Assessment method and system for pollution damage in coastal waters

Abstract

The present invention discloses an evaluation method and system for pollution damage in nearshore waters, belonging to the technical field of pollution evaluation in nearshore waters. The method comprises: constructing a basic association network between environmental pollution behavior and environmental damage in nearshore waters; optimizing the basic association network to obtain an optimized association network, splitting the optimized association network into a pollution causal chain, the pollution causal chain including pollution causes, pollution phenomena and pollution results; collecting image data and water quality data of a target area, and constructing a time series based on the water quality data; analyzing the time series and image data to obtain the actual pollution phenomenon in the target area, and determining the corresponding actual pollution cause in combination with the pollution causal chain; calculating the loss cost of this pollution according to the actual pollution performance, the time series and the basic database, generating charts and reports and displaying them. The present invention can automatically determine the causal relationship between pollution phenomena and pollution behaviors, thereby improving the work efficiency of relevant personnel.

Description

Technical Field

[0001] The present invention belongs to the technical field of coastal waters pollution assessment, and in particular relates to a method and system for assessing coastal waters pollution damage. Background Art

[0002] With the acceleration of industrialization, illegal dumping in nearshore oceans and coastal zones has continued to increase, and the environmental pollution problem caused has become increasingly serious. If marine pollution problems can be discovered in time, it can help relevant departments to track and stop illegal dumping in time to avoid further deterioration of the nearshore marine environment.

[0003] In order to achieve rapid assessment of marine pollution, the following solutions are proposed in the prior art. For example, a Chinese patent document with publication number CN116975574A discloses a method for evaluating heavy metal pollution in the marine environment. The method establishes a joint distribution function of the marine test point and the historical data set based on the distribution characteristics of the historical monitoring data set, and then obtains the predicted value of the marine pollution value of the marine test point. The heavy metal pollution value of the marine monitoring point to be evaluated is further obtained in combination with the joint distribution function, and the marine environment is evaluated based on the heavy metal pollution value. For another example, a Chinese patent document with publication number CN109459549A discloses a real-time wireless marine pollution monitoring system. The system remotely collects various water quality data in seawater through sensor modules deployed in the ocean, and then evaluates the ecological environment of the ocean through the water quality data, so that marine pollution problems can be discovered in a timely manner.

[0004] The above two patent documents mainly focus on marine pollution assessment. In actual scenarios, if the causal relationship between environmental pollution and environmental damage can be determined, it is of great significance for holding polluters accountable and implementing environmental restoration. However, due to the complexity, latency and multi-factor complex of environmental pollution, the causal relationship is currently determined through manual investigation and evidence collection, which is undoubtedly inefficient. Summary of the invention

[0005] To solve the above problems, the present invention provides a method and system for evaluating pollution damage in nearshore waters to solve the problems in the prior art.

[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes a method for assessing pollution damage in coastal waters, comprising:

[0007] Construct a basic correlation network between environmental pollution behavior and coastal marine environmental damage based on network data;

[0008] Optimizing the basic association network to obtain an optimized association network, splitting the optimized association network into a pollution causal chain, wherein the pollution causal chain includes pollution causes, pollution phenomena and pollution results;

[0009] A target area is set in a nearshore sea area, image data and a plurality of water quality data of the target area are collected at intervals of a predetermined length of time, and a time series is constructed based on the water quality data collected at a plurality of time points;

[0010] Setting a trigger condition, and when the trigger condition is met, analyzing the time series and the image data to obtain the actual pollution phenomenon of the target area, and determining the corresponding actual pollution cause in combination with the pollution cause-effect chain;

[0011] Establishing a basic database, wherein the basic database includes a variety of calculation formulas and basic parameters for calculating loss costs, and calculating the loss costs of this pollution according to actual pollution performance, the time series and the basic database;

[0012] The actual pollution phenomenon, actual pollution causes and the loss costs are used as analysis results to generate charts and reports and display them.

[0013] Furthermore, establishing the basic association network includes the following steps:

[0014] An initial network structure is established, wherein the initial network structure includes a root node, a cause node, a manifestation node, and a result node, and a plurality of search statements are preset. The search statements are used to search the network data to obtain a plurality of first description statements, and a dependency analysis is performed on each of the first description statements to obtain a plurality of keywords and dependency relationships. The keywords are filled into the initial network structure according to the dependency relationships to obtain a plurality of sub-networks, and each of the sub-networks is connected with the root node as a connection point to complete the construction of the basic association network.

