Reservoir capacity intelligent analysis method and system based on multi-source data fusion

Abstract

The invention relates to the technical field of reservoir storage capacity analysis, in particular to an intelligent reservoir storage capacity analysis method and system based on multi-source data fusion, and the method comprises the following steps: collecting ground water level sensor data, meteorological station data, LiDAR point cloud data and aerial image data, synchronizing the data through a unified timestamp, integrating the data, summarizing the integrated data, and carrying out the analysis of the reservoir storage capacity. And generating a comprehensive multi-source data set. According to the invention, through real-time updating and accurate tracking of daily reservoir capacity changes, the accuracy of reservoir capacity estimation is significantly improved, daily cumulative statistics of water level, rainfall and evaporation capacity data is integrated, dynamic and detailed management is promoted, seasonal changes and long-term trends of the reservoir water level are analyzed, and the reservoir capacity estimation accuracy is improved. The method can accurately reveal the direct effect of manual intervention and the influence of the manual intervention on the natural supply mode of the reservoir, can make a scientific decision based on a comprehensive data view, integrates short-term prediction and real-time data flow monitoring to enable an early warning system to quickly respond, and optimizes the efficiency and effectiveness of disaster management.

Description

Technical Field

[0001] The present invention relates to the technical field of reservoir capacity analysis, and in particular to a reservoir capacity intelligent analysis method and system based on multi-source data fusion. Background Art

[0002] The field of reservoir capacity analysis technology involves methods and systems for real-time or historical data analysis of reservoir capacity, water level, inflow and outflow, etc., combining hydrology, environmental engineering and information technology, and using multi-source data such as sensor data, satellite remote sensing images and geographic information systems (GIS) for comprehensive analysis. The purpose is to optimize water resources management, predict changes in reservoir water volume, and support flood warning and drought management decisions, including but not limited to the construction of watershed models, accurate data fusion processing technology, and the development of prediction models based on artificial intelligence.

[0003] Among them, the intelligent analysis method of reservoir capacity is a method that uses advanced data analysis technology to evaluate and predict the storage capacity and water volume status of the reservoir. It mainly includes the collection of multi-source data (such as meteorological data, hydrological data, etc.), and data fusion and analysis through intelligent algorithms to achieve real-time monitoring and prediction of reservoir water level and storage capacity. Its main purpose is to help water conservancy management departments to effectively allocate and manage water resources, especially during floods or droughts. It can provide a scientific basis for rapid decision-making and response to ensure the rational use and safety of water resources.

[0004] Existing technologies often lack the necessary real-time and flexibility when processing multi-source data sets, resulting in insufficient response speed during data integration and analysis. In emergencies such as sudden floods, reservoir capacity information cannot be adjusted and updated in a timely manner, delaying key response measures. In addition, the analysis of existing technologies often does not fully consider the long-term and seasonal impacts of climate change and human activities, resulting in decision-making relying on incomplete or outdated information, and the limitations of predicting changes in reservoir water levels and storage capacity, especially when responding to rapidly changing environmental conditions, cannot meet the needs of modern water resources management for high efficiency and accuracy. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a reservoir capacity intelligent analysis method based on multi-source data fusion.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a reservoir capacity intelligent analysis method based on multi-source data fusion, comprising the following steps: S1: Collect ground water level sensor data, weather station data, LiDAR point cloud data, and aerial image data, synchronize the data using a unified timestamp, integrate the data, and generate a comprehensive multi-source data set; S2: Based on the comprehensive multi-source data set, water level and rainfall data are extracted, rainfall and evaporation data are accumulated and counted daily, water level change values ​​at corresponding time points are calculated synchronously, water level rise and fall and rainfall evaporation difference are analyzed according to the calculation results, actual reservoir capacity of the day is estimated, daily reservoir capacity update is captured, and a daily dynamic table of reservoir capacity is constructed; S3: Based on the daily dynamic table of reservoir storage capacity, reveal the change pattern of reservoir water level, determine the change of reservoir storage capacity with seasonal water level, analyze the correlation between reservoir water level change and climate change and artificial water release activities, identify the natural recharge and human intervention mode of the reservoir according to the analysis results, combine periodicity and trend data, draw a comprehensive view of seasonal changes and trends, and generate reservoir storage capacity trend analysis results; S4: Based on the reservoir capacity trend analysis results, monitor the real-time data flow, collect weather forecast and remote sensing monitoring data, analyze the reservoir capacity changes in the short term, and generate reservoir capacity intelligent analysis results.

[0007] As a further solution of the present invention, the steps of acquiring the comprehensive multi-source data set are: S111: Install water level sensors, set up weather station equipment, configure remote sensing satellites, collect data, and generate initial data sets; S112: performing time standardization processing on the initial data set, applying time synchronization to correct time deviation, confirming that data timestamps are unified, and generating a time synchronized data set; S113: Performing data cleaning on the time-synchronized data set, including outlier processing and data missing filling, statistical analysis and data fusion, to generate a comprehensive multi-source data set.

