Hydrology and water quality comprehensive monitoring system based on multi-source data fusion

Through the multi-source data fusion system, the hydrological and water quality monitoring data is integrated, and the problems of data contradictions and accuracy differences in the existing technology are solved, and more accurate and timely water quality monitoring and early warning are achieved, providing scientific support for water resource management.

CN119066610APending Publication Date: 2024-12-03CHINASOFT INTERNATIONAL SOFTWARE (SUZHOU) CO LTD
View PDF 0 Cites 24 Cited by

Patent Information

Application Number
CN202411107470.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The data sources in existing hydrological and water quality monitoring are diverse, with conflicts and differences in accuracy, affecting the accuracy of comprehensive monitoring.

Method used

Design a comprehensive monitoring system for hydrological and water quality based on multi-source data fusion, including data acquisition, preprocessing, quality control, fusion consolidation, feature extraction, model construction and monitoring and evaluation modules, through the integration of fixed monitoring stations, mobile devices, satellite remote sensing and The data of the weather station eliminates noise, calibrates data, integrates different data sources, extracts characteristic data and builds a water quality prediction model to conduct real-time monitoring and early warning.

Benefits of technology

Significantly improve the accuracy and reliability of water quality monitoring, provide more comprehensive and timely hydrological and water quality information, enhance the response speed to emergencies, reduce the impact of environmental and human activities, and evaluate the quantity and quality of water resources. situation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066610A_ABST
    Figure CN119066610A_ABST
Patent Text Reader

Abstract

The invention discloses a hydrology and water quality comprehensive monitoring system based on multi-source data fusion, and relates to the technical field of hydrology and water quality monitoring. The hydrology and water quality monitoring platform is in communication connection with a data acquisition module, a data preprocessing module, a data quality control module, a data fusion assimilation module, a feature extraction module, a model construction module, a monitoring evaluation module and a user interface and interaction module. According to the hydrology and water quality comprehensive monitoring system based on multi-source data fusion, the accuracy and reliability of water quality monitoring are remarkably improved by integrating data from a fixed monitoring station, mobile monitoring equipment, satellite remote sensing and a meteorological station, and the accuracy and reliability of water quality monitoring are remarkably improved by integrating monitoring data from different sources and different types, namely water quality indexes, hydrology parameters and meteorological data. The possible limitation of a single data source is avoided, so that the comprehensiveness and accuracy of monitoring data are greatly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of hydrology and water quality monitoring, and in particular to a hydrology and water quality comprehensive monitoring system based on multi-source data fusion. Background Art

[0002] Hydrological and water quality monitoring is an important means to ensure the security of water resources. By real-time monitoring of hydrological and water quality conditions, potential water pollution problems can be discovered in a timely manner to ensure the safety of drinking water sources and other water uses. Hydrological monitoring is also of great significance for preventing and mitigating water and drought disasters. By real-time monitoring of hydrological elements such as rainfall and water levels, the risks of natural disasters such as floods and droughts can be predicted and assessed, providing a scientific basis for disaster prevention and mitigation. In addition, water quality monitoring helps to understand the health of water ecosystems and provide data support for ecological protection. By monitoring indicators such as pollutant concentrations and dissolved oxygen content in water bodies, the stability and resilience of water ecosystems can be assessed.

[0003] In the existing technology, the data sources involved in hydrological and water quality monitoring are diverse, and there may be contradictions and conflicts in the hydrological and water quality monitoring data from different sources. In addition, there are certain differences in the accuracy and timeliness of the data, which will affect the accuracy of the comprehensive monitoring process. Therefore, how to effectively handle contradictory data and weaken the impact of data differentiation to improve the accuracy of the comprehensive monitoring process is the problem we need to solve. To this end, a comprehensive hydrological and water quality monitoring system based on multi-source data fusion is proposed. Summary of the invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a comprehensive hydrological and water quality monitoring system based on multi-source data fusion, including a hydrological and water quality monitoring platform, wherein the hydrological and water quality monitoring platform is communicatively connected with a data acquisition module, a data preprocessing module, a data quality control module, a data fusion and assimilation module, a feature extraction module, a model building module, a monitoring and evaluation module, and a user interface and interaction module, wherein the modules are connected by electrical signals;

[0005] The data acquisition module is used to collect raw hydrological and water quality data from a variety of data sources, including fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations;

[0006] The data preprocessing module is used to preprocess the collected raw hydrological and water quality data. The preprocessing process includes data cleaning, calibration and time synchronization processing to eliminate noise, errors, incompleteness and time differences in the data;

[0007] The data quality control module is used to evaluate data accuracy, perform data calibration, verification and error analysis on pre-processed multi-source data, and improve the credibility of monitoring results;

