A method for determining intelligent river surface source pollution load based on GIS vector map conversion

Abstract

The present application relates to the technical field of source pollution load determination, and particularly relates to a smart river source pollution load determination method based on GIS vector conversion, comprising the following steps: obtaining river and lake water area shoreline data through GIS vector conversion, collecting water environment treatment data, water pollution prevention and control data, water ecological restoration data, water law enforcement supervision data and video monitoring data, and constructing a basic data information base; dividing the pollution area into several partitions, and preliminarily positioning the partitions where the pollution sources are located by using different data respectively; calculating the reliability scores based on different data positioning methods, constructing a pollution source positioning comprehensive score formula, and calculating the comprehensive reliability scores of different partitions; pushing the partition position with the highest comprehensive reliability score to relevant personnel by an information pushing management module, determining the final position of the pollution source, and calculating the total pollution load according to the type of the pollution source. The present application positions the pollution source by comprehensively using multiple methods, improves the positioning accuracy, and improves the accuracy of the source pollution load.

Description

Technical Field

[0001] This invention relates to the field of source pollution load determination technology, and in particular to a smart river non-point source pollution load determination method based on GIS vector map conversion. Background Technology

[0002] In the treatment of river pollution, accurately determining the source pollution load is the core of formulating an effective treatment strategy. However, due to the inaccurate location of pollution sources, the determination of source pollution load by existing technologies has a large deviation.

[0003] In the early stages, the location of pollution sources was determined by human experience, which was highly subjective and arbitrary with low accuracy. The pollution load determined based on this lacked scientific basis, resulting in a waste of treatment resources and poor treatment effects.

[0004] While location methods based on single data points have been applied after technological advancements, they suffer from significant problems. In the data collection phase, river and lake shoreline data are not updated in a timely manner, failing to reflect new land use changes promptly. This affects the identification of potential pollution sources, leading to biases in pollution load calculations and irrational resource allocation. Water quality monitoring data is also unsatisfactory, with uneven station distribution and long monitoring intervals, making it difficult to detect sudden pollution events promptly. Delayed and missed pollution source location results in pollution load calculations that fail to reflect the actual pollution situation.

[0005] At the analytical method level, single-data location models are simple, relying solely on water pollution indices to identify pollution sources. They cannot comprehensively consider multiple factors and are difficult to pinpoint accurately in complex pollution situations. Consequently, the determination of pollution load cannot fully consider the contribution of each pollution source, and the treatment strategy does not match the actual pollution.

[0006] Single technologies such as water ecological restoration, water administration law enforcement supervision, and video surveillance also have shortcomings. Biodiversity indicators are affected by natural changes, making it difficult to accurately locate pollution sources and affecting the assessment of pollution loads. Water administration law enforcement supervision data is lagging behind, failing to track changes in pollution sources in a timely manner, resulting in a lack of timeliness in determining pollution loads. Video surveillance is susceptible to environmental factors, with unstable identification capabilities, leading to inaccurate determination of pollution loads. These problems result in inaccurate determination of pollution loads, a lack of targeted remediation efforts, irrational resource allocation, and significant waste. Therefore, there is an urgent need for innovative integrated positioning technologies to correct the shortcomings of traditional methods in locating pollution sources, improve the accuracy of determining pollution loads, and promote the scientific and efficient management of river pollution. Summary of the Invention

[0007] This invention combines multiple positioning methods to improve the accuracy and efficiency of pollution source positioning, thereby improving the accuracy of source pollution load determination.

[0008] The technical solution proposed in this invention is: a method for determining non-point source pollution load in smart rivers based on GIS vector map conversion, the method comprising:

[0009] River and lake shoreline data are obtained through GIS vector conversion, and water environment management data, water pollution prevention and control data, water ecological restoration data, water administration law enforcement and supervision data, and video surveillance data are collected to build a basic data information database.

[0010] The polluted area is divided into several zones, and the pollution sources are initially located in the zones using river and lake shoreline data, water environment management data, water pollution prevention and control data, water ecological restoration data, water administration law enforcement and supervision data, and video surveillance data.

[0011] Calculate the reliability score based on different data positioning methods, determine the reliability weight of different positioning methods through the entropy weight method, construct the comprehensive score formula for pollution source positioning, and calculate the comprehensive reliability score for different zones;

[0012] The information push management module pushes the location of the partition with the highest comprehensive reliability score to relevant personnel, who then conduct inspections in that partition to determine the final location and type of pollution sources and calculate the total pollution load based on the pollution source type.

[0013] Verification monitoring points were set up in each zone. Model validation and optimization techniques were used to calculate the concentration error. When the error exceeded the threshold, parameters were adjusted and weights were optimized.

[0014] Preferably, the method for acquiring the river and lake shoreline data is as follows:

[0015] The process involves storing and processing river and lake maps using scanners and computers; determining the proportional relationship between the scanned river and lake maps and GIS vector images; converting raster data into vector data; storing the vector data; and finally, converting the vector data back into raster data.

[0016] Preferably, the construction process of the basic data information database is as follows:

[0017] Water quality monitoring stations are set up to collect routine water quality indicators, while also collecting meteorological and surrounding pollution source information; water pollution prevention and control data are collected according to the type of pollution source to construct a water pollution prevention and control database; water quality and quantity changes before and after the implementation of water ecological restoration projects, as well as the restoration of ecological functions at key ecological nodes, are monitored; law enforcement personnel monitor the overall condition of the river in real time, and the public uploads photos, videos, and related information on pollution problems they discover; different types of high-definition cameras are installed at key locations around rivers and lakes to collect video data; the above data are integrated to construct a basic data information database.

[0018] Preferably, the preliminary positioning process is as follows:

[0019] Divide the polluted area into Each zone is divided into several sections. Buffer zones are constructed using GIS buffer analysis technology. Spatial difference algorithms are used to determine pollution-related attribute values ​​and identify potential pollution source locations. Scoring formulas for river and lake shoreline factors are constructed to calculate scores. A comprehensive pollution index is calculated using the water pollution index method. Scoring formulas for water environment factors are set, and scores are calculated based on the comparison between the comprehensive pollution index and set thresholds. Pollution source types and locations are determined using material balance technology and actual removal volume. The QUAL2K model is used to predict water quality change trends and pollution source development dynamics. Scoring formulas for water pollution prevention and control factors are constructed to calculate scores.

[0020] Preferably, the preliminary positioning process further includes:

[0021] The Shannon-Wiener index is used to measure regional biodiversity. A scoring formula for water ecological restoration factors is constructed by integrating information on vegetation coverage and the presence or absence of specific sensitive species to calculate the score. The pollution frequency index is calculated using data statistical analysis technology. Video image analysis technology is used to count the number of frames of suspected pollution sources and the number of frames of actual confirmed pollution sources to calculate the identification accuracy. The video surveillance credibility coefficient is obtained by weighting the calculation based on equipment performance parameters and installation conditions.

[0022] Preferably, the comprehensive scoring formula for pollution source location is as follows:

[0023] ;

[0024] in: , , , , , , These represent the weights of each factor, with values ​​ranging from [0,1]. Scoring is given for factors related to river and lake shorelines. Scoring for water environment governance factors, Water pollution prevention and control factor scores Scoring for water ecological restoration factors, The pollution frequency index, To improve recognition accuracy, This represents the credibility coefficient of video surveillance.

[0025] Preferably, the specific calculation process for the total pollution load is as follows:

[0026] The daily emission load of the pollutant from the enterprise is calculated based on the enterprise's pollutant concentration and wastewater discharge volume; the daily pollution load of the agricultural pollutant is calculated based on the crop varieties, area, fertilizer and pesticide usage and frequency collected from farmland surveys, combined with water quality monitoring data of farmland irrigation water and drainage; the daily pollution load of the pollutant from domestic pollution sources is calculated based on the urban population, domestic sewage generation, sewage treatment plant capacity and operation status; the total pollution load calculation formula is shown below:

[0027] ;

[0028] in: For total pollution load, This represents the company's daily emission load for this pollutant; This represents the daily pollution load of this pollutant in agriculture; The daily pollution load of this pollutant from domestic pollution sources. For the first Family businesses.

[0029] Preferably, the specific process of parameter adjustment and weight optimization is as follows:

[0030] Multiple verification monitoring points are set up in each zone to collect actual pollutant concentrations regularly. The collected actual pollutant concentrations are compared with the calculated pollutant concentrations to calculate the concentration error rate. An error threshold is set. If the concentration error rate is greater than the error threshold, the parameters are adjusted. Using feedback data, the weights of each factor are re-determined using the analytic hierarchy process. If special influencing factors in certain areas are found to be not included in the model, corresponding influencing factors and calculation formulas are added to improve the model.