[0008] As a further solution of the present invention, the steps for calculating the water level change value at the corresponding time point are: S211: Based on the comprehensive multi-source data set, the reservoir water level data and rainfall data of the day are called to generate an initial hydrological data set; S212: Using the initial hydrological data set, calculating the cumulative sum of the rainfall and evaporation on the day to generate a daily cumulative rainfall and evaporation statistical result; S213: Based on the daily cumulative rainfall evaporation statistical results, the formula is used: ; Calculate the water level change at a time point , generate water level change data; among them, Represents the water level at the current moment. Represents the water level at the previous moment. represents the cumulative rainfall, Represents the cumulative evaporation.

[0009] As a further solution of the present invention, the steps for obtaining the daily dynamic table of reservoir capacity are: S221: Based on the water level change data, combined with the rainfall evaporation difference data, a correlation analysis is performed to determine the statistical relationship between the two, and a correlation analysis record of water level change and meteorological conditions is generated; S222: using the water level change and meteorological condition correlation analysis record, combined with a known reservoir capacity curve, determining the actual reservoir capacity corresponding to the current water level, and generating an actual reservoir capacity estimation result; S223: Integrate the actual storage capacity estimation results with the historical storage capacity data, record daily reservoir storage capacity changes, water level and meteorological conditions related data, and update and output the reservoir storage capacity daily dynamic table.

[0010] As a further solution of the present invention, the analysis steps of the correlation between the reservoir water level change and climate change and artificial water release activities are: S311: Analyze the reservoir capacity daily dynamic table data, extract the reservoir water level daily reading, perform linear regression analysis to reveal the water level change trend, and obtain the reservoir water level change pattern; S312: Based on the reservoir water level variation pattern, a periodic analysis is applied to identify typical water level variations in each season, and the seasonal water level variation pattern is revealed through time series analysis to obtain seasonal water level characteristics; S313: Combined with the seasonal water level characteristics, climate data and reservoir water release records, the formula is used. ; Calculate the water level change response coefficient , analyzed the correlation between water level changes and climate and water release activities, and obtained the analysis results of factors affecting water level changes, among which, and Represent the quantitative indicators of climate and water release activities, Indicates the number of sample days.

[0011] As a further solution of the present invention, the steps for obtaining the reservoir capacity trend analysis result are: S321: Based on the analysis results of the factors affecting water level change, distinguish the impact of natural recharge and human intervention on the reservoir water level, distinguish natural and human factors, and generate a report on the contribution of natural recharge and human intervention; S322: Based on the natural recharge and human intervention contribution report, the historical seasonal and long-term trend data of the reservoir are integrated to reveal the relationship between periodic changes and trend changes, and to construct a seasonal change and trend relationship diagram; S323: Based on the seasonal change and trend relationship diagram, reservoir capacity trend analysis is performed to predict future changes in reservoir water levels and generate reservoir capacity trend analysis results.

[0012] As a further solution of the present invention, the steps for obtaining the intelligent analysis result of the reservoir capacity are: S411: using the reservoir capacity trend analysis result and combining it with the real-time data stream of the sensor, performing data fusion, comparing the current reservoir water level with the historical level during the same period, and generating real-time reservoir capacity change information; S412: Based on the real-time reservoir capacity change information, the information is integrated with the synchronously updated meteorological forecast data, LiDAR point cloud and aerial image data, and the reservoir capacity change in the short term in the future is predicted through time series analysis to obtain a short-term reservoir capacity prediction result; S413: Based on the short-term reservoir capacity prediction results, taking into account the changes in rainfall, evaporation and inflow, an optimal water and reservoir adjustment strategy is formulated, and the reservoir capacity intelligent analysis results are output.

