Water level monitoring and management method based on multi-source information fusion

Abstract

The invention discloses a water level monitoring and management method based on multi-source information fusion, and relates to the technical field of water level monitoring and management. Comprising the following steps: acquiring a water level and water level related data by adopting a sensor, and acquiring surface feature information and meteorological data through GIS data; performing data cleaning and standardization on the collected data; a convolutional neural network is adopted to analyze infrared cloud picture data obtained by satellites to predict precipitation, a weighted average method is adopted to carry out multi-source information fusion, and water level prediction is carried out through a hydrological model; and performing real-time water level monitoring and water level management decision according to accurate water level data obtained by data fusion. Through the above steps, the system can provide a more comprehensive view angle for water level management by fusing multi-source information, and can more accurately predict the water level change trend, thereby optimizing the water resource management strategy and flood control measures.

Description

Technical Field

[0001] The present invention relates to the technical field of water level monitoring and management, and in particular to a water level monitoring and management method based on multi-source information fusion. Background Art

[0002] Traditional water level monitoring often relies on a single data source or method, which has many limitations. For example, relying solely on ground monitoring stations, although the measurement is direct, real-time and highly accurate, in many areas of China, hydrological and meteorological stations and agricultural monitoring stations are sparsely distributed, there are data delay problems, and station observations can only represent local information, which is time-consuming and labor-intensive, and it is difficult to meet the needs of large-scale dynamic and efficient monitoring.

[0003] Satellite remote sensing monitoring has the advantages of large-scale coverage, high temporal resolution, and near-real-time global information, making it the only source of information in areas without stations, but it also has its own disadvantages, such as low spatial resolution and large data uncertainty. Model simulation data can provide high temporal resolution and spatiotemporal continuous water resource element data, but its spatial resolution is often low and the uncertainty may be large.

[0004] Through multi-source information fusion, we can integrate the advantages of different data sources and overcome their respective limitations. For example, the fusion of ground monitoring, satellite remote sensing and model simulation data can more comprehensively and accurately monitor water level changes, providing strong support for the refined and dynamic management of water resources. Summary of the invention

[0005] In order to solve the above problems, the purpose of the present invention is to provide a water level monitoring and management method based on multi-source information fusion, which relates to the technical field of water level monitoring and management, and includes the following steps: Use sensors to obtain water level and water level related data, and use GIS data to obtain information on surface characteristics and meteorological data; Step 2: Use statistical methods to remove noise from the collected data by mean filtering, and then perform data cleaning and normalization. Step 3: Use convolutional neural network to analyze infrared cloud image data obtained by satellite to predict precipitation, use weighted average method to fuse multi-source information and predict water level through hydrological model; Step 4: Taking human influence factors into consideration, predict water level changes through scenario analysis, establish a water level risk warning mechanism based on real-time monitoring data, and release warning information through multiple channels.

[0006] Furthermore, the process of obtaining the water level and water level related data in step 1 includes: Ground monitoring data, satellite remote sensing data, and model simulation data; ground water level gauges provide accurate local water level values, satellite remote sensing data can cover a large area of ​​water, and model simulation data conducts predictive analysis of water levels based on historical data and related parameters; Reasonably arrange various types of sensors in the monitoring area, including water level sensors, flow rate sensors, and temperature sensors; water level sensors are used to directly measure water level height, flow rate sensors are used to obtain water flow velocity, and temperature sensors monitor water temperature. The above parameters will affect the water level; The distribution of sensors should take into account the geographical features and water flow characteristics of the monitoring area, and should be focused on the bends of the river, the places where the width changes, and the key locations of the water inlet and outlet of the reservoir; Satellite remote sensing data relies on sensors carried on various satellites, including optical sensors and synthetic aperture radar SAR; model simulation data is obtained based on computer simulation software by inputting parameters of relevant terrain, climate and watershed characteristics.

[0007] Furthermore, the process of obtaining the water level and water level related data in step 1 includes: In addition to sensor data, Geographic Information System (GIS) data is also obtained, including information on topography, river direction, and watershed area; Meteorological data is also an important source of data. Data on rainfall, air humidity, and air pressure are obtained through the meteorological department. Meteorological conditions are closely related to water levels, and the amount of rainfall directly affects the rise and fall of river water levels.