[0015] Furthermore, optimizing the basic association network includes the following steps:

[0016] Selecting a group of the cause nodes, the performance nodes, and the result nodes having a connection relationship in the basic association network as the first node, the second node, and the third node, respectively; using the first node and the second node to form an extended phrase; using the extended phrase to search the network data to obtain a plurality of second description sentences; generating an extended network based on the second description sentence; generating a network to be optimized according to the first node, the second node, and the third node; and selecting the extended network having at least one node identical to the network to be optimized as the network to be processed;

[0017] The same nodes in the network to be optimized and the network to be processed are merged, and the connection relationship between the nodes before the fusion is retained. After the fusion is completed, the network to be processed that has not been merged is continuously selected to continue the fusion until the fusion with all the networks to be processed is completed, and the network to be optimized at this time is defined as the initial optimized network;

[0018] In the initial optimization network, two nodes with a connection relationship are selected to form the extended phrase, and the above steps are repeated to continue the expansion until a new initial optimization network can no longer be obtained, and the initial optimization network at this time is used as the optimized association network.

[0019] Furthermore, splitting the optimized association network into the pollution causal chain comprises the following steps:

[0020] A plurality of causal paths are generated starting from the cause node in the optimized association network, the causal path is defined as the contaminated causal chain, the node at the tail end of the contaminated causal chain is defined as the result node, and the node between the cause node and the result node is defined as the performance node.

[0021] Further, calculating the loss cost includes the following steps:

[0022] The loss cost includes production loss, ecological loss and processing loss. The designated breeding area is obtained and the production loss L1 is calculated based on the first formula. The first formula is: , for the overlapping area between the breeding area and the target area, N is the number of breeding species in it, is the annual production per unit of the nth farmed species, is the area where the breeding area of ​​the nth breeding species overlaps with the target area, is the farming cost per unit of the nth farmed species, is the correction factor for the nth cultured species;

[0023] Set a plurality of natural species, and calculate the ecological loss L2 based on a second formula, wherein the second formula is: , for the target area, K is the number of natural species existing therein, is the number of units of the kth natural species existing before pollution, is the number of the kth natural species per unit after pollution, H is the average water depth of the target area, The cost per unit of restoration of the kth natural species;

[0024] A unit processing cost is set for each pollution phenomenon, and the processing loss L3 is calculated based on the third formula, which is: , where A is the unit treatment cost corresponding to the current pollution phenomenon, and D is the pollution area in the target area.

[0025] Furthermore, the analysis results include the pollution time point when the pollution first occurs, obtaining historical pollution data corresponding to the pollution causal chain, analyzing the historical pollution data to generate a response process of the nearshore marine environment under each pollution cause, the response process including the change trend of the water quality data within a preset time period after the pollution occurs, selecting the corresponding response process according to the actual pollution cause, combining the time series and the response process for inversion calculation to obtain the pollution time point.

[0026] Furthermore, the trigger condition includes a first condition and a second condition, and a pollution threshold is set for each type of water quality data. The first condition is met when the current water quality data is greater than the pollution threshold, and the second condition is met when the time series of the water quality data shows a continuous upward trend.

[0027] Furthermore, the water quality data includes the concentrations of nutrients, organic matter and heavy metals.

[0028] Furthermore, the water quality data collection method includes remote sensing image collection and buoy network collection.

[0029] The present invention also provides a system for evaluating pollution damage in coastal waters, which is used to implement the above-mentioned method for evaluating pollution damage in coastal waters. The system comprises:

[0030] A knowledge base module is used to construct a basic correlation network between environmental pollution behavior and coastal marine environmental damage based on network data, optimize the basic correlation network to obtain an optimized correlation network, and split the optimized correlation network into a pollution causal chain, wherein the pollution causal chain includes pollution causes, pollution phenomena and pollution results;

[0031] A collection module, used to set a target area in the nearshore waters, the collection module collects image data and multiple water quality data of the target area at intervals of a predetermined length of time, and constructs a time series based on the water quality data collected at multiple time points;

[0032] A verification module, used to set a trigger condition. When the trigger condition is met, the verification module analyzes the time series and the image data to obtain the actual pollution phenomenon of the target area, and determines the corresponding actual pollution cause in combination with the pollution cause-effect chain;

[0033] A quantification module is used to establish a basic database, wherein the basic database includes a variety of calculation formulas and basic parameters for calculating loss costs. The quantification module calculates the loss costs of this pollution according to the actual pollution performance, the time series and the basic database;

[0034] The display module generates charts and reports based on the actual pollution phenomenon, the actual pollution cause and the loss cost as analysis results and displays them.