[0013] A reservoir capacity intelligent analysis system based on multi-source data fusion, comprising: The data acquisition and correction module installs water level sensors, sets up weather station equipment, combines LiDAR point clouds and aerial images to collect data, applies time synchronization to correct time deviations, performs data cleaning, statistical analysis and data fusion, and generates a comprehensive multi-source data set; The water level data generation module calculates the cumulative sum of the rainfall and evaporation on the day based on the comprehensive multi-source data set, calculates the water level change value at the time point, and generates water level change data; The reservoir capacity dynamic update module performs correlation analysis based on the water level change data and combined with the rainfall evaporation difference data to determine the actual reservoir capacity corresponding to the current water level, records the daily reservoir capacity change, water level and meteorological conditions related data, and updates and outputs the reservoir capacity daily dynamic table; The water level trend analysis module analyzes the daily dynamic table data of the reservoir storage capacity, performs linear regression analysis to reveal the trend of water level changes, applies periodic analysis to identify typical water level changes in each season, integrates climate data and reservoir water release records, analyzes the correlation between water level changes and climate and water release activities, and obtains the analysis results of factors affecting water level changes; The influencing factor diagnosis module distinguishes the influence of natural recharge and human intervention on the reservoir water level based on the analysis results of the influencing factors of the water level change, draws a seasonal change and trend relationship diagram, predicts future changes in the reservoir water level, and generates a reservoir storage capacity trend analysis result; The future trend prediction module uses the reservoir capacity trend analysis results to compare the current reservoir water level with the historical level during the same period, and integrates it with the synchronously updated meteorological forecast data, LiDAR point cloud and aerial image data to predict the changes in reservoir capacity in the short term in the future, comprehensively consider the changes in rainfall, evaporation and inflow, and output the intelligent analysis results of the reservoir capacity.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, the accuracy of reservoir capacity estimation is significantly improved by real-time updating and accurate tracking of daily storage capacity changes, and the daily cumulative statistics of water level, rainfall and evaporation data are integrated to promote dynamic and detailed management. The seasonal changes and long-term trends of reservoir water levels can be analyzed, and the direct effects of human intervention and their impact on the natural recharge pattern of the reservoir can be accurately revealed. Scientific decisions can be made based on a comprehensive data view, and the integration of short-term prediction and real-time data flow monitoring enables the early warning system to respond quickly, thereby optimizing the efficiency and effectiveness of disaster management. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a main step flow chart of the present invention; Figure 2 A flowchart for obtaining a comprehensive multi-source data set according to the present invention; Figure 3 It is a flow chart of calculating the water level change value at a corresponding time point of the present invention; Figure 4 The present invention is a flowchart for obtaining a daily dynamic table of reservoir storage capacity; Figure 5 It is a flow chart of the correlation analysis of reservoir water level changes of the present invention; Figure 6 A flow chart for obtaining the reservoir capacity trend analysis results of the present invention; Figure 7 The present invention is a flowchart for obtaining the results of intelligent analysis of reservoir capacity. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0018] See also Figure 1, a reservoir capacity intelligent analysis method based on multi-source data fusion, comprising the following steps: S1: Collect ground water level sensor data, weather station data, LiDAR point cloud data, and aerial image data, synchronize the data using a unified timestamp, integrate the data, and generate a comprehensive multi-source data set; S2: Based on the comprehensive multi-source data set, water level and rainfall data are extracted, rainfall and evaporation data are accumulated and counted daily, and the water level change value at the corresponding time point is calculated synchronously. According to the calculation results, the water level rise and fall and rainfall evaporation difference are analyzed to estimate the actual reservoir capacity of the day, capture the daily storage capacity update, and construct a daily dynamic table of reservoir capacity; S3: Based on the daily dynamic table of reservoir storage capacity, reveal the change pattern of reservoir water level, determine the change of reservoir storage capacity with seasonal water level, analyze the correlation between reservoir water level change and climate change and artificial water release activities, identify the natural recharge and human intervention mode of reservoir based on the analysis results, combine periodic and trend data, draw a comprehensive view of seasonal changes and trends, and generate reservoir storage capacity trend analysis results; S4: Based on the reservoir capacity trend analysis results, monitor the real-time data stream, collect meteorological forecast and remote sensing monitoring data, analyze the changes in reservoir capacity in the short term, and generate intelligent analysis results of reservoir capacity.

[0019] The comprehensive multi-source data set includes synchronized timestamps, data integration results, and data source category identification records; the daily dynamic table of reservoir capacity includes daily water level change records, daily rainfall statistics, and daily evaporation statistics; the reservoir capacity trend analysis results include seasonal change graphs, long-term trend graphs, and manual intervention analysis results; the reservoir capacity intelligent analysis results include short-term change prediction results and real-time reservoir capacity data update records.

[0020] See also Figure 2 ,The steps to obtain the comprehensive multi-source dataset are: S111: Install water level sensors, set up weather station equipment, and collect data in combination with LiDAR point clouds and aerial images to generate initial data sets; When installing ground water level sensors, a location at a certain distance from the water body is selected to ensure that the sensor can accurately capture data when the water level changes; the meteorological station equipment is placed in an open area to avoid being blocked by surrounding tall buildings, ensuring the accurate collection of meteorological data such as wind speed and temperature; at the same time, LiDAR point cloud data and aerial image data are combined to ensure time synchronization and precise alignment of images. The data outputs of these devices must all go through a verification procedure to check the integrity and accuracy of the data, and then be transmitted to the data center through a secure network protocol to generate a preliminary data set. This data set records various environmental parameters and their timestamps, providing basic data for subsequent analysis.

[0021] S112: Perform time standardization on the initial data set, apply time synchronization to correct time deviation, confirm that the data timestamps are unified, and generate a time synchronized data set; In the data center, high-performance servers are used to synchronize the collected initial data sets. During the process, the timestamps of each data point are compared through time synchronization software, and the network time protocol (NTP) is used to correct the time error to ensure the consistency of all data on the time axis. This can effectively solve the data time difference problem caused by different sensor locations. The synchronized data is verified and invalid data with time deviation greater than the set threshold is eliminated, thereby ensuring data quality and generating a time-synchronized data set, which provides a solid time foundation for accurate data analysis and decision-making.