[0008] The data fusion and assimilation module is used to fuse the pre-processed multi-source data, integrate data from different sources and different types to generate a comprehensive monitoring data set, resolve conflicts between data, and weaken the impact of data differentiation;

[0009] The feature extraction module is used to extract feature data of water quality indicators, hydrological parameters and meteorological data from the fused and assimilated monitoring data to reduce the data dimension, wherein the water quality indicators include pH value, dissolved oxygen, algae density and total bacteria count; the hydrological parameters include water level, flow velocity and flow; and the meteorological data include rainfall, temperature and wind speed;

[0010] The model building module builds a water quality prediction and evaluation model based on the comprehensive monitoring data set and the feature data extracted from the fused and assimilated monitoring data, predicts the water quality trend, and provides support for decision-making;

[0011] The monitoring and evaluation module, based on the integrated monitoring data and the constructed water quality prediction and evaluation model, conducts real-time comprehensive monitoring and early warning of hydrology and water quality, including water quality status analysis, change trend prediction, potential risk prediction, and provides timely and accurate hydrology and water quality information;

[0012] The user interface and interaction module are used to provide a visual interactive interface to display monitoring results and prediction information, improve user experience, facilitate users to obtain required information, and enhance the usability and accessibility of the system.

[0013] Preferably, in the data acquisition module, the process of collecting the original hydrological and water quality data is as follows:

[0014] Deploy fixed monitoring stations in water areas and equip them with sensors to monitor hydrological parameters and water quality parameters. Hydrological parameters include water level, flow rate, and flow velocity, while water quality parameters include pH value, dissolved oxygen, and ammonia nitrogen content.

[0015] Using ships and drones to carry monitoring equipment to assist monitoring stations in conducting on-site measurements of hydrology and water quality, and using satellite images and remote sensing technology to obtain information on the distribution of water bodies and water level changes over a large area;

[0016] Obtain meteorological data related to hydrology and water quality from weather stations, including rainfall, wind speed, and temperature;

[0017] Aggregate the original hydrological and water quality data collected by fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations to obtain the original hydrological and water quality data set;

[0018] A data processing center is established to transmit the collected raw hydrological and water quality data to the data processing center through a wireless communication network.

[0019] Preferably, in the data preprocessing module, the process of preprocessing the original hydrological and water quality data is:

[0020] Extract raw hydrological and water quality data from the data processing center, clean the data, and identify and remove random noise in the data;

[0021] For the hydrological and water quality data after data cleaning, identify the outliers in the data set and check the missing values ​​in the data, and correct the missing data;

[0022] Add timestamps to all data points, align the time series for data collected at different times, and infer the timestamps for data missing timestamps based on the data collection frequency and context information. For data with inconsistent time intervals, interpolate and resample to reconstruct the time series and restore the time continuity of the data to ensure time consistency for time series analysis.

[0023] Preferably, in the data quality control module, the process of data accuracy assessment is:

[0024] The preprocessed multi-source data is converted into a unified format through mapping rules, and the preprocessed multi-source data is compared with known standard historical data to establish a calibration curve, and the data points are fitted using the least squares method to obtain a linear regression equation for data calibration;

[0025] By calculating the deviation of data points from the average value, data exploration is performed to identify the strengths and weaknesses of the data, analyze the errors and problems of the data, and use the Z-Score method to detect outliers. According to the deviation size and the set threshold, outliers in the data are detected and marked, and the source of errors is analyzed. Among them, strengths refer to high accuracy and consistency, and weaknesses refer to high deviations and outliers.

[0026] Use the root mean square error to perform error analysis and evaluate data errors to analyze the degree of data deviation, determine the availability of the data, and then analyze whether the data meets the accuracy requirements.

[0027] Preferably, the linear regression equation is expressed as:

[0028] y=mx+b;

[0029] Among them, y represents the dependent variable, which refers to the response value of the monitoring data; x represents the independent variable, which refers to the explanatory variable of the monitoring data; m represents the slope of the regression line, which refers to the expected change in y for every unit change in x; and b represents the intercept, which is the value of y when x = 0;

[0030] The expression for detecting outliers using the Z-Score method is:

[0031]

[0032] Among them, Z represents the Z-Score value, which refers to the number of standard deviations of the data point relative to the mean, X represents the observed value, μ represents the mean of the data set, and σ represents the standard deviation of the data set.

[0033] Preferably, the root mean square error is used for error analysis to evaluate the data error, and its expression is:

[0034]

[0035] Among them, R represents the root mean square error, which is used to measure the difference between the model prediction value and the actual observation value, and X i represents the i-th observation value, Y i Represents the i-th predicted value, n represents the total number of observations, and the smaller R is, the smaller the difference between the predicted value of the model and the actual observed value is, and the better the prediction performance of the model is.