[0035] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0036] The present invention first automatically constructs a pollution causal chain of pollution causes, pollution phenomena and pollution results based on network data. On this basis, the system collects image data and water quality data of the target sea area through buoys or remote sensing satellites, thereby realizing the automatic collection and storage of marine water quality data. When the trigger conditions are met, the image data and water quality data will also be analyzed, so that on the one hand, the pollution situation in the nearshore waters can be objectively evaluated, and on the other hand, the fact that pollution situation in the nearshore waters has occurred can be verified, thereby reducing the investigation burden of relevant personnel.

[0037] After determining the fact of marine pollution, the present invention determines the behavior that leads to the pollution through the pollution causal chain, thereby helping relevant personnel understand the causal relationship between the polluting behavior and the damage to the marine ecological environment, reducing their burden of collecting information. Finally, the present invention also establishes a basic database to calculate the loss cost of marine pollution according to the actual pollution level and the basic database. By calculating the loss cost, it realizes the quantitative analysis of the damage to the ecological environment, supports the establishment of the final damage assessment results of relevant personnel, and provides technical support for public interest litigation related to ecological environmental damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The present invention is a flowchart of the steps of the method for assessing pollution damage in coastal waters;

[0039] Figure 2 It is a schematic diagram of the structure of the system for evaluating pollution damage in nearshore waters according to the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0041] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0042] like Figure 1 As shown, a method for assessing pollution damage in coastal waters includes:

[0043] S1: Construct a basic correlation network between environmental pollution behavior and coastal waters environmental damage based on network data.

[0044] S2: Optimize the basic association network to obtain the optimized association network, and split the optimized association network into a pollution causal chain, which includes pollution causes, pollution phenomena and pollution results.

[0045] The network data is Internet data, including papers and journals, etc. The present invention automatically captures data from the Internet to construct a causal chain of pollution causes, pollution phenomena, and pollution results. Then, in order to increase the richness and accuracy of the basic association network, it continues to be optimized on the basis of the basic association network. It can be optimized using expert knowledge rules, or optimized using the method provided by this embodiment. The optimization method provided by this embodiment will be introduced later. By constructing an optimized association network, the causal relationship between various polluting behaviors that lead to damage to the ecological environment can be stored in the system in advance, thereby laying the foundation for subsequent analysis by assisting relevant personnel.

[0046] S3: A target area is set in the nearshore waters, image data and various water quality data of the target area are collected at predetermined time intervals, and a time series is constructed based on the water quality data collected at multiple time points.

[0047] The target area is manually set by the operator. After setting, the system can only analyze the sea area within the target area, thereby improving the analysis flexibility of the system; the predetermined time length is also manually set, which can be 1 day, 7 days, etc. The image data is the remote sensing image data of the sea level, and the water quality data includes suspended solids, organic matter, total phosphorus, total nitrogen and chlorophyll-a, etc. The method for collecting water quality data can use hyperspectral remote sensing images, and obtain it through the spectral characteristics corresponding to various elements in the hyperspectral remote sensing images. The specific details are existing technologies and will not be introduced here. In addition, buoy sensors can also be set up in nearshore waters to obtain water quality data. After collecting water quality data at multiple time points, the corresponding time series is constructed according to the order in which the water quality data appears.

[0048] S4: Set trigger conditions. When the trigger conditions are met, analyze the time series and image data to obtain the actual pollution phenomenon in the target area, and determine the corresponding actual pollution cause in combination with the pollution cause-effect chain.

[0049] If the trigger conditions are met, it indicates that there is a possibility of pollution in the current area. To further verify whether the sea area is polluted, the image data is analyzed in detail to determine whether there is floating garbage, algae blooms, etc. on the sea surface. The time series is analyzed to determine whether there is an increase in heavy metal elements or a decrease in organic elements in the water quality. After determining the actual pollution phenomenon, the possible cause of pollution can be determined based on the pollution cause-and-effect chain. For example, the actual pollution phenomenon of red tide determines that the cause of pollution is the dumping of phosphorus-containing industrial wastewater.

[0050] S5: Establish a basic database, which includes a variety of calculation formulas and basic parameters for calculating loss costs. The loss costs of this pollution are calculated based on actual pollution performance, time series and basic database.

[0051] S6: Generate charts and reports based on the actual pollution phenomenon, actual pollution causes and loss costs as analysis results and display them.