[0022] S113: Perform data cleaning on the time-synchronized data set, including outlier processing and data missing filling, statistical analysis and data fusion, and generate a comprehensive multi-source data set; In the data cleaning and fusion stage, the time-synchronized data set is first subjected to a detailed data quality analysis, including the identification and processing of outliers in the data, by calculating the standard deviation and mean of the data, and then using statistical indicators to identify outliers that exceed three times the standard deviation and remove them. Next, the missing data points are filled by linear interpolation to ensure the integrity of the data set. In addition, for the fusion of multi-source data, the weighted average method is used to consider the reliability and relevance of different data sources. This method integrates meteorological station data, ground water level data, and satellite remote sensing data, and finally generates a comprehensive multi-source data set, which not only contains various environmental and geographic information, but also ensures the accuracy and availability of the data through time synchronization and data cleaning, providing a solid foundation for subsequent data analysis and decision-making.

[0023] See also Figure 3 , the calculation steps of the water level change value at the corresponding time point are: S211: Based on the comprehensive multi-source data set, the reservoir water level data and rainfall data of the day are called to generate the initial hydrological data set; The process of extracting water level and rainfall data from comprehensive multi-source datasets begins with the data integration and preprocessing stage. First, data cleaning techniques are applied to remove invalid and abnormal data, such as removing noise data generated by water level sensors due to heavy rain or equipment failure. Next, the raw data collected by the sensors are calibrated, including time synchronization and range calibration, to ensure that all data are stored in a unified standard and format. In addition, data interpolation methods are used to deal with data missing problems, especially in uncommon but critical time periods in rainfall data. Then, the data is spatially calibrated in combination with the geographic information system (GIS) to correspond the data to specific geographic locations to support subsequent hydrological analysis and decision making, accurately generate initial hydrological datasets that meet the analysis requirements, and provide a reliable foundation for the next step of hydrological modeling and resource assessment.

[0024] S212: using the initial hydrological data set, calculating the cumulative sum of the rainfall and evaporation on the day, and generating daily cumulative rainfall and evaporation statistics; After obtaining the initial hydrological data set, the rainfall and evaporation data are accumulated daily. The first step is to align the data to ensure that the data from different sensors and sources are aligned according to a unified time standard. Then the data for the whole day is accumulated hour by hour, including the sum of the hourly rainfall and evaporation data. Under extreme weather conditions, the accuracy and representativeness of the data need to be continuously monitored. At the same time, data smoothing technology is used to reduce the impact of mutations and anomalies to ensure the stability and reliability of the statistical output. After the accumulation is completed, the data is verified again to ensure that it accurately reflects the actual rainfall and evaporation conditions, thereby ensuring that the statistical data can effectively support water resource management and the formulation of flood prevention measures.

[0025] S213: Based on the daily cumulative rainfall evaporation statistics, the formula is used: ; Calculate the water level change at a time point , generate water level change data; among them, Represents the water level at the current moment. Represents the water level at the previous moment. represents the cumulative rainfall, Represents the cumulative evaporation; At a certain monitoring point, the current water level The water level at the last moment is 5.2 meters. The total rainfall for the day was 5.0 meters. is 8 mm, the cumulative evaporation is 3 mm. First, calculate the square root of the absolute value of the water level change, and we get , and then perform multiplication and division operations, m, which means that under given conditions, the change in water level is about 1.192 m. This result shows that the change in water level under the influence of rainfall and evaporation provides a dynamic response of actual measurement for the hydrological model.

[0026] See also Figure 4 , the steps to obtain the daily dynamic table of reservoir capacity are: S221: Based on the water level change data, combined with the rainfall evaporation difference data, a correlation analysis is performed to determine the statistical relationship between the two, and a correlation analysis record of water level change and meteorological conditions is generated; When combining water level change data and rainfall-evaporation difference data, the data are first normalized to eliminate dimensional influence and scale effect. Next, advanced statistical software, such as R or Python, is used to calculate the Pearson correlation coefficient between water level and rainfall to determine the strength of the linear relationship between the two. At the same time, linear regression analysis is applied to predict future water level change trends. Based on the regression model results, scatter plots and trend lines are drawn to show in detail the dynamic relationship between water level changes and rainfall-evaporation differences. Finally, the analysis results are compiled into a detailed correlation analysis report, which includes statistical test results, charts, model parameters and their interpretations, providing a scientific basis for reservoir management decisions and enhancing the data support for decision-making and the accuracy of predictions.

[0027] S222: using the correlation analysis records between water level changes and meteorological conditions, combined with known reservoir capacity curves, determine the actual reservoir capacity corresponding to the current water level, and generate actual reservoir capacity estimation results; Using the water level change and meteorological conditions correlation analysis report, combined with the reservoir volume curve, reservoir capacity estimation is performed. First, the accuracy and completeness of the water level measurement data are carefully checked, and then the reservoir capacity corresponding to each water level reading is calculated through professional hydrological software, such as HEC-HMS. This calculation takes into account the reservoir terrain and reservoir characteristics. The model used can reflect the actual reservoir's ability to respond to rainfall events. Furthermore, through historical data comparison and analysis, abnormal trends or periodic changes in current water level changes are identified and integrated into an actual reservoir capacity estimation report, providing managers with data support and scientific basis for operational decision-making.