[0031] Preferably, the push process of the information push management module is as follows:

[0032] Information is categorized and organized according to message type. Different push channels or multiple channels are used simultaneously to push messages based on their urgency and nature. Relevant personnel set message receiving preferences and priorities according to their needs, and the information push management module automatically pushes important information according to the priority.

[0033] The present invention also provides a computer-readable storage medium storing a computer program, which is executed by a processor to implement the aforementioned method for determining non-point source pollution loads in smart rivers based on GIS vector map conversion.

[0034] The beneficial effects of this invention are:

[0035] This method comprehensively utilizes information from multiple sources, including river and lake shorelines and water environment management, to pinpoint pollution sources. River and lake shorelines, combined with land use and human activity intensity, identify potential pollution areas. Water quality monitoring data clarifies the degree of pollution, and multi-source data complement each other, overcoming the limitations of single-source location and accurately locating pollution sources. This provides an accurate basis for determining source pollution loads, avoiding calculation errors caused by inaccurate source location. Precise location allows for accurate acquisition of the location and type of each pollution source, making subsequent pollution load calculations more consistent with reality and improving the accuracy of source pollution load determination.

[0036] Multiple algorithms and models, such as spatial difference algorithm and water pollution index method, are employed for location calculation to assess the probability of pollution sources from different dimensions. The weights of each factor are determined through standardization and entropy weighting, constructing a comprehensive scoring formula. This multi-algorithm fusion and scientific weight determination method enhances the reliability of location calculation and provides a reliable basis for determining source pollution load. Accurate pollution source location enables pollution load calculation to comprehensively consider the contribution of each pollution source, ensuring the reliability of source pollution load calculation and allowing for a more rational allocation of treatment resources.

[0037] The positioning system integrates real-time monitoring data to achieve dynamic positioning. As the environment changes and the location and intensity of pollution sources shift, the system captures these changes in real time. It automatically issues warnings and notifies relevant departments when the overall score is abnormal or the area enters a high-risk level. This not only promptly curbs pollution spread but also makes the determination of pollution load more timely. When pollution sources change dynamically, the system can adjust the positioning results in a timely manner, allowing the pollution load to reflect the actual situation. Relevant departments can then quickly adjust their governance strategies, improve governance efficiency, and minimize the impact of pollution on river ecosystems. Attached Figure Description

[0038] Figure 1 This is a flowchart of a smart river non-point source pollution load determination method based on GIS vector map conversion according to the present invention;

[0039] Figure 2 This is a flowchart illustrating the pollution source location process of a smart river non-point source pollution load determination method based on GIS vector map conversion, according to the present invention. Detailed Implementation

[0040] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0041] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0042] like Figure 1 and Figure 2 As shown, this scheme collects information on existing river and lake shorelines, water environment management, water pollution prevention and control, water ecological restoration, water administration law enforcement supervision, and video surveillance to construct a basic data information database. A certain area near detected pollutants is designated as a polluted area (this can be determined based on the type of pollutant detected). The entire polluted area is divided into several zones. Information collected through these different methods is used to locate the pollution source, thus obtaining the location of the pollution source for each method. The reliability of each calculation method is calculated, and these reliability scores are combined to determine which zone the pollutant is most likely to be in. Finally, manual investigation is conducted in the area to determine the specific type and location of the pollution source. Then, based on the specific type and location of the pollution source, the source pollution load of the entire polluted area is determined. This comprehensive location method reduces the influence of other factors on a single location method, thereby effectively improving the accuracy of pollution source location and the accuracy of source pollution load calculation.

[0043] When using river and lake shoreline information for location analysis, the difficulty in obtaining shoreline data in some areas, especially remote or topographically complex regions, makes it challenging to comprehensively and accurately acquire detailed information about the surrounding area, such as land use types and human activities. This can lead to incomplete identification of potential pollution sources. The condition of river and lake shorelines changes over time, due to factors such as erosion, siltation, and development in surrounding areas. If data is not updated promptly, it cannot reflect the impact of these changes on pollution sources, potentially leading to inaccurate identification. Relying solely on shoreline information for pollution source identification often only allows for preliminary judgments, lacking in-depth analysis of pollution generation mechanisms, migration and transformation patterns, making it difficult to accurately assess the specific impact of pollution sources on water bodies.

[0044] When using water pollution prevention and control information for location analysis, measures are typically formulated based on known pollution source types and characteristics. However, for some new pollution sources or complex pollution situations, it may be difficult to accurately determine their type and source. For example, some emerging industrial production processes or the use of new pesticides and fertilizers may generate unknown pollutants, which existing prevention and control information cannot effectively identify. Even if the type of pollution source is known, the pollutant emission patterns of different enterprises may vary, influenced by factors such as production processes, production scale, and environmental protection measures. Without accurately grasping these emission patterns, it is difficult to monitor and identify pollution sources at the appropriate time and place, potentially leading to missed or incorrect identification. Over time, the emission situation of pollution sources may change, such as improvements in enterprise production processes or the strengthening or weakening of environmental protection measures. If water pollution prevention and control information lacks a dynamic updating mechanism, these changes cannot be reflected in a timely manner, thus affecting the accuracy of pollution source identification.

[0045] When using water ecological restoration information for location analysis, the relationship between ecological changes and pollution sources is not a simple one-to-one correspondence because the aquatic ecosystem is a complex whole. Ecological changes observed during water ecological restoration may be caused by multiple factors working together, and may not be directly and accurately traced to a specific pollution source. For example, eutrophication may be caused by a combination of factors, including agricultural non-point source pollution, domestic sewage discharge, and natural factors, making it difficult to accurately identify the main pollution source. The assessment of the effectiveness of water ecological restoration is also subject to uncertainty; different assessment methods and indicators may yield different results. This makes it difficult to accurately determine whether pollution sources have been effectively controlled or treated, and whether other potential pollution sources exist, when using water ecological restoration information to identify pollution sources. Water ecological restoration is a long-term process that requires continuous monitoring data to assess its effectiveness and identify pollution sources. However, many regions currently lack sufficiently long-term monitoring data, making it difficult to accurately grasp the long-term changes and trends of pollution sources, further increasing the difficulty of identifying pollution sources.

[0046] When using water environment management information for location analysis, incomplete or missing historical data may exist, affecting the accurate analysis of water pollution status and trends, and consequently making it difficult to accurately identify pollution sources. For example, early water quality monitoring data in some areas may not have been effectively preserved due to equipment malfunctions, personnel changes, or other reasons. The implementation of water environment management measures may have certain impacts on water quality and the ecological environment, thus interfering with the identification of pollution sources. For example, river dredging and ecological restoration projects may alter the physical and chemical properties of water bodies, changing their original pollution characteristics and increasing the difficulty of identifying pollution sources. Water environment management involves multiple aspects of data and indicators, such as water quality indicators, water quantity changes, and ecological indicators. These data are interconnected and complex, and accurately interpreting them and transforming them into effective information for identifying pollution sources is a challenge. Different professionals may have different interpretation methods, leading to uncertainty in the identification results.

[0047] When using water administration law enforcement and supervision information for location tracking, the limited resources and capabilities of these agencies make it difficult to comprehensively and in real-time monitor all rivers, lakes, and pollution sources. Some remote areas, small businesses, or hidden pollution sources may not be detected in a timely manner, leading to some sources being missed. Information sharing between water administration law enforcement and supervision departments and other relevant departments may be untimely or insufficient. For example, pollution problems discovered by environmental protection departments may not be promptly communicated to water administration law enforcement departments, causing delays in the identification and treatment of pollution sources. Differences in water administration law enforcement standards across different regions can lead to inconsistent results in the identification and treatment of pollution sources. Some regions may impose lenient penalties for certain polluting behaviors, encouraging businesses or individuals to continue illegally discharging pollutants, further complicating pollution source identification.

[0048] When using video surveillance for location tracking, the effectiveness is affected by environmental factors such as weather, lighting, and water turbidity. For example, in adverse weather conditions like heavy rain or fog, camera image quality deteriorates, making it difficult to clearly identify pollution sources; turbid water also hinders the observation and assessment of pollutants in the water. Video surveillance generates massive amounts of data, and effectively storing, managing, and analyzing this data is a challenge. Current technologies may not be able to quickly and accurately extract useful information from massive amounts of video data to identify pollution sources, resulting in low information utilization. Most video surveillance systems rely primarily on manual observation to identify pollution sources, lacking intelligent identification algorithms and models. Manual identification is not only inefficient but also prone to subjective errors, making it difficult to guarantee the accuracy and consistency of the identification results.