[0008] Furthermore, in step 2, the process of using statistical methods to filter the collected data to remove noise, perform data cleaning and perform normalization processing includes: For the various sensor data and other information source data collected, data cleaning is first performed to remove abnormal values ​​and erroneous data; in the ground water level meter data, thresholds are set and statistical methods are used to identify and eliminate obviously unreasonable values ​​due to equipment failures and external interference; corresponding cleaning methods are used for different types of data, and statistical methods-based mean filtering is used to remove noise for time series sensor data; Convert data from different sources and different magnitudes into a unified standard form; normalize the water level-related values ​​in satellite remote sensing data and the values ​​measured by ground water level gauges so that they are on the same order of magnitude.

[0009] Furthermore, the process of using satellite remote sensing to measure the water level of high data in step 3 includes: First, obtain the optical satellite image data of the target water area and the transit orbit data of the altimeter satellite closest to the target water area; pre-process the optical satellite image, and then use the visible light and infrared bands in the multispectral image data to extract water body information; use the object-oriented method to first segment the image into different objects, and then analyze the spectral, shape, and texture characteristics of the above objects for classification, so as to distinguish the water body part from the non-water body part, and randomly select a number of satellite altimeter points as the initial measurement points of the current measurement according to the boundary range of the water body part; judge and eliminate the initial measurement points with abnormal imaging based on the principle of satellite altimeter, and the retained ones are the target measurement points; then it is necessary to set the elimination judgment criteria for the target measurement points, and further retain the compliant target measurement points as valid measurement points; The satellite-measured water level verification value and average value of this measurement are calculated for the valid measuring points, and the average value is used as the satellite-observed water level of this measurement; the satellite-observed water level time series and the measured water level time series of multiple measurements are obtained, and the average deviation of the two time series is used as the calibration parameter. By optimizing the calibration parameters, the water level is finally calculated and calibrated.

[0010] Furthermore, the process of using satellite images to predict water level changes in step 3 includes: Convolutional neural networks are used to analyze infrared cloud image data obtained by satellites, and combined with ground-measured precipitation records, a precipitation calculation model is established to estimate precipitation. A large amount of satellite image data provided by the meteorological department is used for training to learn to identify the characteristics of different precipitation levels. The model is established by testing the generalization ability of the model on data from different regions. The infrared cloud image data obtained by satellites is analyzed to predict precipitation. Prediction of precipitation is the premise of water level change analysis; a distributed hydrological model is used to simulate the flow and accumulation of precipitation in the basin; the model includes the influence of temperature, topography, soil type, and vegetation coverage on precipitation runoff; a river network model is constructed based on the topography and water system structure of the basin, and the parameters of different river sections are determined; preliminary settings are made based on empirical values, the predicted results of the model are compared with the measured data, and the model parameters are adjusted to make the predicted results as close as possible to the measured results; an automatic optimization algorithm is used to automatically search for the optimal parameter combination by setting the objective function to minimize the error between the model output and the actual observed value; The collected data is divided into a training set and a validation set; the training set is used to calibrate the model parameters, and the validation set is used to test the accuracy and reliability of the model; 80% of the data is used as the training set, and 20% of the data is used as the validation set; The root mean square error (RMSE) is selected to verify the model, and the root mean square of the sum of squares of the errors between the model prediction value and the actual observation value is calculated; the smaller the RMSE value, the higher the accuracy of the model prediction; the verification result analysis shows that the accuracy and reliability of the model meet the requirements, and the model is used for water level prediction; if the verification result is not ideal, the model parameters are readjusted and then verified again; The processed data is input into the constructed hydrological model, and the model is run for simulation calculations. During the operation, the model will calculate the hydrological process, evaporation, infiltration, and runoff in the basin based on the input meteorological data, terrain data, and soil data, and finally obtain the predicted results of the water level.

[0011] Furthermore, the process of multi-source information fusion in step 3 includes: Calculate the water level difference, which is obtained by subtracting the sensor collection value from the satellite remote sensing data value. When the water level difference is less than the threshold, the next step of data fusion is performed; when the water level difference is greater than the threshold, an alarm is issued and data is collected again; The weighted average method is used for data fusion; the weights of different data sources in the fusion are determined according to their reliability and accuracy factors; higher weights are given to ground monitoring data with higher accuracy; and the weights are relatively low for satellite remote sensing data with greater uncertainty; that is, the weights are determined based on the error analysis of the data, and the data sources with small errors have larger weights; higher weights are given to the data from the water level sensor because it directly measures the water level; and the data from the rain gauge are weighted according to the degree of correlation with the water level change to obtain the water level data after data fusion.