[0036] Preferably, in the data fusion and assimilation module, the process of generating the comprehensive monitoring data set is:

[0037] Integrate pre-processed multi-source data and evaluate the similarities and differences between different data sources, where the similarities and differences between different data sources include data accuracy, sampling frequency, and time and space coverage;

[0038] Check the timestamps, spatial locations and data formats of different data sources so that data can be compared at the same temporal and spatial scales;

[0039] Use data assimilation technology to merge data from different sources into a consistent data set, thereby generating a comprehensive monitoring data set to ensure the continuity and integrity of the data set in time and space, and apply the comprehensive monitoring data set to actual monitoring tasks.

[0040] Preferably, in the feature extraction module, the feature data extraction process is:

[0041] Traverse the integrated and assimilated monitoring data in the comprehensive monitoring data set, analyze the characteristic indicators of water health status, which are water quality indicators, hydrological parameters and meteorological data, analyze the correlation of variables in the comprehensive monitoring data set, and identify the characteristic indicator data with high correlation;

[0042] For feature extraction of water quality indicators, the hydrogen ion concentration in the water body is measured by a pH meter to obtain the pH value. The dissolved oxygen sensor is used to measure and detect the dissolved oxygen content in the water body and record it. The spectral data of remote sensing data is used to extract the spectral features related to algae to reflect the density and species distribution of algae in the water body. The edge detection and texture analysis technology combined with the convolutional neural network are used to identify the algae species and density through the captured image or video data. The statistics of the algae species and density are calculated to reflect its overall characteristics. The laboratory PCR amplification technology is used to obtain the quantitative data of the total number of bacteria and specific bacterial groups, and the genetic information of bacteria is extracted using gene sequence analysis technology as feature data. The activity level and species information of bacteria are reflected by detecting their metabolites.

[0043] For the feature extraction of hydrological parameters, water level gauges are used to measure and record water level heights, and flow meters are used to measure the Doppler effect of sound waves reflected by water flow to calculate flow. Based on the observation results of water level and flow velocity, flow meters are used to measure flow.

[0044] For feature extraction of meteorological data, a rain gauge is used to record rainfall by collecting precipitation and converting it into electrical signals, a temperature sensor is used to record the temperature of the surrounding environment, and a wind speed sensor is used to calculate the wind speed using the ultrasonic principle. Anemometers are usually installed at weather stations or high places to obtain more accurate measurement results;

[0045] The extracted feature data were standardized to eliminate the impact of different dimensions and magnitudes, and the importance of each feature to water quality assessment was analyzed. The PCA dimensionality reduction technique was applied to reduce the number of features while retaining important information as much as possible.

[0046] Preferably, in the model building module, the construction process of the water quality prediction and evaluation model is:

[0047] Based on the comprehensive monitoring data set and the fused and assimilated monitoring data, feature data is extracted, and a data set containing the extracted features is constructed. The feature data set is divided into a training set and a test set. The training set is used for model training, and the test set is used for the final evaluation of model performance. The feature data set includes the associated data of water quality indicators, hydrological parameters, and meteorological data.

[0048] According to the needs of water quality prediction, a water quality prediction and evaluation model is constructed using the training set in combination with the linear regression algorithm;

[0049] Use the test set to evaluate the prediction accuracy of the model, calculate and compare different performance indicators to comprehensively evaluate the model performance and ensure that the model maintains good prediction capabilities even on unseen data;

[0050] Analyze the difference between the model prediction results and the actual values, identify potential sources of error, adjust the model, and deploy the trained model to the production environment.

[0051] Preferably, in the monitoring and evaluation module, the process of real-time comprehensive monitoring and early warning of hydrology and water quality is:

[0052] Receive hydrological and water quality data from various monitoring stations in real time, fuse and assimilate the received multi-source data, and perform pre-processing operations such as cleaning, denoising, and interpolation on the fused and assimilated data to eliminate the impact of outliers and missing values ​​on subsequent analysis;

[0053] Use sensors and automated monitoring equipment to monitor water quality in real time, obtain instant data on current water quality conditions, and conduct statistical analysis on real-time monitoring data, calculate the statistical quantities of water quality indicators, and evaluate whether the current water quality conditions meet relevant standards. The analysis results are intuitively displayed in the form of charts and reports, so that decision makers can quickly understand the current water quality conditions;

[0054] The constructed water quality prediction and evaluation model is used to analyze the real-time monitoring data, predict the changing trend of water quality indicators in the future, and analyze the prediction results to identify the changing direction and speed of water quality indicators and evaluate their impact on the water environment;

[0055] Combine historical data with real-time monitoring data to assess potential risks in water quality, including identifying possible sources of pollution and assessing the spread and speed of pollutants;

[0056] Based on the prediction results and the assessed potential risks, water quality warning information and risk warning information are released in a timely manner, and emergency plans and response measures are formulated.