[0049] Therefore, by comprehensively considering the above methods for positioning, the impact of their respective defects can be effectively reduced, thereby improving the final positioning accuracy and the accuracy of the final calculation results when determining the source pollution.

[0050] When constructing buffer zones based on river and lake shorelines using GIS buffer analysis technology, the basic data, such as shoreline geographic coordinates, is obtained through GIS vector map conversion. Information such as river and lake maps is stored and processed using scanners and computers, converting the scanned river and lake map information (raster data) into vector data. Assuming the river and lake management area is defined as a polygon consisting of N sides connected by multiple inflection points, this polygon can be further subdivided into N-2 triangles. Using the area and side length ratios of similar triangles, the proportional relationship between the scanned river and lake map and the GIS vector map can be determined as follows: .in, This represents the ratio of the side lengths of two similar polygons. , These represent the areas of two similar polygons. In the actual process of determining the location of potential pollution sources, such as when analyzing a certain area... ( =1,2, , When obtaining land use type codes, and the quantitative value of human activity intensity In the analysis and processing of geographic data, including data from remote sensing, cameras, sensors, and their networks, the direct application of vector data is quite complex. Therefore, it is necessary to convert vector data into raster data to facilitate computational processing such as intersection analysis and overlay analysis. This conversion allows for more accurate information integration and helps in determining the location of potential pollution sources. When determining pollution-related attribute values ​​using spatial difference algorithms, the converted vector data can more accurately determine the location of sample points and the distance between the point to be estimated and the sample points.

[0051] In different areas of rivers, lakes, and reservoirs, water quality monitoring stations are scientifically and rationally deployed based on factors such as water flow characteristics, pollution source distribution, and water function, covering key locations such as the upper, middle, and lower reaches of rivers, different functional zones of lakes, and inlets and outlets. Advanced equipment such as multi-parameter water quality monitors and automatic samplers are used to conduct high-frequency monitoring of conventional indicators such as water temperature, pH, dissolved oxygen, chemical oxygen demand (COD), ammonia nitrogen, total phosphorus, and total nitrogen, as well as specific indicators such as heavy metals and organic pollutants. Some stations achieve real-time online monitoring to ensure timely understanding of water quality changes. In addition to water quality indicators, meteorological data, including air temperature, precipitation, wind speed and direction, as well as information on surrounding pollution sources, such as the location, discharge volume, and types of pollutants emitted by industrial enterprises, are collected simultaneously. Wireless transmission technology is used to transmit the monitoring data to a data center in real time, where it is visualized and analyzed using a Geographic Information System (GIS) to promptly detect water quality anomalies and provide a scientific basis for water environment governance decisions.

[0052] To address water pollution prevention and control, a multi-dimensional data collection approach is employed. For industrial pollution sources, in-depth investigations are conducted within enterprises to collect detailed information on production processes, raw material usage, and equipment operation. Automatic monitoring devices installed at discharge points acquire real-time data on wastewater discharge volume and major pollutant concentrations, and this data is connected to the environmental protection department's monitoring system to ensure accuracy and timeliness. For agricultural pollution sources, extensive farmland surveys are conducted to collect information on crop varieties, acreage, fertilizer and pesticide usage, and frequency of application. Simultaneously, the water quality of irrigation and drainage is monitored. For domestic pollution sources, collaboration with municipal departments and wastewater treatment plants yields data on urban population size, domestic sewage generation, wastewater treatment plant capacity, and operational status. Furthermore, drones equipped with high-resolution imaging equipment and spectral analyzers are used for rapid surveys of large water areas to identify the location and extent of suspected pollution sources. By integrating multi-source data, a comprehensive water pollution prevention and control database is constructed, providing strong support for analyzing pollution sources and assessing the effectiveness of prevention and control measures.

[0053] To comprehensively assess the effectiveness of aquatic ecological restoration, a comprehensive monitoring approach was employed for data collection. Regarding biodiversity, a combination of quadrat methods, transect methods, and drone monitoring was used to conduct regular surveys of the species, quantity, and distribution of flora and fauna in different ecological regions. DNA barcoding technology and a biodiversity monitoring system were utilized to accurately identify species and establish a database of species dynamic changes. For aquatic organisms, water samples were collected for analysis of plankton and benthic organisms to assess the health of the aquatic ecosystem. Regarding ecosystem structure and function, remote sensing technology and Geographic Information Systems (GIS) were used to obtain information such as vegetation cover and wetland area, combined with field measurements of soil texture and nutrient content, to analyze the material cycling and energy flow of the ecosystem. Simultaneously, changes in water quality and quantity before and after the implementation of aquatic ecological restoration projects were monitored, as well as the restoration of ecological functions at key ecological nodes, such as the connectivity of river ecological corridors and the purification capacity of wetlands, providing data support for optimizing aquatic ecological restoration plans.

[0054] Law enforcement officers use drones equipped with advanced equipment such as high-definition cameras, infrared sensors, and spectrometers to conduct aerial patrols of the river channels, monitoring their overall condition in real time. All data collected by law enforcement officers, including patrol records and reports, is first aggregated into a dedicated data management system. Within the system, data cleaning algorithms are used to remove duplicates and correct errors, ensuring the accuracy and completeness of the data. Data analysis tools are employed to deeply mine the water administration law enforcement supervision data. For example, by statistically analyzing the frequency and type distribution of pollution incidents in different regions, pollution hotspot maps are created, visually presenting areas with high pollution incidence; time-series data is analyzed to identify patterns in pollution incidents, such as seasonal trends. These analytical results will provide a scientific basis for developing targeted enforcement strategies. A public participation mechanism has been established, encouraging the public to participate in the collection of water administration law enforcement supervision data through mobile applications, online platforms, and other channels. The public can upload photos, videos, and related information about pollution problems they discover; this data, after review, is included in the water administration law enforcement supervision database. Simultaneously, the results of law enforcement actions are promptly communicated to the public, enhancing public trust and support for water administration law enforcement and fostering a positive atmosphere of public participation in river management.

[0055] Based on the geographical environment, water flow characteristics, and distribution of key monitoring areas of rivers and lakes, different types of high-definition cameras, including fixed cameras, pan-tilt cameras, and panoramic cameras, are rationally installed at key locations around rivers and lakes, such as major sewage outlets, pollution-prone areas, and near water conservancy facilities. According to monitoring needs, appropriate resolution, frame rate, and shooting angle are set for each camera to ensure comprehensive monitoring coverage and clear images. Geographic Information System (GIS) technology is used for precise positioning and labeling of cameras to facilitate rapid subsequent querying and management. Cameras continuously collect video data according to set time intervals or event triggering mechanisms. The collected video data is transmitted in real-time to the data center via wired networks (such as fiber optics) or wireless networks (such as 4G / 5G). During transmission, efficient video encoding technologies (such as H.265) are used to compress the data, reducing data transmission volume, improving transmission efficiency, and ensuring the real-time performance and smoothness of the video data. Simultaneously, a data backup mechanism is established to perform off-site backups of important video data to prevent data loss. In the data center, large-capacity storage devices (such as disk arrays) are used for long-term storage of video data. Distributed storage technology is adopted to improve the reliability and scalability of data storage. A comprehensive video data management system has been established to classify and store video data, indexing it according to information such as time, location, and monitoring area, facilitating quick querying and retrieval of historical video materials for users. Strict access permissions are set to ensure data security and confidentiality. Advanced intelligent video analysis algorithms are used to analyze the collected video data in real time. Object detection algorithms identify floating objects, vessels, and personnel in the water, tracking and analyzing their behavior. Image recognition technology is used to determine the presence of pollution events, such as sewage discharge and garbage dumping, and to issue timely alerts. Through deep learning of video data, the intelligent analysis model is continuously optimized to improve the accuracy and timeliness of event detection. Video surveillance data is integrated and analyzed with other monitoring data (such as water quality monitoring data and water level monitoring data) to provide more comprehensive decision-making basis for river management. For example, when video surveillance detects an anomaly in a certain area, it is combined with water quality monitoring data for that area to determine whether it has affected the water environment. Simultaneously, video surveillance data is applied to business scenarios such as law enforcement supervision and emergency management. Through video playback and evidence extraction, it provides strong support for water administration law enforcement, improving emergency response speed and processing efficiency.