[0012] Furthermore, in step 4, taking into account human influence factors, the water level changes are predicted through scenario analysis, combined with real-time monitoring data, a water level risk early warning mechanism is established, and the process of issuing early warning information through multiple channels includes: When there are human factors affecting the water level, such as flood discharge and artificial rainfall, the detection personnel input them into the system. In step 3, when predicting the water level change, the human factors are comprehensively considered to obtain the data after human influence; After obtaining accurate water level data through data fusion, real-time monitoring is carried out; Establish a water level risk warning mechanism; set different levels of water level warning values, set the flood level as a, the actual water level data as b, when , the water level reaches 90% of the flood level, a Level 1 alarm is issued; , the water level reaches the flood level and a Level 2 alarm is issued; , the water level exceeds the flood level by 10%, and a level 3 alarm is issued; and according to step 3, the water level impact range corresponding to floods of different scales is analyzed; Early warning information is released through various means on the Internet, including text messages notifying relevant management personnel, setting up alarm devices near the monitoring area, and detecting management web pages.

[0013] The beneficial effects of the present invention are: A water level monitoring and management method based on multi-source information fusion is provided. The method can integrate various types of sensor data including water level sensors, flow rate sensors, meteorological sensors, and other related information geographic information system data through multi-source information fusion technology to improve the accuracy and reliability of water level monitoring. The method of the invention provides a more comprehensive perspective for water level management and more accurately predicts water level change trends, thereby optimizing water resource management strategies and flood control measures.

[0014] A water level risk warning mechanism is established to send alarm information to the WeChat server and PC. Once the water level is detected to enter a dangerous range, the corresponding flood control emergency plan is activated, the embankment is reinforced, and the flood discharge channel is opened. When the water level continues to drop to a certain warning value, the decision-making department will take measures such as restricting water use and adjusting the reservoir release strategy. The water level change trend is predicted based on multi-source information, and water use plans for different water-using departments are formulated to achieve sustainable use of water resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a water level monitoring and management method based on multi-source information fusion according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0017] like Figure 1 As shown, a water level monitoring and management method based on multi-source information fusion relates to the technical field of water level monitoring and management, including the following steps: Use sensors to obtain water level and water level related data, and use GIS data to obtain information on surface characteristics and meteorological data; Step 2: Use statistical methods to remove noise from the collected data by mean filtering, and then perform data cleaning and normalization. Step 3: Use convolutional neural network to analyze infrared cloud image data obtained by satellite to predict precipitation, use weighted average method to fuse multi-source information and predict water level through hydrological model; Step 4: Taking human influence factors into consideration, predict water level changes through scenario analysis, establish a water level risk warning mechanism based on real-time monitoring data, and release warning information through multiple channels.

[0018] The process of obtaining the water level and water level related data in step 1 includes: Ground monitoring data, satellite remote sensing data, and model simulation data; ground water level gauges provide accurate local water level values, satellite remote sensing data can cover a large area of ​​water, and model simulation data conducts predictive analysis of water levels based on historical data and related parameters; Reasonably arrange various types of sensors in the monitoring area, including water level sensors, flow rate sensors, and temperature sensors; water level sensors are used to directly measure water level height, flow rate sensors are used to obtain water flow velocity, and temperature sensors monitor water temperature. The above parameters will affect the water level; The distribution of sensors should take into account the geographical features and water flow characteristics of the monitoring area, and should be focused on the bends of the river, the places where the width changes, and the key locations of the water inlet and outlet of the reservoir; Satellite remote sensing data relies on sensors carried on various satellites, including optical sensors and synthetic aperture radar SAR; model simulation data is obtained based on computer simulation software by inputting parameters of relevant terrain, climate and watershed characteristics.

[0019] The process of obtaining the water level and water level related data in step 1 includes: In addition to sensor data, Geographic Information System (GIS) data is also obtained, including information on topography, river direction, and watershed area; Meteorological data is also an important source of data. Data on rainfall, air humidity, and air pressure are obtained through the meteorological department. Meteorological conditions are closely related to water levels, and the amount of rainfall directly affects the rise and fall of river water levels.

[0020] The process of using statistical method mean filtering to remove noise from the collected data in step 2 to clean the data and perform normalization processing includes: For the various sensor data and other information source data collected, data cleaning is first performed to remove abnormal values ​​and erroneous data; in the ground water level meter data, thresholds are set and statistical methods are used to identify and eliminate obviously unreasonable values ​​due to equipment failures and external interference; corresponding cleaning methods are used for different types of data, and statistical methods-based mean filtering is used to remove noise for time series sensor data; Convert data from different sources and different magnitudes into a unified standard form; normalize the water level-related values ​​in satellite remote sensing data and the values ​​measured by ground water level gauges so that they are on the same order of magnitude.