[0028] S223: Integrate the actual storage capacity estimation results with the historical storage capacity data, record the daily reservoir storage capacity changes, water level and meteorological conditions related data, and update and output the reservoir storage capacity daily dynamic table; After integrating the reservoir capacity estimation report and historical data, the daily dynamic table of reservoir capacity is updated and maintained. First, an automated data integration system is used to ensure that all newly acquired hydrological data are entered into the database in a timely and accurate manner. Then, database management software, such as Oracle or SQLServer, is used to regularly update the daily dynamic table, including data verification, processing and storage, to ensure the accuracy and timeliness of the data in the table. In addition, an automatic report generation tool is set up to generate a daily reservoir capacity change report. The report includes not only reservoir capacity data, but also water level, meteorological conditions and the output of its prediction model, making reservoir management more efficient, providing the ability to quickly respond to environmental changes, and helping managers optimize water resource allocation and adjust emergency measures.

[0029] See also Figure 5 , the analysis steps of the correlation between reservoir water level changes and climate change and artificial water release activities are: S311: Analyze the reservoir capacity daily dynamic table data, extract the reservoir water level daily readings, perform linear regression analysis to reveal the water level change trend, and obtain the reservoir water level change pattern; Before conducting a linear regression analysis, it is first necessary to extract water level data from the reservoir capacity daily dynamic table, including the highest and lowest water levels of each day; next, preprocess the data, including removing obvious abnormal data points, such as incorrect water level values ​​recorded due to equipment failure, and using the average water level of adjacent days to fill in missing data records to ensure data continuity and consistency; then, apply the preprocessed data to the linear regression model, and construct the relationship between data points and time to estimate the trend line of water level changes. The main outputs include the slope and intercept of the trend line, which are crucial to understanding the change pattern of reservoir water level over time. It depicts how the water level increases or decreases over time, and draws a water level change trend graph, from which the change pattern of water level over time can be intuitively seen.

[0030] S312: Based on the reservoir water level change pattern, periodic analysis is applied to identify the typical water level changes in each season, and the seasonal water level change pattern is revealed through time series analysis to obtain seasonal water level characteristics; Before applying periodic analysis techniques to identify typical water level changes in each season, the water level data is first normalized to ensure that the data is analyzed on the same scale. This is done by subtracting the mean of the data and dividing by the standard deviation. The normalized data is then processed using a Fast Fourier Transform (FFT) to convert the water level time series from the time domain to the frequency domain, making the periodic components more obvious. In the frequency domain analysis, the main cycles of water level changes are determined by identifying the main frequency components in the spectrum. Each peak represents an important periodic component, which reflects how the water level changes with seasonal factors. Then, the main periodic components in the frequency domain are selected for inverse transformation, and the components are converted back to a time series, resulting in a clear seasonal water level change pattern, which reveals how the water level rises or falls in each season and provides a basis for reservoir management. Finally, the seasonal change pattern is plotted into a chart to vividly show the typical fluctuations of water levels in different seasons, allowing reservoir managers to better predict and regulate water levels.

[0031] S313: Combined with seasonal water level characteristics, climate data and reservoir release records, the formula is used to ; Calculate the water level change response coefficient , analyzed the correlation between water level changes and climate and water release activities, and obtained the analysis results of factors affecting water level changes, among which, and Represent the quantitative indicators of climate and water release activities, represents the number of sample days; The climate data for one day is set to consist of temperature and rainfall, with the temperature being 30 degrees Celsius and the rainfall being 5 mm. is calculated as the average of temperature and rainfall, that is, ; represents the water release data for that day, set to 100 cubic meters, then ; is the number of observation sample days, set to 30 days, then . Substitute the values ​​into the formula to calculate: ; The results show that, based on consideration of climate data and water release volume, the water level change response coefficient is 2.19, which means that water level changes are significantly affected by climate and water release activities, and this value reflects the degree of water level change under given climate conditions and water release activities.

[0032] See also Figure 6 , the steps to obtain the reservoir capacity trend analysis results are: S321: Based on the analysis results of factors affecting water level changes, distinguish the impact of natural recharge and human intervention on reservoir water level, distinguish natural and human factors, and generate a report on the contribution of natural recharge and human intervention; Different data sources were integrated to identify the impact patterns of natural recharge and human intervention. First, hydrological and meteorological data including rainfall, snowmelt, flood discharge and irrigation discharge were collected. Data preprocessing included outlier detection, missing value filling and data standardization to improve the accuracy of subsequent analysis. Classification algorithms such as decision trees and random forests were used to analyze how factors affect reservoir water levels individually and collectively. By comparing the results with historical water level changes, the impact strength and pattern of each factor were verified, thus determining that rainfall and snowmelt have a significant impact on water levels mainly in spring and early summer, while human discharge is significant in the water demand season, providing a preliminary scientific basis for subsequent trend analysis and reservoir management decisions.