[0056] To achieve precise pollution source location, the entire area was first divided into zones. The study area was divided into... A small area, denoted as The following is a detailed process for locating pollution sources based on multiple factors:

[0057] Preliminary location of pollution sources is determined based on river and lake shoreline data: Using GIS buffer analysis technology, buffer zones are constructed based on river and lake shorelines to identify potentially affected areas. Simultaneously, leveraging its powerful spatial analysis capabilities, geographic data such as land use type codes and quantified human activity intensity are integrated to aid in determining the location of potential pollution sources. For example, when determining the buffer zone range, an appropriate buffer distance is set based on the geographical coordinates of the shoreline, and the land use type and human activity intensity within this range are analyzed to identify potential pollution areas. For each sub-region... ( =1,2, , Obtain land use type code. (such as arable land) Industrial land (etc.) and quantitative values ​​of human activity intensity (Value range [0,10]). Pollution-related attribute values ​​are determined using spatial interpolation algorithms (such as inverse distance weighted interpolation). These algorithms are then used to calculate pollution-related attribute values ​​within the region. Based on the attribute values ​​of known sample points and the distance between the point to be estimated and the sample points, the pollution attribute value of the point to be estimated is calculated using formulas, thereby determining the pollution distribution within the region and providing data support for pollution source location. For example, during the calculation process, based on pollution data and distance information from different sample points, the pollution concentration and other attribute values ​​of unknown points can be accurately calculated. Let the point to be estimated be... In the region The formula for calculating its pollution attribute value is as follows:

[0058] ;

[0059] in: It is a region Internal estimation point Pollution attribute values ​​(such as pollution concentration, etc., used to determine the source of pollution); It is a region Known sample points The attribute value; It is the area to be estimated. Known sample points The distance; It is a region The number of known sample points within the area; This is the weighting coefficient, typically set to 2 (with a value generally between 1 and 3). Based on this, the scoring formula for the river and lake shoreline factor is set as follows:

[0060] ;

[0061] in: The number of pollution assessment indicators related to river and lake shorelines (such as land use type-related indicators, human activity intensity indicators, etc.); For the region The Middle The weights of each indicator are determined based on its importance in identifying pollution sources, and their values ​​range from [0,1]. ; For the standardized area The Middle Each indicator has a value ranging from [0,1]. The larger the value, the higher the potential pollution level reflected by the indicator.

[0062] Preliminary location of pollution sources is determined based on water environment governance data: Using water quality testing technology and various monitoring methods, measured concentrations of pollutants at different locations within the region are obtained. This data forms the basis for subsequent analysis; by monitoring various pollutant indicators in water bodies, the water environment quality is directly reflected. For example, professional water quality monitoring instruments are used to regularly collect water samples for analysis, obtaining data on pollutant concentrations such as chemical oxygen demand (COD) and ammonia nitrogen. In each region... Within the region, the water pollution index method is used to determine the pollution situation. Based on water quality monitoring data, a comprehensive pollution index is calculated using this method. By comparing measured concentrations with evaluation standards, it is determined whether the area is polluted, and the degree of pollution is judged based on the pollution index, thus providing a basis for locating pollution sources. For example, in the calculation process, the measured concentrations of various pollutants are substituted into the formula to obtain the comprehensive pollution index, which is used to assess the regional pollution status. The formula for calculating the comprehensive pollution index is:

[0063] ;

[0064] in: For the region The comprehensive pollution index; The number of evaluation indicators; For the region Inner Measured concentrations of various pollutants; For the region Inner Evaluation criteria for various pollutants. When Values ​​exceeding a certain threshold (set to) ,like A score of 1 (which can be adjusted according to local water quality standards and actual conditions) indicates that the area is polluted. The scoring formula for water environment governance factors is set as follows:

[0065] ;

[0066] Preliminary source localization of pollution based on water pollution control data: Taking organic pollution treated by biological methods as an example, theoretical removal rates are calculated using material balance techniques, based on the principles and design parameters of biological treatment. These calculations are then compared with actual removal rates obtained from monitoring to determine the type and location of the pollution source. In practical applications, detailed calculations of pollutant inputs and outputs during the biological treatment process are performed to analyze differences in removal rates and identify the pollution source. In regional... Taking the biological treatment of organic pollution as an example (other types of pollution are analyzed similarly), we determine the type and location of pollution sources.

[0067] ;

[0068] in: For the region The actual amount of pollutants removed is obtained through monitoring; For the region The theoretical removal capacity is calculated based on the principles of biological treatment and design parameters to determine the removal capacity of a specific organic pollutant. It is the threshold for judgment, and the value range is generally between 0.7 and 0.9.

[0069] Predicting regions using the QUAL2K model (simplified form) This model is used to predict regional water quality trends and pollution source development dynamics. By simulating changes in indicators such as biochemical oxygen demand (BOD) and dissolved oxygen (DO) over time, it analyzes the transformation and migration patterns of pollutants in water bodies, providing dynamic data support for pollution source location. For example, by inputting regional water flow, meteorological, and other conditions, it simulates water quality changes at different time points and predicts the diffusion paths and impact range of pollution sources. The specific calculation formula is as follows:

[0070] ;

[0071] ;

[0072] in: For the region Endogenous oxygen demand (BOD, mg / L); For the region Dissolved oxygen (DO, mg / L); For time ( ); For the region Internal BOD attenuation coefficient ( The value typically ranges from 0.1 to 0.5. between; For the region Internal reoxygenation coefficient ( The value is usually between 0.5 and 3. between; For the region Sources and sinks of internal BOD (mg / L / d); For the region Sources and sinks of internal DO (mg / L / d).

[0073] Based on the above analysis, the scoring formula for water pollution prevention and control factors is set as follows:

[0074] ;

[0075] in: The number of indicators used to assess the effectiveness of water pollution prevention and control and to locate pollution sources; For the first The weights of each indicator range from [0,1], and =1; For the region predicted by the model Inner The value of each indicator; For the region Inner The actual monitoring values ​​of each indicator.

[0076] Preliminary location of pollution sources based on water ecological restoration data: in the region Within this region, the Shannon-Wiener index is used to measure biodiversity. By calculating the Shannon-Wiener index, the biodiversity of the area is assessed to evaluate the health of the ecosystem. Changes in the biodiversity index can reflect whether the ecosystem has been disturbed, and thus infer the existence and potential location of pollution sources. For example, in the calculation process, the number of species and the percentage of individuals of each species in the area are statistically analyzed and substituted into the formula to calculate the biodiversity index, thereby determining the degree of impact of pollution on the ecosystem.

[0077] ;

[0078] in: For the region The Shannon-Wiener index; For the region Number of species within the organism; For the region Inner The proportion of individuals of each species to the total number of individuals. The value is lower than the threshold under normal ecological conditions in this area. This indicates that the ecosystem has been disturbed.

[0079] Assuming vegetation coverage The normal range is [ , Standardize it to :

[0080] ;

[0081] When specific sensitive species are present When it does not exist .

[0082] The formula for the comprehensive impact score of water ecological restoration factors is set as follows:

[0083] ;

[0084] in: , , These are the weights of the Shannon-Wiener index, vegetation cover, and presence of specific sensitive species, respectively, with values ​​ranging from [0,1]. .

[0085] Comparison of areas before and after ecological restoration Changes in indicators such as biodiversity index, ecosystem structure, and function. Assuming the biodiversity index before ecological restoration was... After repair, it is The expected recovery index is The quantitative indicators of ecosystem structure and function before restoration were After repair, it is The expected value is Set an evaluation coefficient for the effectiveness of ecological restoration. :

[0086] ;

[0087] in: and These represent the weights of changes in the biodiversity index and changes in ecosystem structure and function in the evaluation coefficient, respectively, with values ​​ranging from [0,1]. .when ( When the set effect evaluation threshold is reached (e.g., 0.6), for Make corrections:

[0088] ;

[0089] in: This is a correction factor, and its value ranges from [0,1].

[0090] Data such as vegetation coverage are standardized to ensure comparability. Simultaneously, information regarding the presence or absence of specific sensitive species is incorporated to comprehensively assess the impact of aquatic ecological restoration factors on pollution source location. For example, vegetation coverage data can be standardized according to its normal range, placing it within the [0,1] interval, facilitating comprehensive analysis with other indicators.