[0052] The loss costs in the present invention include production losses, ecological losses and treatment losses. Production losses are the loss amounts of fishery farming, ecological losses are the loss amounts of the number of biological species distribution, and treatment losses are the amounts required to control pollution. The specific calculation formula and specific parameters will be introduced later. After obtaining the actual pollution phenomenon, the present invention can automatically calculate the loss costs according to the basic database, thereby realizing the quantification of environmental ecological losses.

[0053] Finally, the present invention also automatically generates charts and reports based on actual pollution phenomena, actual pollution causes and loss costs to achieve visual display, thereby assisting relevant personnel to understand the system analysis results more intuitively.

[0054] The present invention first automatically constructs a pollution causal chain of pollution causes, pollution phenomena and pollution results based on network data. On this basis, the system collects image data and water quality data of the target sea area through buoys or remote sensing satellites, thereby realizing the automatic collection and storage of marine water quality data. When the trigger conditions are met, the image data and water quality data will also be analyzed. On the one hand, the pollution situation in the nearshore waters can be objectively evaluated, and on the other hand, the fact that pollution has occurred in the nearshore waters can be verified, thereby reducing the investigation burden of relevant personnel.

[0055] After determining the fact of marine pollution, the present invention determines the behavioral causes that lead to the pollution through the pollution causal chain, thereby helping relevant personnel understand the causal relationship between the polluting behavior and the damage to the marine ecological environment, reducing their burden of collecting information. Finally, the present invention also establishes a basic database to calculate the loss cost of marine pollution based on the actual pollution level and the basic database. By calculating the loss cost, it realizes the quantitative analysis of the damage to the ecological environment, supports the establishment of the final damage assessment results of relevant personnel, and provides technical support for public interest litigation related to ecological environmental damage.

[0056] It is particularly noteworthy that the causal relationship between pollution phenomena and pollution behaviors can be automatically determined through the present invention, thereby improving the work efficiency of relevant personnel.

[0057] In this embodiment, establishing a basic association network includes the following steps:

[0058] An initial network structure is established, which includes a root node, a cause node, a manifestation node, and a result node. A plurality of search statements are preset, and the search statements are used to search in the network data to obtain a plurality of first description statements. A dependency analysis is performed on each first description statement to obtain a plurality of keywords and dependency relationships. The keywords are filled into the initial network structure according to the dependency relationships to obtain a plurality of sub-networks. The sub-networks are connected with the root node as the connection point to complete the construction of the basic association network.

[0059] In the initial network results, the root node, the cause node, the performance node and the result node are connected in sequence. The search statements set are, for example, "What are the causes of pollution in the nearshore waters?" and "What are the causes of fish deaths in the nearshore waters?". Multiple search results are obtained through these search statements, and the search results include web articles, journals and papers. The answer to the search question in the search results can be extracted by introducing a third-party large language model (LLM), that is, the first description sentence. Then, the first description sentence is segmented and dependency analyzed using a deep learning algorithm to obtain the dependency relationship of the keywords. For example, the first description sentence is "Eutrophication will lead to red tides, which will lead to fish deaths." The keywords "eutrophication", "red tides" and "fish deaths" are extracted through dependency analysis. The dependency relationship is "eutrophication" leads to "red tides" and "red tides" lead to "fish deaths." They are filled into the corresponding nodes according to the dependency relationship, thereby completing the construction of a subnetwork. Repeating this step to analyze each first description sentence can obtain multiple subnetworks. Connecting multiple subnetworks through the root node can obtain a larger basic association network.

[0060] In this embodiment, optimizing the basic association network includes the following steps:

[0061] A group of cause nodes, performance nodes and result nodes with connection relationships are selected in the basic association network as the first node, the second node and the third node respectively, the first node and the second node are combined into an extended phrase, the extended phrase is used to search in the network data to obtain a variety of second description sentences, an extended network is generated based on the second description sentence, a network to be optimized is generated according to the first node, the second node and the third node, and an extended network that has at least one node identical to the network to be optimized is screened as a network to be processed.

[0062] The above steps are illustrated by an example below. It is assumed that the three nodes "eutrophication", "red tide" and "fish death" are selected as the first node, the second node and the third node respectively, and the combination of "eutrophication" and "red tide" is used as the extension number. "Eutrophic red tide" is used to search in the network data to obtain multiple second description statements, and then the method of generating subnetworks before is used to generate an extended network based on the second description statements. For example, the second description statement is "excessive discharge of ammonia nitrogen will lead to eutrophication, and then fish death will occur", and the generated extended network is "ammonia nitrogen discharge" leads to "eutrophication", "eutrophication" leads to "fish death", and because there are two nodes in common with the subnetwork, the extended network is the network to be processed.