[0033] S322: Based on the report on natural recharge and human intervention contribution, integrated with the historical seasonal and long-term trend data of the reservoir, the relationship between periodic changes and trend changes is revealed, and a seasonal change and trend relationship diagram is constructed; Combine natural and human-influenced patterns with seasonal and long-term trends in historical water level data. By using advanced statistical methods and graphical visualization tools such as SAS and Tableau, time series graphs of water level changes are drawn, which clearly show the water level fluctuation patterns in different seasons and years. In addition, linear and nonlinear regression models are applied to explore the intrinsic factors of water level changes and how they develop over time, not only identifying the main periodic changes, but also revealing long-term trends. Trends and periodic patterns are integrated in a comprehensive view, providing a dual perspective for reservoir operation and management, and enhancing the understanding and prediction capabilities of the complex dynamics of reservoir water level changes.

[0034] S323: Based on the seasonal change and trend relationship diagram, reservoir capacity trend analysis is performed to predict future reservoir water level changes and generate reservoir capacity trend analysis results; Comprehensive views and forecasting models are used to comprehensively analyze reservoir storage trends. The process uses machine learning techniques, especially neural networks and support vector machines (SVM) to predict future water level changes. In this process, we first use historical water level data to train and verify the model to ensure the accuracy and reliability of the prediction. Through continuous model optimization and parameter adjustment, such as changing the network structure or adjusting the kernel function of SVM, it can adapt to different data characteristics and changing trends. The final report generated shows in detail the comparison between the model prediction results and the actual observations, as well as charts of future trends, providing reservoir managers with accurate data support, enabling them to make more scientific decisions and effectively plan water resource utilization and risk management strategies.

[0035] See also Figure 7 , the steps to obtain the intelligent analysis results of reservoir capacity are: S411: Using the reservoir capacity trend analysis results and combining them with the real-time data stream of the sensor, data fusion is performed to compare the current reservoir water level with the historical level during the same period to generate real-time reservoir capacity change information; Using the results of reservoir capacity trend analysis and real-time sensor monitoring data streams, we first collect water level and flow rate data from monitoring sites, and transmit the data to the central processor in real time via wireless networks. We use data fusion technology to integrate real-time data with historical capacity data, and use a weighted average method to process data at different time points, focusing on the instantaneous impact of extreme weather events on water levels and the long-term effects of seasonal changes on flow rates. By comparing the current reservoir water level with the historical level for the same period, we analyze the real-time water level changes and expected range of changes, and generate real-time capacity change information. This information will directly support the next step of capacity forecasting and analysis, provide a scientific basis for reservoir management, and ensure safe and effective reservoir operations.

[0036] S412: Based on the real-time reservoir capacity change information, integrated with the synchronously updated meteorological forecast data, LiDAR point cloud and aerial image data, the reservoir capacity change in the short term in the future is predicted through time series analysis to obtain the short-term reservoir capacity prediction result; When predicting short-term storage capacity changes, the formula is used: ; here, represents the predicted storage capacity change, , and Represent the variables obtained from real-time reservoir capacity change information, weather forecast data, LiDAR point cloud and aerial image data. Model coefficients It is determined through statistical analysis of historical data.

[0037] Method for determining the coefficient: First, collect historical storage capacity data , and the corresponding , , Data. Use the data to build a multiple linear regression model, where As the dependent variable, , , as independent variable.

[0038] Statistical analysis: The following system of equations was solved to minimize the sum of squares of the prediction errors using the least squares method: ; Optimization problems are solved by numerical methods such as gradient descent or directly using regression analysis functions in statistical software packages.

[0039] Model validation: After solving the model coefficients, evaluate the predictive performance of the model through cross-validation or retaining part of the data as a test set to ensure the validity of the coefficients and the generalization ability of the model.

[0040] Apply the model: Once the model coefficients is determined and applied to the latest collection , , Data to predict future storage capacity changes .

[0041] For example, if historical data is used to determine , , , , and the current , , ,but: ; The result is the predicted storage capacity change value, which guides reservoir management decisions. The detailed calculation process and coefficient determination method ensure the accuracy and adaptability of the model, making it a powerful tool for reservoir management.

[0042] S413: Based on the short-term storage capacity prediction results, taking into account the changes in rainfall, evaporation and inflow, the optimal water and reservoir adjustment strategy is formulated, and the intelligent analysis results of the reservoir capacity are output; Based on the short-term storage capacity forecast results, a comprehensive analysis is conducted on the possible impact of meteorological factors such as rainfall, evaporation and inflow on the reservoir water level. First, recent rainfall forecast data and real-time evaporation rate are collected, and combined with the hydrological model to predict the reservoir inflow and outflow in the next week. Then, the data is analyzed based on a multi-objective planning algorithm to find the optimal reservoir release and storage balance point to prevent excessively high water levels from causing overflows or excessively low water levels from affecting power generation and irrigation. Finally, the above analysis results are integrated through a decision support system to generate intelligent analysis results of reservoir capacity, provide dynamic adjustment strategies for reservoir management, optimize water resource allocation, and ensure the economy and ecological safety of reservoir operations.