[0091] Preliminary location of pollution sources is determined based on water administration law enforcement and supervision data: Statistical analysis techniques are used to count the number of pollution incidents and total number of inspections within a region, calculating a pollution frequency index. Through statistical analysis of law enforcement and supervision data, areas with frequent pollution are identified, providing clues for locating pollution sources. For example, within a certain time period, the number of pollution incidents and inspections in different areas are counted to calculate the pollution frequency index for each area, identifying high-incidence pollution areas. Within the region... Internally, during the process of water administration law enforcement and supervision, the number of pollution incidents is counted. Total number of inspections Calculate the pollution frequency index :

[0092] ;

[0093] Preliminary pollution source localization based on video surveillance data: Using video image analysis technology, the number of frames showing suspected pollution sources and the number of frames actually confirmed as pollution sources are counted, the identification accuracy is calculated, and the video surveillance's ability to identify pollution sources is assessed. By analyzing video images, relevant pollution source information is extracted, providing intuitive evidence for localization. For example, image recognition algorithms are used to analyze water bodies and the surrounding environment in the video, identifying potential pollution sources and counting the number of frames. In the video surveillance system analysis, the statistical area... The number of frames appearing from suspected pollution sources inside And the number of frames of the actual confirmed pollution sources Calculate the recognition accuracy :

[0094] ;

[0095] A weighted calculation method is used to obtain the video surveillance reliability coefficient based on performance parameters such as device resolution, frame rate, coverage angle, and device distance, as well as installation conditions. By comprehensively considering these factors, the reliability of video surveillance data is evaluated, improving the accuracy of pollution source location. For example, during the calculation process, the weights of each parameter are set according to different scenario requirements to calculate the reliability coefficient, providing a quantitative indicator for the reliability of video surveillance data. The video surveillance reliability coefficient is obtained through weighted calculation based on device performance parameters and installation conditions.

[0096] ;

[0097] in: , , , These are, respectively, resolution weight, frame rate weight, coverage angle weight, and device distance weight, and These weights can be set according to actual needs, such as in scenarios where high detail is required. It can be set to 0.4; in scenarios where it is necessary to quickly capture dynamic images, It can be appropriately increased. For the region The resolution of the internal video surveillance equipment The highest resolution among all devices; For device frame rate, The highest frame rate; For the equipment coverage angle, For ideal full coverage angle (e.g., 360°); The distance from the equipment to the key monitoring area. The maximum permissible effective monitoring distance.

[0098] Data table transformation and entropy weighting were used to standardize the scores of each factor in each region, eliminating dimensional differences and making different factors comparable. Then, the entropy weighting method was used to determine the weight of each factor. Based on the magnitude of the information entropy of each factor, its importance in pollution source location was objectively reflected, and a comprehensive scoring formula was constructed to achieve precise location. For example, after standardizing the scores of each factor, the information entropy was calculated, the weight was determined based on the information entropy, and then substituted into the comprehensive scoring formula to calculate the comprehensive score of each region, thus determining the region where the pollution source is most likely to exist. For each region... Various factors Scoring is performed, and the data is standardized to obtain the original data matrix. .in This indicates the number of regions, corresponding to the division of the study area. this Sub-regions; This indicates the number of factors influencing the location of pollution sources, including factors such as river and lake shorelines, water environment management, water pollution prevention and control, water ecological restoration, water administration law enforcement and supervision, and video surveillance. Indicates the first Each region ( =1,2, , ) One factor ( =1,2, , The score, for example Representable region River and lake shoreline factor score , Indicates the area Water environment governance factor score wait, Indicates the area Water pollution prevention and control factor score , Indicates the region Water ecological restoration factor score wait, Indicates the area Water administration law enforcement supervision pollution frequency index wait, Indicates the area The accuracy of video surveillance recognition. Indicates the region The credibility coefficient of video surveillance.

[0099] Standardized data It is the raw data The result after standardization is shown in the formula. .in It is the first The minimum value of each factor in all regional scores. It is the first The maximum value of each factor's score across all regions. Standardization aims to eliminate dimensional differences between factor scores, making the factors comparable. For the original data matrix... Each column (i.e., each factor) in the formula The data is standardized to obtain a standardized data matrix. For example, for the river and lake shoreline factor (column 1), a standardized score is calculated for each region. This ensures that the scores for all factors are within the range of [0,1].

[0100] Reflects the first The distribution of scores for each factor across different regions is calculated using the following formula: , express In the region, the first The proportion of each factor.

[0101] For the Each factor, its information entropy Information entropy is used to measure the first... The degree of disorder of information of a factor across all regions. The smaller the information entropy, the greater the difference of the factor between regions, the more information it provides, and the greater its impact on the location of pollution sources.

[0102] No. The weights of each factor are given by the formula: The weight reflects the relative importance of this factor in the comprehensive assessment of pollution source location.

[0103] Taking all factors into account, the region The formula for the comprehensive score of pollution source location is:

[0104] ;

[0105] in: , , , , , , These are the weights corresponding to each factor, according to the formula. The value obtained is within the range [0,1], and By comparing different regions size, The higher the value, the greater the likelihood of the presence of pollution sources, thus enabling precise location of pollution sources.

[0106] To ensure the accuracy of the positioning results, the overall score of each region needs to be calculated. Verification was conducted. Multiple verification monitoring points were set up in each area, and water and soil samples were collected regularly, along with meteorological data. The actual pollutant concentrations obtained from the monitoring were then analyzed. Compared with pollutant concentrations predicted by the model Compare the results and calculate the concentration error rate. :

[0107] ;

[0108] like Exceeding the set error threshold (e.g., 15%, which can be adjusted according to actual monitoring accuracy requirements), indicating that the positioning results in this area may have deviations, and the model needs to be optimized.

[0109] Model validation and optimization techniques are used to collect actual data and compare it with model predictions by setting up validation monitoring points in various regions, calculating the concentration error rate. If the error exceeds a threshold, the model parameters are adjusted, weights are optimized, and the model is improved to continuously enhance positioning accuracy. For example, actual data such as water and soil samples are collected regularly and compared with the pollutant concentrations predicted by the model. Model parameters are adjusted and the model structure is optimized based on the error. The optimization process includes parameter adjustment, weight optimization, and model improvement. The parameters in the scoring formulas for each factor are re-evaluated. For example, the evaluation criteria in the water pollution index method... The data can be updated according to the latest environmental quality standards or the specific environmental needs of the region; if any omissions or errors are found in the species statistics during the calculation of the biodiversity index, the species count should be corrected in a timely manner. and the ratio of individuals of species Parameters such as these are used. The weights of each factor are re-determined using the analytic hierarchy process (AHP) or machine learning algorithms, based on feedback data. Taking machine learning as an example, historical validation data and corresponding comprehensive scores are used. As training samples, the algorithm learns the actual contribution of different factors to the accuracy of pollution source location, and then adjusts the weights accordingly. , This process makes the model more closely reflect reality. If certain areas are found to have special influencing factors that have not been included in the model, such as sudden geological activity leading to groundwater pollution, the corresponding influencing factors and calculation formulas need to be added to the model. For areas with complex terrain, a more refined geomorphological model can be introduced to optimize pollutant diffusion calculations.

[0110] Over time and with environmental changes, the location and intensity of pollution sources may alter. A positioning system based on GIS vector maps can access real-time monitoring data to achieve dynamic positioning. When the overall score of a certain area... Significant changes occur within a short period of time (e.g., the magnitude of the change exceeds a set threshold). When the score calculated from new monitoring data causes the area to enter a higher risk level, the system automatically triggers a real-time warning. Relevant departments are notified via SMS, app push notifications, etc., so that timely countermeasures can be taken to effectively curb the spread of pollution.

[0111] In addition to locating pollution sources in a single area, the system can also perform collaborative analysis of location results from multiple areas. For example, by analyzing the comprehensive pollution source scores and pollutant types in adjacent areas, it can determine whether there is cross-regional diffusion of pollution sources; and by combining factors such as water system connectivity and wind direction between areas, it can predict the transmission path of pollution. This provides decision support for joint governance between regions, enabling the formulation of unified pollution prevention and control plans, improving overall governance efficiency, and achieving collaborative protection and restoration of the ecological environment within the watershed. When relevant personnel receive zoning information indicating the most likely pollution sources, they can first investigate the locations of pollution sources within that zoning, located using different methods, and then investigate other locations, thereby improving the efficiency of pollution source location determination.