[0057] The present invention provides a comprehensive hydrological and water quality monitoring system based on multi-source data fusion.

[0058] Beneficial effects:

[0059] 1. This comprehensive hydrological and water quality monitoring system based on multi-source data fusion significantly improves the accuracy and reliability of water quality monitoring by integrating data from fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations. By integrating monitoring data from different sources and different types, namely water quality indicators, hydrological parameters, and meteorological data, it avoids the limitations of a single data source, thereby greatly improving the comprehensiveness and accuracy of monitoring data. By integrating information from multiple data sources, it can more comprehensively reflect the actual situation of hydrology and water quality, providing a more reliable data basis for subsequent decision support.

[0060] 2. This comprehensive hydrological and water quality monitoring system based on multi-source data fusion improves the response speed to emergencies by monitoring the changing trends of water quality and hydrological conditions in real time and issuing early warnings in a timely manner when abnormalities or potential risks are found. It also formulates response measures in advance by comprehensively analyzing early warning information on water pollution disasters to reduce the impact on the environment and human activities. It uses comprehensive analysis of water quality, water volume, meteorological data to evaluate the quantity, quality and distribution of water resources, monitors and evaluates the effectiveness of water resource protection measures, and provides feedback for continuous optimization of management.

[0061] 3. This comprehensive hydrological and water quality monitoring system based on multi-source data fusion extracts key features from the fused and assimilated monitoring data, reduces data dimensions, and improves the efficiency of model training and prediction accuracy. The extracted feature data provides important input for the water quality prediction and assessment model, and can perform water quality status analysis, change trend prediction and potential risk prediction, providing timely and accurate hydrological and water quality information. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a module structure diagram of a hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to the present invention;

[0063] Figure 2 Flow chart of data accuracy assessment of the present invention. DETAILED DESCRIPTION

[0064] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0065] Examples, such as Figure 1 , Figure 2 As shown, the present invention provides a technical solution: a comprehensive hydrological and water quality monitoring system based on multi-source data fusion, including a hydrological and water quality monitoring platform, the hydrological and water quality monitoring platform is communicatively connected with a data acquisition module, a data preprocessing module, a data quality control module, a data fusion and assimilation module, a feature extraction module, a model building module, a monitoring and evaluation module, and a user interface and interaction module, wherein the modules are connected by electrical signals;

[0066] The data acquisition module is used to collect raw hydrological and water quality data from a variety of data sources, including fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations, to ensure the comprehensiveness and real-time nature of the data and provide a basis for subsequent data processing and analysis. Fixed monitoring stations are deployed in water areas and equipped with sensors to monitor hydrological parameters and water quality parameters. Hydrological parameters include water level, flow, and flow velocity, and water quality parameters include pH value, dissolved oxygen, and ammonia nitrogen content. Monitoring stations are deployed at river confluences, key areas of lakes, and key locations at estuaries. Hydrological parameter sensors include water level meters and flow meters. Pressure, ultrasonic, or radar water level meters are used to monitor water level changes in real time. Flow meters use electromagnetic induction, ultrasonic, or laser principles to measure flow. Water quality parameter sensors include pH meters, dissolved oxygen sensors, and ammonia nitrogen sensors. pH meters are used to measure the pH of water bodies, dissolved oxygen sensors are used to assess the content of dissolved oxygen in water bodies, and ammonia nitrogen sensors are used to detect ammonia nitrogen concentrations in water bodies to reflect the degree of water pollution. Monitoring equipment is carried out by ships and drones to assist monitoring stations in conducting hydrological and water quality monitoring. On-site measurement of water quality, and use satellite images and remote sensing technology to obtain information on water distribution and water level changes over a large area. In the process of ship monitoring, high-precision sensors and water quality sampler equipment are used to carry out depth profile monitoring and sampling analysis of water bodies along the preset route. UAV monitoring uses drones equipped with high-definition cameras and multi-spectral / infrared sensors to quickly obtain water surface images, temperature distribution, and algae growth information. Through high-resolution satellite images, macro views of water distribution and changes are regularly obtained to analyze the dynamic changes of water bodies. Remote sensing technology combines multi-spectral, radar and other remote sensing data to monitor parameters such as water surface temperature, turbidity, and chlorophyll content to further analyze water quality conditions; meteorological data related to hydrology and water quality are obtained from meteorological stations, including rainfall, wind speed, and temperature; the original hydrological and water quality data collected by fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations are aggregated to obtain original hydrological and water quality data sets; a data processing center is established to transmit the collected original hydrological and water quality data to the data processing center through a wireless communication network;