[0021] The process of using satellite remote sensing to measure the water level in step 3 includes: First, obtain the optical satellite image data of the target water area and the transit orbit data of the altimeter satellite closest to the target water area; pre-process the optical satellite image, and then use the visible light and infrared bands in the multispectral image data to extract water body information; use the object-oriented method to first segment the image into different objects, and then analyze the spectral, shape, and texture characteristics of the above objects for classification, so as to distinguish the water body part from the non-water body part, and randomly select a number of satellite altimeter points as the initial measurement points of the current measurement according to the boundary range of the water body part; judge and eliminate the initial measurement points with abnormal imaging based on the principle of satellite altimeter, and the retained ones are the target measurement points; then it is necessary to set the elimination judgment criteria for the target measurement points, and further retain the compliant target measurement points as valid measurement points; The satellite-measured water level verification value and average value of this measurement are calculated for the valid measuring points, and the average value is used as the satellite-observed water level of this measurement; the satellite-observed water level time series and the measured water level time series of multiple measurements are obtained, and the average deviation of the two time series is used as the calibration parameter. By optimizing the calibration parameters, the water level is finally calculated and calibrated.

[0022] The process of using satellite images to predict water level changes in step 3 includes: Convolutional neural networks are used to analyze infrared cloud image data obtained by satellites, and combined with ground-measured precipitation records, a precipitation calculation model is established to estimate precipitation. A large amount of satellite image data provided by the meteorological department is used for training to learn to identify the characteristics of different precipitation levels. The model is established by testing the generalization ability of the model on data from different regions. The infrared cloud image data obtained by satellites is analyzed to predict precipitation. Prediction of precipitation is the premise of water level change analysis; a distributed hydrological model is used to simulate the flow and accumulation of precipitation in the basin; the model includes the influence of temperature, topography, soil type, and vegetation coverage on precipitation runoff; a river network model is constructed based on the topography and water system structure of the basin, and the parameters of different river sections are determined; preliminary settings are made based on empirical values, the predicted results of the model are compared with the measured data, and the model parameters are adjusted to make the predicted results as close as possible to the measured results; an automatic optimization algorithm is used to automatically search for the optimal parameter combination by setting the objective function to minimize the error between the model output and the actual observed value; The collected data is divided into a training set and a validation set; the training set is used to calibrate the model parameters, and the validation set is used to test the accuracy and reliability of the model; 80% of the data is used as the training set, and 20% of the data is used as the validation set; The root mean square error (RMSE) is selected to verify the model, and the root mean square of the sum of squares of the errors between the model prediction value and the actual observation value is calculated; the smaller the RMSE value, the higher the accuracy of the model prediction; the verification result analysis shows that the accuracy and reliability of the model meet the requirements, and the model is used for water level prediction; if the verification result is not ideal, the model parameters are readjusted and then verified again; The processed data is input into the constructed hydrological model, and the model is run for simulation calculations. During the operation, the model will calculate the hydrological process, evaporation, infiltration, and runoff in the basin based on the input meteorological data, terrain data, and soil data, and finally obtain the predicted results of the water level.

[0023] The process of fusing multi-source information in step 3 includes: Calculate the water level difference, which is obtained by subtracting the sensor collection value from the satellite remote sensing data value. When the water level difference is less than the threshold, the next step of data fusion is performed; when the water level difference is greater than the threshold, an alarm is issued and data is collected again; The weighted average method is used for data fusion; the weights of different data sources in the fusion are determined according to their reliability and accuracy factors; higher weights are given to ground monitoring data with higher accuracy; and the weights are relatively low for satellite remote sensing data with greater uncertainty; that is, the weights are determined based on the error analysis of the data, and the data sources with small errors have larger weights; higher weights are given to the data from the water level sensor because it directly measures the water level; and the data from the rain gauge are weighted according to the degree of correlation with the water level change to obtain the water level data after data fusion.