[0112] After determining the location of the final pollution source, the pollution load is calculated. Errors in source location will directly affect the accuracy of the pollution load calculation. When locating pollution sources based on river and lake shoreline data, if data errors or inaccurate analysis methods mistakenly identify a region as a pollution source when the actual source is elsewhere, the pollution load calculation for that region will include the non-existent pollution load, leading to an overestimation of the result. Conversely, omitting the true pollution source region will result in an underestimation of the pollution load. When dividing small areas for pollution source location, inaccurate boundary delineation may place some pollution sources in incorrect areas, affecting the accuracy of the pollution load calculation for each area. Misjudging the type of pollution source also affects the accuracy of pollution load determination. When using water pollution control data to determine the type of pollution source, if organic pollution is misclassified as other types of pollution, the calculation methods and parameters used in calculating the pollution load will be incorrect. When using biological treatment methods for organic pollution, the calculation methods for theoretical and actual removal capacities differ depending on the type of pollution. Misjudging the type can lead to incorrect pollution load calculations, failing to accurately reflect the actual pollution situation. Insufficient accuracy in identifying pollution sources results in inaccurate determination of source pollution loads, thus affecting the formulation and implementation of pollution control decisions. Underestimating the pollution load may result in insufficient control measures, failing to effectively improve environmental quality; while overestimating the pollution load may lead to a waste of treatment resources. Inaccurate determination of source pollution loads when formulating the treatment scale and processes for wastewater treatment plants may result in treatment facilities that cannot meet actual needs or excessive construction, affecting the scientific and economic efficiency of the treatment work.

[0113] Based on the law of conservation of mass, this study integrates data from various sources, including river and lake shorelines, water environment management, water pollution prevention and control, water ecological restoration, water administration law enforcement and supervision, and video surveillance, to comprehensively consider the input, output, and migration and transformation processes of pollution sources within the environment, thereby determining the pollution load. For industrial pollution sources, information collected within the enterprise regarding production processes, raw material usage, and equipment operation status is combined with data on wastewater discharge and major pollutant concentrations obtained from automatic monitoring equipment at discharge outlets. The pollution load is calculated based on the principle of mass balance. When determining the pollution load of a chemical enterprise, the amount of pollutants contained in the raw materials used in its production process is subtracted from the amount of pollutants contained in the products, and then the amount of pollutants generated and discharged into the environment during the production process is added to obtain the enterprise's pollution load.

[0114] The specific calculation process is as follows: For the region First, we need to compile relevant data on various pollution sources in the region. For industrial pollution sources, let's assume that enterprises... The concentration of a certain pollutant emitted is (mg / L), wastewater discharge volume is ( If so, then the daily emission load of the pollutant of the enterprise is... For agricultural pollution sources, the agricultural non-point source pollution load is calculated based on information collected from farmland surveys, including crop varieties, area, fertilizer and pesticide usage, and frequency of use, combined with water quality monitoring data for irrigation and drainage. Assume a certain type of fertilizer is used in a certain farmland, and the proportion of a certain pollutant in its effective component is... The usage amount is M (kg), and the concentration of this pollutant in farmland drainage is... (mg / L), drainage flow rate is ( Then, the daily pollution load of this pollutant from agricultural non-point sources is... For domestic pollution sources, calculations are based on data such as urban population, domestic sewage generation, sewage treatment plant capacity, and operational status. Assume the urban population is... The per capita domestic sewage generation is ( The removal rate of a certain pollutant by a wastewater treatment plant is... The concentration of this pollutant in domestic sewage is (mg / L), then the daily pollution load of this pollutant from domestic pollution sources. The sum of the pollution loads from all industrial, agricultural, and domestic pollution sources within the region yields the regional pollution load. Total pollution load of the pollutant .

[0115] An application support platform was constructed to manage data and information. Information resources, serving as the data support layer, constitute the information source and foundation of the regulatory platform. To avoid redundant construction and ensure effective sharing of data resources, the existing basic information databases established by relevant departments were fully utilized. Through the collection, integration, and improvement of existing information on river and lake shorelines, water environment management, water pollution prevention and control, water ecological restoration, water administration law enforcement supervision, and video surveillance, a practical, reliable, and advanced comprehensive database was constructed to meet the information service requirements of the information management system. Based on the needs of the regulatory platform construction and the actual conditions and characteristics of rivers and lakes, this project constructed basic, operational, monitoring, spatial, and unstructured databases. Given the large volume of data and the need for data exchange with various units, this study adopted MySQL database software, established an efficient data update mechanism, integrated information resources, and ensured data integrity and consistency. The construction content covered data resource planning, collection, compilation, entry, maintenance, and management, as well as database construction.

[0116] Establish a unified application support platform to provide a business development and operation support environment based on a unified technical architecture. This platform will provide a basic framework and underlying general services for upper-layer application construction, and an operational platform for data storage and integration, enabling information sharing. Specifically, based on the base map of the relevant unit's GIS platform, a customized river map service will be developed. A workflow engine will be used for the customization and execution of business processes such as business management. Through the workflow engine, processes can be easily and quickly formulated for relevant businesses. The workflow engine helps users adapt to the changing needs of processes and maintains maintainability and low cost when processes change.

[0117] The middleware provides efficient and flexible message synchronization, as well as asynchronous transmission processing, store-and-forward, and reliable transmission capabilities, ensuring secure, reliable, and efficient message delivery in complex network and application system environments. During data exchange, different forms of data are sent or uploaded to the target system via WebService and REST interface calls, thereby enabling data exchange between heterogeneous systems across regions, platforms, and security authentication methods. A data exchange architecture is established from both technical and business perspectives, realizing the service-oriented architecture of IT capabilities and laying a solid foundation for the establishment of an SOA architecture.

[0118] The application support platform is built on a web platform, with mobile app and WeChat official account access, providing corresponding services for inspections. The web platform is based on a backend database, ensuring a single data source and efficient data exchange.

[0119] The main functions of the system are as follows:

[0120] Basic Parameter Profile: This module is used to input and manage basic project data. Users can enter basic project information through this module, including project overview, location, and responsible person. This information provides necessary data support for daily business management. The data entry process supports multiple formats and batch import, ensuring the accuracy and efficiency of data entry.

[0121] Attendance Management: The attendance management module records the attendance of cleaning vessels using electronic attendance technology. This module utilizes GPS positioning and clock-in / out recording functions to meticulously record the time and location of vessels entering and leaving each section of the river. This method ensures the accuracy of attendance data and provides managers with reliable data. The system automatically calculates the working hours of the cleaning vessels and exports the data to the required format according to the client's needs. Through detailed attendance data recording, managers can track and analyze the operational status of the cleaning vessels, promptly identifying and resolving attendance issues. The system also supports attendance anomaly handling functions to help managers address attendance exceptions, ensuring the accuracy and completeness of attendance records.

[0122] Patrol Management: The patrol management module records and saves patrol routes and information. Through GPS positioning and mobile device upload capabilities, the system can update patrol records in real time, ensuring the accuracy and completeness of the data. This module supports live drone patrols, further improving coverage and efficiency. Drones can transmit patrol footage in real time, helping managers better understand the situation on-site. Problems discovered during patrols can be immediately reported through the system, forming a closed-loop management system. This feature ensures that problems are detected and resolved promptly, improving the effectiveness and response speed of patrol management. The system also supports historical querying of patrol records, helping managers analyze patrol data and optimize patrol strategies and workflows.

[0123] Cleaning Turnover Management: This module optimizes the assignment of cleaning tasks and the scheduling of cleaning vessels. Through this module, managers can view the real-time progress of cleaning tasks and allocate personnel and resources as needed.

[0124] Track Query Management: The track query management module records and queries the operational tracks of cleaning vessels. Through this module, users can view the historical operational tracks of cleaning vessels at any time to understand their work status and routes. The system provides detailed track recording and query functions, and users can generate various reports and statistical data as needed. By analyzing the track data, managers can identify potential problems in cleaning work, optimize operational routes and workflows, and improve work efficiency and effectiveness.

[0125] Performance Indicator Management: This module is used for assessing and evaluating daily work performance. The system digitizes performance indicators and evaluates them in conjunction with actual work conditions. This module supports the automatic generation of assessment reports and performance evaluation results based on different assessment standards and requirements, ensuring fairness and transparency in the assessment process. The system provides flexible assessment standard setting functions, allowing managers to define performance indicators according to actual needs. Assessment results can generate detailed reports and charts, and the module supports historical record and query functions for assessment results, facilitating long-term performance tracking and analysis.

[0126] Carbon Degradation Management: The carbon degradation management module calculates the carbon conversion of waste collection and displays the results in charts. This module processes the weighing data of collected waste to calculate the corresponding carbon conversion, thereby assessing the environmental benefits of the sanitation work. The system provides detailed carbon conversion calculation and display functions. Through the analysis of carbon conversion, managers can assess the environmental impact of sanitation work, formulate corresponding environmental protection strategies, and further improve the environmental benefits of the work.