[0067] The data preprocessing module is used to preprocess the collected raw hydrological and water quality data. The preprocessing process includes data cleaning, calibration and time synchronization processing to eliminate noise, errors, incompleteness and time differences in the data, improve the accuracy and consistency of the data, and provide a high-quality data source for subsequent data fusion. The raw hydrological and water quality data are extracted from the data processing center, cleaned, and random noise in the data is identified and removed; for the cleaned hydrological and water quality data, outliers in the data set are identified and missing values ​​in the data are checked, and missing data are corrected; timestamps are added to all data points, and time series are aligned for data collected at different times. For data with missing timestamps, timestamps are inferred based on data collection frequency and context information. For data with inconsistent time intervals, interpolation and resampling are performed to reconstruct the time series and restore the time continuity of the data to ensure time consistency for time series analysis;

[0068] The data quality control module is used to evaluate data accuracy, perform data calibration, verification and error analysis on the preprocessed multi-source data, ensure the reliability and accuracy of the data set, improve the credibility of the monitoring results, convert the preprocessed multi-source data into a unified format through mapping rules, and compare the preprocessed multi-source data with known standard historical data to establish a calibration curve, use the least squares method to fit the data points, obtain the linear regression equation, and perform data calibration; perform data exploration by calculating the deviation between the data point and the average value, identify the strengths and weaknesses of the data, analyze data errors and problems, use the Z-Score method to detect outliers, detect and mark outliers in the data according to the deviation size and the set threshold, and analyze the source of errors, where strengths refer to high accuracy and consistency, and weaknesses refer to high deviations and outliers; use the root mean square error to perform error analysis, evaluate data errors, analyze the degree of data deviation, judge the availability of the data, and then analyze whether the data meets the accuracy requirements;

[0069] Furthermore, the linear regression equation is expressed as:

[0070] y=mx+b;

[0071] Among them, y represents the dependent variable, which refers to the response value of the monitoring data; x represents the independent variable, which refers to the explanatory variable of the monitoring data; m represents the slope of the regression line, which refers to the expected change in y for every unit change in x; b represents the intercept, that is, the value of y when x = 0. It should be noted that the purpose of the least squares method is to minimize the sum of the squares of the vertical distances from all data points to the fitting line. The range of values ​​of the formula depends on the data itself. The slope m and the intercept b can be any real number;

[0072] The expression for detecting outliers using the Z-Score method is:

[0073]

[0074] Among them, Z represents the Z-Score value, which refers to the number of standard deviations of the data point relative to the mean, X represents the observed value, μ represents the mean of the data set, and σ represents the standard deviation of the data set. It should be noted that the Z-Score method detects outliers by calculating the deviation of each data point from the mean. |Z|>3 is considered an outlier. The value range of Z-Score is from negative infinity to positive infinity. A positive value indicates that the data point is higher than the mean, and a negative value indicates that it is lower than the mean.

[0075] Furthermore, the root mean square error is used for error analysis to evaluate the data error, and its expression is:

[0076]

[0077] Among them, R represents the root mean square error, which is used to measure the difference between the model prediction value and the actual observation value, and X i represents the i-th observation value, Y i Represents the i-th predicted value, and n represents the total number of observed values. It should be noted that the smaller R is, the smaller the difference between the model's predicted value and the actual observed value is, and the better the model's prediction performance is. The value range of R is 0 to positive infinity, where 0 represents perfect prediction and a positive value represents the existence of prediction error.

[0078] The data fusion and assimilation module is used to fuse the pre-processed multi-source data, integrate data from different sources and different types to generate a comprehensive monitoring data set, resolve conflicts between data, weaken the impact of data differentiation, and improve the accuracy and reliability of monitoring results. It integrates the pre-processed multi-source data and evaluates the similarities and differences between different data sources, including data accuracy, sampling frequency, and time and space coverage; checks the timestamps, spatial locations, and data formats of different data sources to compare data at the same time and space scale; uses data assimilation technology to fuse data from different sources into a consistent data set, and then generates a comprehensive monitoring data set to ensure the continuity and integrity of the data set in time and space, and applies the comprehensive monitoring data set to actual monitoring tasks;