[0063] The network to be optimized is merged with the same nodes in the network to be processed, and the connection relationship between the nodes before the fusion is retained. After the fusion is completed, it is continued to be selected as the unfused network to be processed to continue the fusion until the fusion with all the networks to be processed is completed. The network to be optimized at this time is defined as the initial optimized network.

[0064] Combined with the previous example, since both the sub-network and the network to be processed have "eutrophication" leading to "fish death", the two are merged. In the network to be processed, "ammonia nitrogen emission" is before "eutrophication", and in the sub-network, "red tide" is after "eutrophication". Therefore, after the fusion, the initial optimized network will be obtained in which "ammonia nitrogen emission" leads to "eutrophication", "eutrophication" leads to "red tide", and "red tide" leads to "fish death".

[0065] In the initial optimization network, two nodes with a connection relationship are selected to form an extended phrase, and the above steps are repeated to continue the expansion until a new initial optimization network can no longer be obtained. The initial optimization network at this time is used as the optimized association network.

[0066] For example, the combination of "ammonia nitrogen emission" and "eutrophication" is selected as an extended phrase to continue searching, thereby continuously serving the initial optimization network. When the initial optimization network can no longer be expanded according to the results, it indicates that the initial optimization network already has a high richness. Then the initial optimization network at this time is used as the optimized association network after the optimization is completed.

[0067] It can be seen that the above steps enrich the nodes of the association network, and through continuous expansion, each pollution cause and pollution process becomes more detailed, and the whole system is executed in an automated manner. Relevant personnel can better understand the root cause of pollution and the specific pollution process when using the optimized association network.

[0068] In this embodiment, splitting the optimized association network into contamination causal chains includes the following steps:

[0069] Taking the cause node in the optimized association network as the starting point, multiple causal paths are generated, and the causal paths are defined as contaminated causal chains. The nodes at the tail end of the contaminated causal chains are defined as result nodes, and the nodes between the cause nodes and the result nodes are defined as performance nodes.

[0070] Specifically, first select a cause node, and then take the cause node as the starting point to obtain multiple causal paths to the outermost node of the optimized association network. The outermost node is a node that no longer has subordinate nodes. Each path includes at least three nodes. The node located at the end of the causal path, that is, the outermost node in the optimized association network is taken as the result node. The node between the cause node and the result node is defined as a manifestation node. It can be seen that there may be multiple manifestation nodes. For example, eutrophication and red tide are both manifestations of pollution in coastal waters. Further, it can be seen that eutrophication and red tide are both manifestations of pollution in nearshore waters.

[0071] The calculation of the loss cost in this embodiment includes the following steps:

[0072] The loss cost includes production loss, ecological loss and processing loss. The designated breeding area is obtained and the production loss L1 is calculated based on the first formula. The first formula is: , for the overlapping area between the breeding area and the target area, N is the number of breeding species in it, is the annual production per unit of the nth farmed species, is the area where the breeding area of ​​the nth breeding species overlaps with the target area, is the farming cost per unit of the nth farmed species, is the correction factor for the nth cultured species.

[0073] The breeding area can be obtained in advance through manual surveys, including which species are farmed and in what areas. The distribution of the breeding areas obtained from the survey is entered into the system. If there is an overlapping area between the breeding area and the polluted target area, the area of ​​the overlapping area is calculated, and the production loss is estimated through the first formula. In the first formula, the annual output per unit of each breeding species can be set by the system default, or can be manually input through surveys. The correction factor can also be set by the system default, or can be determined manually. When the correction factor is manually determined, it can be the difference between the natural survival rate of the breeding species and the survival rate under the pollution state. This formula can quantify the damage to the production environment in the target area.

[0074] Set multiple natural species and calculate the ecological loss L2 based on the second formula, which is: , for the target area, K is the number of natural species existing therein, is the number of units of the kth natural species existing before pollution, is the number of the kth natural species per unit after pollution, H is the average water depth of the target area, Restoration cost per unit of the kth natural species.

[0075] Natural species include animals and plants that exist naturally in coastal waters. The number of animals and plants per unit is obtained through surveys. The restoration cost can be set by the system default or determined manually. This formula can be used to quantify the damage to the ecological environment in the target area.