[0024] In step 4, the process of predicting water level changes through scenario analysis and combining real-time monitoring data to establish a water level risk early warning mechanism and release early warning information through multiple channels includes: When there are human factors affecting the water level, such as flood discharge and artificial rainfall, the detection personnel input them into the system. In step 3, when predicting the water level change, the human factors are comprehensively considered to obtain the data after human influence; After obtaining accurate water level data through data fusion, real-time monitoring is carried out; Establish a water level risk warning mechanism; set different levels of water level warning values, set the flood level as a, the actual water level data as b, when , the water level reaches 90% of the flood level, a Level 1 alarm is issued; , the water level reaches the flood level and a Level 2 alarm is issued; , the water level exceeds the flood level by 10%, and a level 3 alarm is issued; and according to step 3, the water level impact range corresponding to floods of different scales is analyzed; Early warning information is released through various means on the Internet, including text messages notifying relevant management personnel, setting up alarm devices near the monitoring area, and detecting management web pages.

[0025] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A water level monitoring and management method based on multi-source information fusion, characterized in that: The following steps are involved: Step 1: Use sensors to obtain water level and water level related data, and obtain information on surface characteristics and meteorological data through GIS data; Step 2: Use statistical methods to remove noise from the collected data by mean filtering, and then perform data cleaning and normalization. Step 3: Use convolutional neural network to analyze infrared cloud image data obtained by satellite to predict precipitation, use weighted average method to fuse multi-source information and predict water level through hydrological model; Step 4: Taking human influence factors into consideration, predict water level changes through scenario analysis, establish a water level risk warning mechanism based on real-time monitoring data, and release warning information through multiple channels.

2. The water level monitoring and management method based on multi-source information fusion according to claim 1 is characterized in that: The process of obtaining the water level and water level related data in step 1 includes: Ground monitoring data, satellite remote sensing data, and model simulation data; ground water level gauges provide accurate local water level values, satellite remote sensing data can cover a large area of ​​water, and model simulation data conducts predictive analysis of water levels based on historical data and related parameters; Reasonably arrange various types of sensors in the monitoring area, including water level sensors, flow rate sensors, and temperature sensors; water level sensors are used to directly measure water level height, flow rate sensors are used to obtain water flow velocity, and temperature sensors monitor water temperature. The above parameters will affect the water level; The distribution of sensors should take into account the geographical features and water flow characteristics of the monitoring area, and should be focused on the bends of the river, the places where the width changes, and the key locations of the water inlet and outlet of the reservoir; Satellite remote sensing data relies on sensors carried on various satellites, including optical sensors and synthetic aperture radar SAR; model simulation data is obtained based on computer simulation software by inputting parameters of relevant terrain, climate and watershed characteristics.

3. The water level monitoring and management method based on multi-source information fusion according to claim 2 is characterized in that: The process of obtaining the water level and water level related data in step 1 includes: In addition to sensor data, Geographic Information System (GIS) data is also obtained, including information on topography, river direction, and watershed area; Meteorological data is also an important source of data. Data on rainfall, air humidity, and air pressure are obtained through the meteorological department. Meteorological conditions are closely related to water levels, and the amount of rainfall directly affects the rise and fall of river water levels.

4. The water level monitoring and management method based on multi-source information fusion according to claim 3 is characterized in that: The process of using statistical method mean filtering to remove noise from the collected data and performing data cleaning and normalization processing in step 2 includes: For the various sensor data and other information source data collected, data cleaning is first performed to remove abnormal values ​​and erroneous data; in the ground water level meter data, thresholds are set and statistical methods are used to identify and eliminate obviously unreasonable values ​​due to equipment failures and external interference; corresponding cleaning methods are used for different types of data, and statistical methods-based mean filtering is used to remove noise for time series sensor data; Convert data from different sources and different magnitudes into a unified standard form; normalize the water level-related values ​​in satellite remote sensing data and the values ​​measured by ground water level gauges so that they are on the same order of magnitude.

5. The water level monitoring and management method based on multi-source information fusion according to claim 4 is characterized in that: The process of using satellite remote sensing to measure the water level in step 3 includes: First, obtain the optical satellite image data of the target water area and the transit orbit data of the altimeter satellite closest to the target water area; pre-process the optical satellite image, and then use the visible light and infrared bands in the multispectral image data to extract water body information; use the object-oriented method to first segment the image into different objects, and then analyze the spectral, shape, and texture characteristics of the above objects for classification, so as to distinguish the water body part from the non-water body part, and randomly select a number of satellite altimeter points as the initial measurement points of the current measurement according to the boundary range of the water body part; judge and eliminate the initial measurement points with abnormal imaging based on the principle of satellite altimeter, and the retained ones are the target measurement points; then it is necessary to set the elimination judgment criteria for the target measurement points, and further retain the compliant target measurement points as valid measurement points; The satellite-measured water level verification value and average value of this measurement are calculated for the valid measuring points, and the average value is used as the satellite-observed water level of this measurement; the satellite-observed water level time series and the measured water level time series of multiple measurements are obtained, and the average deviation of the two time series is used as the calibration parameter. By optimizing the calibration parameters, the water level is finally calculated and calibrated.