[0079] The feature extraction module is used to extract the characteristic data of water quality indicators, hydrological parameters and meteorological data from the fused and assimilated monitoring data, reduce the data dimension, and improve the model training efficiency and prediction accuracy. Among them, the water quality indicators include pH value, dissolved oxygen, algae density, and total bacteria count. The hydrological parameters include water level, flow velocity, and flow. The meteorological data include rainfall, temperature, and wind speed. The monitoring data after fusion and assimilation in the comprehensive monitoring data set are traversed to analyze the characteristic indicators of water health status, which are water quality indicators, hydrological parameters, and meteorological data. The variable correlation in the comprehensive monitoring data set is analyzed to identify the characteristic indicator data with high correlation to reduce data redundancy. For the feature extraction of water quality indicators, the hydrogen ion concentration in the water body is measured by a pH meter to obtain the pH value. The dissolved oxygen sensor is used for measurement to detect the dissolved oxygen content in the water body and record it. The spectral data of remote sensing data is used to extract the spectral features related to algae to reflect the density and species distribution of algae in the water body. The algae species and density are identified by using edge detection and texture analysis technology combined with convolutional neural network through the captured image or video data, and the algae species are calculated. The statistics of the total number of bacteria and density reflect its overall characteristics, and the quantitative data of the total number of bacteria and specific bacterial groups are obtained through laboratory PCR amplification technology, and the genetic information of bacteria is extracted by gene sequence analysis technology as characteristic data, and the activity and species information of bacteria are reflected by detecting the metabolites of bacteria; for the feature extraction of hydrological parameters, the water level is measured by a water level meter to record the water level height, and the Doppler effect of the sound wave reflected by the water flow is measured by a flow meter to calculate the flow, and the flow is measured based on the observation results of the water level and flow velocity in combination with the flow meter; for the feature extraction of meteorological data, a rain gauge is used to collect precipitation and convert it into an electrical signal to record the rainfall, a temperature sensor is used to record the temperature of the surrounding environment, and a wind speed sensor is used to calculate the wind speed using the ultrasonic principle. The anemometer is usually installed at a meteorological station or at a high place to obtain more accurate measurement results; the extracted feature data is standardized to eliminate the influence of different dimensions and magnitudes, and the importance of each feature to water quality assessment is analyzed, and the PCA dimensionality reduction technology is applied to reduce the number of features, while retaining important information as much as possible to improve the efficiency of model training;

[0080] Model building module, based on the comprehensive monitoring data set and the feature data extracted from the fused and assimilated monitoring data, builds a water quality prediction and evaluation model, predicts water quality trends, and provides support for decision-making. Based on the comprehensive monitoring data set and the fused and assimilated monitoring data, feature data is extracted, and a data set containing the extracted features is built. The feature data set is divided into a training set and a test set. The training set is used for model training, and the test set is used for the final evaluation of model performance. The feature data set includes the associated data of water quality indicators, hydrological parameters, and meteorological data. According to the needs of water quality prediction, the training set is combined with the linear regression algorithm to build a water quality prediction and evaluation model. The test set is used to evaluate the prediction accuracy of the model, and different performance indicators are calculated and compared to comprehensively evaluate the model performance to ensure that the model can maintain good prediction capabilities on unseen data. The difference between the model prediction results and the actual values ​​is analyzed, potential sources of error are identified, and the model is adjusted. The trained model is deployed to the production environment to ensure that the model can receive new monitoring data in real time and make predictions.

[0081] The monitoring and evaluation module conducts real-time comprehensive monitoring and early warning of hydrology and water quality based on the fused and assimilated monitoring data and the constructed water quality prediction and evaluation model, including water quality status analysis, trend prediction, and potential risk prediction, providing timely and accurate hydrology and water quality information, supporting decision-making and emergency response, receiving hydrology and water quality data from various monitoring sites in real time, and fusing and assimilating the received multi-source data to solve the problems of inconsistent data formats, time asynchrony, and spatial resolution mismatch, ensuring the accuracy and consistency of the data, and performing pre-processing operations such as cleaning, denoising, and interpolation on the fused and assimilated data to eliminate the impact of outliers and missing values ​​on subsequent analysis; using sensors and automated monitoring equipment to monitor water quality in real time and obtain real-time data on the current water quality status. The system can collect data from real-time monitoring data, calculate the statistics of water quality indicators, evaluate whether the current water quality status meets the relevant standards, and display the analysis results in the form of charts and reports, so that decision makers can quickly understand the current water quality status; use the constructed water quality prediction and evaluation model to analyze the real-time monitoring data, predict the changing trend of water quality indicators in the future, analyze the prediction results, identify the changing direction and speed of water quality indicators, and evaluate their impact on the water environment; combine historical data and real-time monitoring data to evaluate the potential risks in water quality, including identifying possible pollution sources and evaluating the diffusion range and speed of pollutants; according to the prediction results and the evaluated potential risks, timely release water quality warning information and risk warning information, and formulate emergency plans and response measures;

[0082] The user interface and interaction module is used to provide a visual interactive interface, display monitoring results and forecast information, improve user experience, facilitate users to obtain the required information, and enhance the usability and accessibility of the system. Through the collaborative work of the above modules, it can effectively handle contradictory data, weaken the impact of data differentiation, and improve the accuracy and reliability of the comprehensive monitoring process, providing more scientific and effective technical support for water resources management, ecological protection, disaster prevention and mitigation and other fields.