[0043] A reservoir capacity intelligent analysis system based on multi-source data fusion, comprising: The data acquisition and correction module installs water level sensors, sets up weather station equipment, combines LiDAR point clouds and aerial images to collect data, applies time synchronization to correct time deviations, performs data cleaning, statistical analysis and data fusion, and generates a comprehensive multi-source data set; The water level data generation module is based on a comprehensive multi-source data set, which counts the cumulative total of rainfall and evaporation on the day, calculates the water level change value at a time point, and generates water level change data; The reservoir capacity dynamic update module performs correlation analysis based on water level change data and rainfall evaporation difference data to determine the actual reservoir capacity corresponding to the current water level, record daily reservoir capacity changes, water level and meteorological conditions, and update and output the reservoir capacity daily dynamic table; The water level trend analysis module analyzes the daily dynamic table data of reservoir storage capacity, performs linear regression analysis to reveal the trend of water level changes, applies periodic analysis to identify typical water level changes in each season, integrates climate data and reservoir water release records, analyzes the correlation between water level changes and climate and water release activities, and obtains the analysis results of factors affecting water level changes; The influencing factor diagnosis module distinguishes the impact of natural recharge and human intervention on reservoir water level based on the analysis results of the influencing factors of water level change, draws a seasonal change and trend relationship diagram, predicts future changes in reservoir water level, and generates reservoir storage capacity trend analysis results; The future trend prediction module uses the results of reservoir capacity trend analysis to compare the current reservoir water level with the historical level during the same period, and integrates it with the synchronously updated meteorological forecast data, LiDAR point cloud and aerial image data to predict the changes in reservoir capacity in the short term in the future, comprehensively consider the changes in rainfall, evaporation and inflow, and output the intelligent analysis results of reservoir capacity.

[0044] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A reservoir capacity intelligent analysis method based on multi-source data fusion, characterized in that: The following steps are involved: Collect ground water level sensor data, weather station data, LiDAR point cloud data, and aerial image data, synchronize the data using a unified timestamp, integrate the data for aggregation, and generate a comprehensive multi-source data set; Based on the comprehensive multi-source data set, water level and rainfall data are extracted, rainfall and evaporation data are accumulated and counted daily, water level change values ​​at corresponding time points are calculated synchronously, water level rise and fall and rainfall evaporation difference are analyzed according to the calculation results, actual reservoir capacity of the day is estimated, daily reservoir capacity update is captured, and a reservoir capacity daily dynamic table is constructed; Based on the daily dynamic table of reservoir storage capacity, reveal the change pattern of reservoir water level, determine the change of reservoir storage capacity with seasonal water level, analyze the correlation between reservoir water level change and climate change and artificial water release activities, identify the natural recharge and human intervention mode of reservoir according to the analysis results, combine periodic and trend data, draw a comprehensive view of seasonal changes and trends, and generate reservoir storage capacity trend analysis results; Based on the reservoir capacity trend analysis results, real-time data streams are monitored, meteorological forecasts and remote sensing monitoring data are collected, short-term reservoir capacity changes are analyzed, and reservoir capacity intelligent analysis results are generated.

2. The reservoir capacity intelligent analysis method based on multi-source data fusion according to claim 1 is characterized in that: The steps of obtaining the comprehensive multi-source data set are: Install water level sensors, set up weather station equipment, and collect data using LiDAR point clouds and aerial images to generate an initial data set; Performing time standardization processing on the initial data set, applying time synchronization to correct time deviation, confirming that the data timestamps are unified, and generating a time synchronized data set; The time-synchronized data set is cleaned, including outlier processing and data missing filling, and statistical analysis and data fusion are performed to generate a comprehensive multi-source data set.

3. The reservoir capacity intelligent analysis method based on multi-source data fusion according to claim 2 is characterized in that: The calculation steps of the water level change value at the corresponding time point are: Based on the comprehensive multi-source data set, the reservoir water level data and rainfall data of the day are called to generate an initial hydrological data set; Using the initial hydrological data set, the cumulative sum of the rainfall and evaporation on the day is calculated to generate a daily cumulative rainfall and evaporation statistical result; Based on the daily cumulative rainfall evaporation statistics, the formula is used: ； Calculate the water level change at a time point , generate water level change data; among them, Represents the water level at the current moment. Represents the water level at the previous moment. represents the cumulative rainfall, Represents the cumulative evaporation.