[0127] Salvaged Material Weighing Management: The salvaged material weighing management module records the weight of salvaged waste in real time using weighing equipment on board and uploads the data to the system. The system supports real-time monitoring and recording of weighing data, enabling managers to understand the waste salvage situation at any time. Through analysis of the weighing data, the system can generate detailed reports and statistical data to assist managers in waste disposal and resource management.

[0128] Information Push Management: The information push management module is used to promptly push important messages and reminders from the system to users. This includes work notifications, early warning information, etc., ensuring that all relevant personnel can receive system messages quickly. The system provides multiple push methods, including SMS, email, and mobile notifications, ensuring that information can be delivered to all relevant personnel in a timely manner. Users can set message receiving preferences according to their needs, and the system will automatically push important information based on priority.

[0129] Real-time Sensing Monitoring: The real-time sensing monitoring module uses cameras installed at key monitoring points to monitor the accumulation of garbage in real time. This module not only supports the integration of newly installed cameras but also the consolidation of existing cameras, ensuring comprehensive coverage of the monitoring system. The system transmits monitoring footage in real time, allowing users to view the on-site situation of garbage accumulation at any time. Through the analysis of video data, managers can promptly identify garbage accumulation problems and take necessary measures to address them.

[0130] Data Analysis and Statistics: The data analysis and statistics module is used to summarize and analyze the entered cleaning data. This module supports data statistics by day, month, quarter, year, or user-selected dates, and exports reports and charts according to required formats. By organizing and analyzing historical data, the system generates various statistical reports and charts, providing strong support for management decisions. Users can use data analysis to understand the trends and status of cleaning work, identify potential problems, and make adjustments.

[0131] Attendance Monitoring and Early Warning: The absence warning function in the attendance monitoring and early warning module is designed to ensure the accuracy of attendance records for cleaning boats. When the system detects that a cleaning boat has not clocked in within the stipulated time, it will automatically trigger an early warning mechanism and immediately send a notification to the relevant management personnel. Through automatic early warning, the system can promptly remind relevant personnel to make up for the missed clock-in, avoiding omissions and incomplete attendance records. Simultaneously, the absence warning function also has detailed recording and report generation capabilities, allowing management personnel to easily query historical attendance data and conduct comprehensive attendance statistical analysis.

[0132] Waste Volume Statistics Management: The waste volume statistics management module accurately records and calculates the amount of waste collected by cleaning vessels, generating detailed daily, monthly, and yearly statistical reports. Utilizing advanced data acquisition and processing technology, the waste volume statistics function ensures the accuracy of waste volume records for each vessel. By summarizing and analyzing this data, the system can intuitively display trends and changes in waste collection, providing important references for management decisions. In practical applications, the waste volume statistics module also supports multi-dimensional data display and comparative analysis. Users can easily view waste collection volumes for different time periods and areas through the system interface, identifying high-incidence areas and peak periods for waste collection, thereby allowing for targeted adjustments to cleaning strategies and resource allocation. Furthermore, the statistical reports generated by the system can be exported to various formats, facilitating data sharing and communication among management personnel.

[0133] Excess Waste Warning Statistics: The excess waste warning statistics module is dedicated to real-time monitoring of waste collection volume. When the collection volume exceeds a preset threshold, the system will immediately issue a warning notification. This function helps managers understand the waste excess situation in a timely manner, quickly take necessary measures to deal with the excess waste, and ensure the smooth progress of waste disposal. By setting reasonable thresholds, the excess waste warning statistics module can effectively prevent waste accumulation and processing delays, ensuring the continuity and efficiency of cleaning work. The waste excess warning function also supports detailed statistics and analysis of warning records. Managers can view historical excess warning data, analyze fluctuations and trends in waste collection volume, and optimize warning thresholds and response strategies. The system can also generate excess warning reports, providing intuitive data displays and multi-dimensional comparative analysis to help managers better understand and address the challenges in waste disposal. Through the excess waste warning statistics module, users can achieve accurate monitoring and efficient management of waste collection volume, improving the overall quality and efficiency of cleaning work.

[0134] Departure from Track Warning: The departure from track warning module monitors the cleaning vessels' trajectories in real time to ensure they operate according to designated routes and times. When the system detects that a cleaning vessel has been docked at a certain location for an extended period exceeding a set threshold or has deviated from its designated track, it automatically issues a warning notification and generates a corresponding report. The departure from track warning module also features detailed trajectory recording and analysis functions. Managers can view the actual travel trajectory of each vessel through the system, identifying potential deviations and abnormal docking situations. The trajectory reports generated by the system provide detailed data and charts, helping users comprehensively understand the vessel's travel status and operational efficiency.

[0135] Waste Disposal Monitoring: The waste disposal monitoring module uses cameras installed at the waste disposal site to monitor the entire waste disposal process in real time. Users can view the on-site situation through the system interface, ensuring that the process meets environmental standards and allowing for timely detection and resolution of problems. This function provides comprehensive supervision of the waste disposal process through efficient video surveillance technology, ensuring transparency and standardization. The waste disposal monitoring module also supports video playback and event marking functions. Managers can easily view historical recordings, analyze problems during the process, and identify areas for improvement. The system also provides event marking and alarm functions; when abnormal processing behavior or equipment malfunction is detected, it automatically issues early warning notifications to help users take timely measures to prevent environmental pollution and resource waste.

[0136] Closed-Loop Status Alert: The closed-loop status alert module monitors each step of the cleaning process to ensure its smooth completion. When the system detects a problem or failure to complete a step according to regulations, it automatically issues an alert to help users identify and resolve issues promptly. The module also supports detailed process recording and analysis. Users can view the completion status and problem records of each process step through the system interface, identifying potential bottlenecks and areas for improvement. The closed-loop alert report generated by the system provides detailed data and charts, helping users fully understand the process's execution status and efficiency.

[0137] Message Push Alerts: The message push alert module promptly delivers alert messages from the system to relevant personnel, including alerts for attendance anomalies, excessive waste, and deviations from designated tracking. Real-time message pushes ensure all important information is quickly disseminated, improving response speed and processing efficiency, and preventing management problems caused by information delays. This function utilizes multiple information push channels, such as SMS, email, and system notifications, to ensure users receive alert information promptly. The message push alert module also supports message logging and analysis. Users can view historical alert messages through the system interface, analyze the frequency and causes of alerts, and optimize alert strategies and response measures. The system-generated alert message reports provide detailed data and charts, helping users fully understand the execution status and effectiveness of alerts.

[0138] For example, based on the geographical features of the watershed, the direction of water flow, and the distribution of functional zones, the entire watershed is divided into 100 smaller regions, each labeled as follows: , ,..., This lays the foundation for accurate positioning in the future.

[0139] Using GIS buffer analysis techniques, buffer zones are constructed based on the shoreline. In the region... The land use types in this area include large industrial parks and densely populated residential areas. The industrial land codes correspond to a high potential pollution level, and the quantified value of human activity intensity in this area reaches 8 (range [0,10]). Assume there are 3 known sample points in this area used for spatial difference algorithms to calculate pollution attribute values. The known pollution attribute values ​​of the sample points... They are respectively =50、 =60、 =70, the distance from the point to be estimated to each sample point. They are respectively =2、 =3、 =4, weighting coefficient =2. Calculate the contamination attribute value of the point to be estimated using the spatial interpolation algorithm formula. :

[0140] ;

[0141] Assuming that the pollution assessment indicators related to river and lake shorelines include two indicators: land use type and intensity of human activities ( =2), weight of land use type index =0.6, weight of human activity intensity index =0.4, the standardized land use type index value =0.8, Human Activity Intensity Index =0.9. According to the scoring formula for river and lake shoreline factors, the result is... Regional river and lake shoreline factor score for:

[0142] ;

[0143] exist In a certain region, the measured concentration of chemical oxygen demand was obtained from a monitoring session. =50 Measured concentration of ammonia nitrogen COD evaluation standards for this region , Assuming the number of evaluation indicators =2. The comprehensive pollution index is calculated using the water pollution index method. :

[0144] ;

[0145] Set threshold ,because According to the scoring formula for water environment governance factors:

[0146] ;

[0147] achievable .