[0083] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A comprehensive hydrological and water quality monitoring system based on multi-source data fusion, including a hydrological and water quality monitoring platform, characterized in that: The hydrological and water quality monitoring platform is communicatively connected with a data acquisition module, a data preprocessing module, a data quality control module, a data fusion and assimilation module, a feature extraction module, a model building module, a monitoring and evaluation module, and a user interface and interaction module, wherein the modules are electrically connected with each other; The data acquisition module is used to collect raw hydrological and water quality data from a variety of data sources, including fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations; The data preprocessing module is used to preprocess the collected raw hydrological and water quality data; The data quality control module is used to evaluate data accuracy and perform data calibration, verification and error analysis on preprocessed multi-source data; The data fusion and assimilation module is used to fuse the pre-processed multi-source data and integrate data from different sources and different types to generate a comprehensive monitoring data set; The feature extraction module is used to extract feature data of water quality indicators, hydrological parameters and meteorological data from the fused and assimilated monitoring data, wherein the water quality indicators include pH value, dissolved oxygen, algae density and total bacteria count; the hydrological parameters include water level, flow velocity and flow; and the meteorological data include rainfall, temperature and wind speed; The model building module builds a water quality prediction and evaluation model based on the comprehensive monitoring data set and the feature data extracted from the fused and assimilated monitoring data to predict the water quality trend; The monitoring and evaluation module performs real-time comprehensive monitoring and early warning of hydrology and water quality based on the integrated monitoring data and the constructed water quality prediction and evaluation model; The user interface and interaction module are used to provide a visual interaction interface to display monitoring results and prediction information.

2. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 1 is characterized by: In the data acquisition module, the collection process of the original hydrological and water quality data is as follows: Deploy fixed monitoring stations in water areas and equip them with sensors to monitor hydrological parameters and water quality parameters. Hydrological parameters include water level, flow rate, and flow velocity, while water quality parameters include pH value, dissolved oxygen, and ammonia nitrogen content. Using ships and drones to carry monitoring equipment to assist monitoring stations in conducting on-site measurements of hydrology and water quality, and using satellite images and remote sensing technology to obtain information on the distribution of water bodies and water level changes over a large area; Obtain meteorological data related to hydrology and water quality from weather stations, including rainfall, wind speed, and temperature; Aggregate the original hydrological and water quality data collected by fixed monitoring stations, mobile monitoring equipment, satellite remote sensing, and meteorological stations to obtain the original hydrological and water quality data set; A data processing center is established to transmit the collected raw hydrological and water quality data to the data processing center through a wireless communication network.

3. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 2 is characterized by: In the data preprocessing module, the process of preprocessing the original hydrological and water quality data is as follows: Extract raw hydrological and water quality data from the data processing center, clean the data, and identify and remove random noise in the data; For the hydrological and water quality data after data cleaning, identify the outliers in the data set and check the missing values ​​in the data, and correct the missing data; Add timestamps to all data points, align the time series for data collected at different times, and infer the timestamps for data with missing timestamps based on the data collection frequency and context information. For data with inconsistent time intervals, interpolate and resample to reconstruct the time series and restore the temporal continuity of the data.

4. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 3 is characterized by: In the data quality control module, the process of data accuracy assessment is as follows: The preprocessed multi-source data is converted into a unified format through mapping rules, and the preprocessed multi-source data is compared with known standard historical data to establish a calibration curve, and the data points are fitted using the least squares method to obtain a linear regression equation for data calibration; By calculating the deviation of data points from the average value, we can explore the data, identify the strengths and weaknesses of the data, analyze the errors and problems of the data, detect outliers using the Z-Score method, detect and mark outliers in the data according to the deviation size and the set threshold, and analyze the source of the error; Use the root mean square error to perform error analysis and evaluate data errors to analyze the degree of data deviation, determine the availability of the data, and then analyze whether the data meets the accuracy requirements.

5. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 4 is characterized by: The linear regression equation is expressed as: y=mx+b; Among them, y represents the dependent variable, which refers to the response value of the monitoring data; x represents the independent variable, which refers to the explanatory variable of the monitoring data; m represents the slope of the regression line, which refers to the expected change in y for every unit change in x; and b represents the intercept, which is the value of y when x = 0; The expression for detecting outliers using the Z-Score method is: Among them, Z represents the Z-Score value, which refers to the number of standard deviations of the data point relative to the mean, X represents the observed value, μ represents the mean of the data set, and σ represents the standard deviation of the data set.

6. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 5 is characterized by: The root mean square error is used for error analysis to evaluate the data error, and its expression is: Among them, R represents the root mean square error, which is used to measure the difference between the model prediction value and the actual observation value, and X i represents the i-th observation value, Y i represents the i-th predicted value and n represents the total number of observations.

7. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 6 is characterized by: In the data fusion and assimilation module, the generation process of the comprehensive monitoring data set is as follows: Integrate pre-processed multi-source data and evaluate the similarities and differences between different data sources, where the similarities and differences between different data sources include data accuracy, sampling frequency, and time and space coverage; Check the timestamps, spatial locations and data formats of different data sources so that data can be compared at the same temporal and spatial scales; Using data assimilation technology, data from different sources are integrated into a consistent data set to generate a comprehensive monitoring data set, which is then applied to actual monitoring tasks.

8. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 7 is characterized by: In the feature extraction module, the feature data extraction process is as follows: Traverse the integrated and assimilated monitoring data in the comprehensive monitoring data set, analyze the characteristic indicators of water health status, which are water quality indicators, hydrological parameters and meteorological data, analyze the correlation of variables in the comprehensive monitoring data set, and identify the characteristic indicator data with high correlation; For the feature extraction of water quality indicators, the hydrogen ion concentration in the water body is measured by a pH meter to obtain the pH value, and the dissolved oxygen sensor is used to measure and detect the dissolved oxygen content in the water body and record it. The spectral data of remote sensing data is used to extract the spectral characteristics related to algae to reflect the density and species distribution of algae in the water body. The edge detection and texture analysis technology combined with the convolutional neural network are used through the captured image or video data to identify the algae species and density, and the statistics of the algae species and density are calculated to reflect its overall characteristics, and Through laboratory PCR amplification technology, quantitative data on the total number of bacteria and specific bacterial groups are obtained, and genetic information of bacteria is extracted using gene sequence analysis technology as characteristic data. The activity level and species information of bacteria are reflected by detecting their metabolites; For the feature extraction of hydrological parameters, water level gauges are used to measure and record water level heights, and flow meters are used to measure the Doppler effect of sound waves reflected by water flow to calculate flow. Based on the observation results of water level and flow velocity, flow meters are used to measure flow. For feature extraction of meteorological data, a rain gauge is used to record rainfall by collecting precipitation and converting it into electrical signals, a temperature sensor is used to record the temperature of the surrounding environment, and a wind speed sensor is used to calculate the wind speed using the ultrasonic principle; The extracted feature data were standardized, and the importance of each feature to water quality assessment was analyzed, and the PCA dimensionality reduction technique was applied to reduce the number of features.

9. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 8 is characterized by: In the model building module, the construction process of the water quality prediction and evaluation model is as follows: Based on the comprehensive monitoring data set and the fused and assimilated monitoring data, feature data are extracted, and a data set containing the extracted features is constructed, and the feature data set is divided into a training set and a test set, wherein the feature data set includes the associated data of water quality indicators, hydrological parameters and meteorological data; According to the needs of water quality prediction, a water quality prediction and evaluation model is constructed using the training set in combination with the linear regression algorithm; Use the test set to evaluate the prediction accuracy of the model and calculate and compare different performance metrics; Analyze the difference between the model prediction results and the actual values, identify potential sources of error, adjust the model, and deploy the trained model to the production environment.

10. The hydrological and water quality comprehensive monitoring system based on multi-source data fusion according to claim 9 is characterized by: In the monitoring and evaluation module, the process of real-time comprehensive monitoring and early warning of hydrology and water quality is as follows: Receive hydrological and water quality data from various monitoring stations in real time, fuse and assimilate the received multi-source data, and perform pre-processing operations such as cleaning, denoising, and interpolation on the fused and assimilated data; Use sensors and automated monitoring equipment to monitor water quality in real time, obtain instant data on current water quality conditions, and perform statistical analysis on real-time monitoring data, calculate the statistical quantities of water quality indicators, evaluate whether the current water quality conditions meet relevant standards, and intuitively display the analysis results in the form of charts and reports; The constructed water quality prediction and evaluation model is used to analyze the real-time monitoring data, predict the changing trend of water quality indicators in the future, and analyze the prediction results to identify the changing direction and speed of water quality indicators and evaluate their impact on the water environment; Combine historical data with real-time monitoring data to assess potential risks in water quality, including identifying possible sources of pollution and assessing the spread and speed of pollutants; Based on the prediction results and the assessed potential risks, water quality warning information and risk warning information are released in a timely manner, and emergency plans and response measures are formulated.

Citation Information

Cited By

  • System and method for analyzing water quality in water area based on safe dynamic evaluation

    CN119251625A

  • Ecological hydrological data processing method and system

    CN119273495A

  • Intelligent water quality monitoring system and method based on turbidity sensor

    CN119291147A

  • Employment distribution monitoring method and system based on multi-source data fusion

    CN119358851A

  • Water area ecological environment supervision system based on artificial intelligence

    CN119363796A