; Spatial grid division: Divide the watershed map into N × N , calculate each grid cell ( m , n ) density value dm , n : ； Where, Sm , n Indicates that it falls into the grid ( m , n ) observation point set, μx , μy represents the grid center coordinates, σ Indicates the Gaussian kernel radius, controls the smoothness, default σ =0.05; (3) Color mapping and texture generation: Color coding, density values dm , n Mapping to RGBA color space uses a piecewise linear function: ； Where, express Density range dynamic threshold; α Indicates transparency, which is positively correlated with density; Texture Compression: Will N × N The color matrix is encoded as a WebGL texture, using the ASTC compression format to reduce video memory usage; (4) WebGL rendering optimization: Shader Programming: Executing Vertex Shaders and Fragment Shaders in the GPU: The vertex shader converts the mesh coordinates to screen positions, the fragment shader mixes colors based on the texture sampling results, and superimposes the timeline markers; then all mesh vertex data is submitted at once using gl.drawArrays().

Images