6. The water level monitoring and management method based on multi-source information fusion according to claim 5 is characterized in that: The process of using satellite images to predict water level changes in step 3 includes: Convolutional neural networks are used to analyze infrared cloud image data obtained by satellites, and combined with ground-measured precipitation records, a precipitation calculation model is established to estimate precipitation. A large amount of satellite image data provided by the meteorological department is used for training to learn to identify the characteristics of different precipitation levels. The model is established by testing the generalization ability of the model on data from different regions. The infrared cloud image data obtained by satellites is analyzed to predict precipitation. Prediction of precipitation is the premise of water level change analysis; a distributed hydrological model is used to simulate the flow and accumulation of precipitation in the basin; the model includes the influence of temperature, topography, soil type, and vegetation coverage on precipitation runoff; a river network model is constructed based on the topography and water system structure of the basin, and the parameters of different river sections are determined; preliminary settings are made based on empirical values, the predicted results of the model are compared with the measured data, and the model parameters are adjusted to make the predicted results as close as possible to the measured results; an automatic optimization algorithm is used to automatically search for the optimal parameter combination by setting the objective function to minimize the error between the model output and the actual observed value; The collected data is divided into a training set and a validation set; the training set is used to calibrate the model parameters, and the validation set is used to test the accuracy and reliability of the model; 80% of the data is used as the training set, and 20% of the data is used as the validation set; The root mean square error (RMSE) is selected to verify the model, and the root mean square of the sum of squares of the errors between the model prediction value and the actual observation value is calculated; the smaller the RMSE value, the higher the accuracy of the model prediction; the verification result analysis shows that the accuracy and reliability of the model meet the requirements, and the model is used for water level prediction; if the verification result is not ideal, the model parameters are readjusted and then verified again; The processed data is input into the constructed hydrological model, and the model is run for simulation calculations. During the operation, the model will calculate the hydrological process, evaporation, infiltration, and runoff in the basin based on the input meteorological data, terrain data, and soil data, and finally obtain the predicted results of the water level.

7. The water level monitoring and management method based on multi-source information fusion according to claim 6 is characterized in that: The process of multi-source information fusion in step 3 includes: Calculate the water level difference, which is obtained by subtracting the sensor collection value from the satellite remote sensing data value. When the water level difference is less than the threshold, the next step of data fusion is performed; when the water level difference is greater than the threshold, an alarm is issued and data is collected again; The weighted average method is used for data fusion; the weights of different data sources in the fusion are determined according to their reliability and accuracy factors; higher weights are given to ground monitoring data with higher accuracy; and the weights are relatively low for satellite remote sensing data with greater uncertainty; that is, the weights are determined based on the error analysis of the data, and the data sources with small errors have larger weights; higher weights are given to the data from the water level sensor because it directly measures the water level; and the data from the rain gauge are weighted according to the degree of correlation with the water level change to obtain the water level data after data fusion.

8. The water level monitoring and management method based on multi-source information fusion according to claim 7 is characterized in that: In step 4, taking into account human influence factors, the process of predicting water level changes through scenario analysis, combining real-time monitoring data, establishing a water level risk early warning mechanism, and issuing early warning information through multiple channels includes: When there are human factors, the detection personnel input data including the flow, time and downstream water level response of flood discharge, as well as the intensity, coverage area and runoff after artificial rainfall; Step 3: When predicting water level changes, according to different flood discharge and artificial rainfall scenarios, scenario analysis is carried out to obtain water level data after human influence; after obtaining accurate water level data through data fusion, real-time monitoring is carried out; Establish a water level risk warning mechanism; set different levels of water level warning values, set the flood level as a, the actual water level data as b, when , the water level reaches 90% of the flood level, a Level 1 alarm is issued; , the water level reaches the flood level and a Level 2 alarm is issued; , the water level exceeds the flood level by 10%, and a level 3 alarm is issued; and according to step 3, the water level impact range corresponding to floods of different scales is analyzed; Early warning information is released through various means on the Internet, including text messages notifying relevant management personnel, setting up alarm devices near the monitoring area, and detecting management web pages.

Images