4. The reservoir capacity intelligent analysis method based on multi-source data fusion according to claim 3 is characterized in that: The steps for obtaining the daily dynamic table of reservoir capacity are as follows: Based on the water level change data, combined with the rainfall evaporation difference data, correlation analysis is performed to determine the statistical relationship between the two, and a correlation analysis record of water level change and meteorological conditions is generated; Using the water level change and meteorological condition correlation analysis records, combined with known reservoir capacity curves, the actual reservoir capacity corresponding to the current water level is determined to generate an actual reservoir capacity estimation result; Integrate the actual storage capacity estimation results with the historical storage capacity data, record daily reservoir storage capacity changes, water level and meteorological conditions related data, and update and output the reservoir storage capacity daily dynamic table.

5. The reservoir capacity intelligent analysis method based on multi-source data fusion according to claim 4 is characterized in that: The analysis steps for the correlation between the reservoir water level change and climate change and artificial water release activities are as follows: Analyze the reservoir capacity daily dynamic table data, extract the reservoir water level daily readings, perform linear regression analysis to reveal the water level change trend, and obtain the reservoir water level change pattern; Based on the reservoir water level variation pattern, periodic analysis is applied to identify the typical water level variation in each season, and the seasonal water level variation pattern is revealed through time series analysis to obtain seasonal water level characteristics; Combined with the seasonal water level characteristics, climate data and reservoir release records, the formula is used: ； Calculate the water level change response coefficient , analyzed the correlation between water level changes and climate and water release activities, and obtained the analysis results of factors affecting water level changes, among which, and Represent the quantitative indicators of climate and water release activities, Indicates the number of sample days.

6. The reservoir capacity intelligent analysis method based on multi-source data fusion according to claim 5 is characterized in that: The steps for obtaining the reservoir capacity trend analysis results are as follows: Based on the analysis results of the factors affecting water level changes, the impact of natural recharge and human intervention on reservoir water level is distinguished, natural and human factors are distinguished, and a report on the contribution of natural recharge and human intervention is generated; Based on the natural recharge and human intervention contribution reports, the historical seasonal and long-term trend data of the reservoir are integrated to reveal the relationship between periodic changes and trend changes, and to construct a seasonal change and trend relationship diagram; Based on the seasonal variation and trend relationship diagram, reservoir capacity trend analysis is performed to predict future reservoir water level changes and generate reservoir capacity trend analysis results.

7. The reservoir capacity intelligent analysis method based on multi-source data fusion according to claim 6 is characterized in that: The steps for obtaining the intelligent analysis results of the reservoir capacity are as follows: Using the reservoir capacity trend analysis results, combined with the real-time sensor data stream, data fusion is performed to compare the current reservoir water level with the historical level over the same period, and real-time reservoir capacity change information is generated; Based on the real-time storage capacity change information, the data is integrated with the synchronously updated meteorological forecast data, LiDAR point cloud and aerial image data, and the reservoir capacity change in the short term in the future is predicted through time series analysis to obtain the short-term storage capacity prediction result; According to the short-term storage capacity prediction results, the optimal water and reservoir adjustment strategy is formulated by comprehensively considering the changes in rainfall, evaporation and inflow, and the intelligent analysis results of the reservoir capacity are output.

8. A reservoir capacity intelligent analysis system based on multi-source data fusion, characterized in that: The system is used to execute the reservoir capacity intelligent analysis method based on multi-source data fusion according to any one of claims 1 to 7, comprising: The data acquisition and correction module installs water level sensors, sets up weather station equipment, combines LiDAR point clouds and aerial images to collect data, applies time synchronization to correct time deviations, performs data cleaning, statistical analysis and data fusion, and generates a comprehensive multi-source data set; The water level data generation module calculates the cumulative sum of the rainfall and evaporation on the day based on the comprehensive multi-source data set, calculates the water level change value at the time point, and generates water level change data; The reservoir capacity dynamic update module performs correlation analysis based on the water level change data and combined with the rainfall evaporation difference data to determine the actual reservoir capacity corresponding to the current water level, records the daily reservoir capacity change, water level and meteorological conditions related data, and updates and outputs the reservoir capacity daily dynamic table; The water level trend analysis module analyzes the daily dynamic table data of the reservoir storage capacity, performs linear regression analysis to reveal the trend of water level changes, applies periodic analysis to identify typical water level changes in each season, integrates climate data and reservoir water release records, analyzes the correlation between water level changes and climate and water release activities, and obtains the analysis results of factors affecting water level changes; The influencing factor diagnosis module distinguishes the influence of natural recharge and human intervention on the reservoir water level based on the analysis results of the influencing factors of the water level change, draws a seasonal change and trend relationship diagram, predicts future changes in the reservoir water level, and generates a reservoir storage capacity trend analysis result; The future trend prediction module uses the reservoir capacity trend analysis results to compare the current reservoir water level with the historical level during the same period, and integrates it with the synchronously updated meteorological forecast data, LiDAR point cloud and aerial image data to predict the changes in reservoir capacity in the short term in the future, comprehensively consider the changes in rainfall, evaporation and inflow, and output the intelligent analysis results of the reservoir capacity.

Images