[0148] Using biological treatment methods to treat organic pollution The actual removal volume was obtained through regional monitoring. The theoretical removal capacity is calculated based on the principles and design parameters of biological treatment. Judgment threshold .because Furthermore, the pollution characteristics are consistent with those of organic pollution, therefore the pollutant type in this area is organic pollution. Assume that the indicators used to assess the effectiveness of water pollution control and the location of pollution sources include two indicators: the ratio of actual removal to theoretical removal and the change in BOD predicted by the QUAL2K model. The weight of the ratio of actual removal amount to theoretical removal amount Weights of BOD change indicators predicted by the QUAL2K model The change in BOD within the region was predicted using the QUAL2K model. Actual monitoring of BOD changes According to the scoring formula for water pollution prevention factors, we can obtain:

[0149] ;

[0150] In the region Within the area, the number of species was statistically obtained through quadratic methods, transect methods, and drone monitoring. And calculate the percentage of individuals for each species. Then calculate the Shannon-Wiener index. : ;

[0151] At the same time, obtain vegetation coverage. and standardize it to Indicators for determining the presence or absence of specific sensitive species (Sensitive species exist). Based on the comprehensive impact score formula for water ecosystem restoration factors, we can obtain:

[0152] ;

[0153] If a water ecological restoration project has been carried out in the area, the changes in indicators such as biodiversity index, ecosystem structure and function before and after the restoration should be compared to calculate the ecological restoration effectiveness evaluation coefficient. (greater than the set threshold) Then there is no need to... Make corrections.

[0154] Over a period of time, statistical areas Number of pollution incidents Number of patrols Next, calculate the pollution frequency index according to the formula. The higher the index, the greater the likelihood of frequent pollution in the area. According to the scoring formula for water administration law enforcement and supervision factors, we can obtain:

[0155] ;

[0156] In the region In video surveillance, the number of frames showing suspected pollution sources was counted. Frames, the actual number of confirmed pollution source frames. Frames, calculate recognition accuracy Meanwhile, the overall score of the parameters of the cameras installed in this area was also considered. The score is 100 points. According to the video surveillance credibility coefficient formula, we can get:

[0157] ;

[0158] Based on the video surveillance factor scoring formula (assuming a total number of frames) ), we can obtain:

[0159] ;

[0160] Data table transformation and entropy weighting were used to standardize the scores of each factor in each region. Assuming standardization has been completed and standardized scores for each factor in each region are obtained, the scores are then calculated based on the region... , , , , , For example, let's assume the standardized scores are as follows:

[0161] , , , , , ;

[0162] , , , , , ;

[0163] , , , , , ;

[0164] , , , , , ;

[0165] , , , , , ;

[0166] , , , , , ;

[0167] , , , , , ;

[0168] Calculate the overall score for each region:

[0169] ;

[0170] ;

[0171] ;

[0172] ;

[0173] ;

[0174] ;

[0175] By comparing the overall scores of each region, the region The area with the highest overall score (0.725) was therefore identified as the most likely source of pollution. Further investigation revealed a chemical plant in the area that was illegally discharging industrial wastewater, and an agricultural plot with excessive pesticide contamination, consistent with the calculated results.

[0176] The process described in the flowchart above can be implemented as a computer software program. Embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wire segments, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination thereof.

[0177] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0178] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the stated principles, the implementation of the present invention may have any variations or modifications.

Claims

1. A method for determining non-point source pollution load in smart rivers based on GIS vector map conversion, characterized in that, The method includes: River and lake shoreline data are obtained through GIS vector conversion, and water environment management data, water pollution prevention and control data, water ecological restoration data, water administration law enforcement and supervision data, and video surveillance data are collected to build a basic data information database. The polluted area is divided into several zones, and the pollution sources are initially located in the zones using river and lake shoreline data, water environment management data, water pollution prevention and control data, water ecological restoration data, water administration law enforcement and supervision data, and video surveillance data. Calculate the reliability score based on different data positioning methods, determine the reliability weight of different positioning methods through the entropy weight method, construct the comprehensive score formula for pollution source positioning, and calculate the comprehensive reliability score for different zones; The information push management module pushes the location of the partition with the highest comprehensive reliability score to relevant personnel, who then conduct inspections in that partition to determine the final location and type of pollution sources and calculate the total pollution load based on the pollution source type. Verification monitoring points were set up in each zone. Model verification and optimization techniques were used to calculate the concentration error. When the error exceeded the threshold, parameter adjustments and weight optimization were performed. The specific process of parameter adjustment and weight optimization is as follows: Multiple verification monitoring points are set up in each zone to collect actual pollutant concentrations regularly. The collected actual pollutant concentrations are compared with the calculated pollutant concentrations to calculate the concentration error rate. An error threshold is set. If the concentration error rate is greater than the error threshold, the parameters are adjusted. Using feedback data, the weights of each factor are re-determined using the analytic hierarchy process. If special influencing factors in certain areas are found to be not included in the model, corresponding influencing factors and calculation formulas are added to improve the model.

2. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 1, characterized in that, The methods for obtaining the river and lake shoreline data are as follows: The process involves storing and processing river and lake maps using scanners and computers; determining the proportional relationship between the scanned river and lake maps and GIS vector images; converting raster data into vector data; storing the vector data; and finally, converting the vector data back into raster data.

3. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 2, characterized in that, The process of constructing the basic data information database is as follows: Set up water quality monitoring stations to collect routine water quality indicators, and collect meteorological and surrounding pollution source information; collect water pollution prevention and control data according to the type of pollution source, and build a water pollution prevention and control database. Monitor changes in water quality and quantity before and after the implementation of water ecological restoration projects, as well as the restoration of ecological functions at key ecological nodes; law enforcement officers monitor the overall condition of the river in real time, and the public can upload photos, videos, and related information about pollution problems they discover. Different types of high-definition cameras were installed at key locations around rivers and lakes to collect video data; the data was then integrated to build a basic data information database.

4. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 3, characterized in that, The specific process of the preliminary positioning is as follows: Divide the polluted area into Each zone is divided into several sections. Buffer zones are constructed using GIS buffer analysis technology. Spatial difference algorithms are used to determine pollution-related attribute values ​​and identify potential pollution source locations. Scoring formulas for river and lake shoreline factors are constructed to calculate scores. A comprehensive pollution index is calculated using the water pollution index method. Scoring formulas for water environment factors are set, and scores are calculated based on the comparison between the comprehensive pollution index and set thresholds. Pollution source types and locations are determined using material balance technology and actual removal volume. The QUAL2K model is used to predict water quality change trends and pollution source development dynamics. Scoring formulas for water pollution prevention and control factors are constructed to calculate scores.

5. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 4, characterized in that, The preliminary positioning process also includes: The Shannon-Wiener index is used to measure regional biodiversity. A scoring formula for water ecological restoration factors is constructed by integrating information on vegetation coverage and the presence or absence of specific sensitive species to calculate the score. The pollution frequency index is calculated using data statistical analysis technology. Video image analysis technology is used to count the number of frames of suspected pollution sources and the number of frames of actual confirmed pollution sources to calculate the identification accuracy. The video surveillance credibility coefficient is obtained by weighting the calculation based on equipment performance parameters and installation conditions.

6. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 5, characterized in that, The specific formula for the comprehensive score of pollution source location is as follows: ； in: , , , , , , These represent the weights of each factor, with values ​​ranging from [0,1]. Scoring is given for factors related to river and lake shorelines. Scoring for water environment governance factors, Water pollution prevention and control factor scores Scoring for water ecological restoration factors, The pollution frequency index, To improve recognition accuracy, This represents the credibility coefficient of video surveillance.

7. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 6, characterized in that, The specific calculation process for the total pollution load is as follows: The daily emission load of the pollutant from the enterprise is calculated based on the enterprise's pollutant concentration and wastewater discharge volume; the daily pollution load of the agricultural pollutant is calculated based on the crop varieties, area, fertilizer and pesticide usage and frequency collected from farmland surveys, combined with water quality monitoring data of farmland irrigation water and drainage; the daily pollution load of the pollutant from domestic pollution sources is calculated based on the urban population, domestic sewage generation, sewage treatment plant capacity and operation status; the total pollution load calculation formula is shown below: ； in: For total pollution load, This represents the company's daily emission load for this pollutant; This represents the daily pollution load of this pollutant in agriculture. The daily pollution load of this pollutant from domestic pollution sources. For the first Family businesses.

8. The method for determining non-point source pollution load in smart rivers based on GIS vector map conversion according to claim 7, characterized in that, The push process of the information push management module is as follows: Information is categorized and organized according to message type. Different push channels or multiple channels are used simultaneously to push messages based on their urgency and nature. Relevant personnel set message receiving preferences and priorities according to their needs, and the information push management module automatically pushes important information according to the priority.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is executed by a processor to implement the method for determining non-point source pollution load in smart rivers based on GIS vector map conversion as described in any one of claims 1-8.

Images