[0076] A unit treatment cost is set for each pollution phenomenon, and the treatment loss L3 is calculated based on the third formula, which is: , where A is the unit treatment cost corresponding to the current pollution phenomenon, and D is the polluted area in the target area.

[0077] The analysis results of this embodiment also include the pollution time point when the pollution first occurred, obtaining historical pollution data corresponding to the pollution cause-effect chain, analyzing the historical pollution data to generate the response process of the nearshore marine environment under each pollution cause, the response process includes the change trend of water quality data within a preset time period after the pollution occurs, selecting the corresponding response process according to the actual pollution cause, and combining the time series and the response process for inversion calculation to obtain the pollution time point.

[0078] The purpose of this step is to infer the time point when the pollution first occurred based on the current pollution level in the nearshore waters. Specifically, first collect a large amount of historical pollution data that has occurred in the past, and analyze the historical pollution data to obtain the response process of the nearshore marine environment under each pollution cause. Historical pollution data include the time point and amount of the first dumping of pollutants, as well as the pollution manifestations at each subsequent time point, including algae outbreaks, biological mortality, etc.; by analyzing a large amount of historical data to obtain the response process of the sea area to each pollution behavior, deep learning algorithms can be used for data analysis, or the median of various historical data under each pollution cause can be taken as the response process; after obtaining the time series, obtain the pollution manifestations that appear at each time point, and by matching the corresponding response process, the time point when the pollution may first occur can be known.

[0079] The trigger conditions include a first condition and a second condition. A pollution threshold is set for each water quality data. The first condition is met when the current water quality data is greater than the pollution threshold, and the second condition is met when the time series of the water quality data shows a continuous upward trend.

[0080] Specifically, different pollution thresholds are set for different parameters such as heavy metals, chlorophyll, and organic matter. When the corresponding water quality data is monitored to exceed the pollution threshold, the time series is analyzed to determine whether there is pollution. When the water quality parameters in the time series do not exceed the pollution threshold, but there is an upward or downward trend, the time series will also be analyzed.

[0081] The water quality data in this implementation includes the concentrations of nutrients, organic matter and heavy metals. The methods for collecting water quality data include remote sensing image collection and buoy network collection, which have been introduced before and will not be repeated here.

[0082] like Figure 2 As shown, the present invention also provides a system for evaluating pollution damage in nearshore waters, which is used to implement the above-mentioned method for evaluating pollution damage in nearshore waters. The system includes:

[0083] The knowledge base module is used to construct a basic correlation network between environmental pollution behavior and coastal marine environmental damage based on network data, optimize the basic correlation network, obtain an optimized correlation network, and split the optimized correlation network into a pollution causal chain, which includes pollution causes, pollution phenomena and pollution results;

[0084] A collection module is used to set a target area in the nearshore waters. The collection module collects image data and multiple water quality data of the target area at intervals of a predetermined length of time, and constructs a time series based on the water quality data collected at multiple time points;

[0085] Verification module, used to set trigger conditions. When the trigger conditions are met, the verification module analyzes the time series and image data to obtain the actual pollution phenomenon in the target area, and determines the corresponding actual pollution cause in combination with the pollution cause-effect chain;

[0086] The quantitative module is used to establish a basic database, which includes a variety of calculation formulas and basic parameters for calculating the loss cost. The quantitative module calculates the loss cost of this pollution based on the actual pollution performance, time series and basic database;

[0087] The display module generates charts and reports based on the actual pollution phenomenon, actual pollution causes and loss costs as analysis results and displays them.

[0088] It should be understood that the various technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0089] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for assessing pollution damage in coastal waters, characterized in that: include: Construct a basic correlation network between environmental pollution behavior and coastal marine environmental damage based on network data; Optimizing the basic association network to obtain an optimized association network, splitting the optimized association network into a pollution causal chain, wherein the pollution causal chain includes pollution causes, pollution phenomena and pollution results; A target area is set in a nearshore sea area, image data and a plurality of water quality data of the target area are collected at intervals of a predetermined length of time, and a time series is constructed based on the water quality data collected at a plurality of time points; Setting a trigger condition, and when the trigger condition is met, analyzing the time series and the image data to obtain the actual pollution phenomenon of the target area, and determining the corresponding actual pollution cause in combination with the pollution cause-effect chain; Establishing a basic database, wherein the basic database includes a variety of calculation formulas and basic parameters for calculating loss costs, and calculating the loss costs of this pollution according to actual pollution phenomena, the time series and the basic database; Generate and display the actual pollution phenomenon, actual pollution causes and the loss costs as analysis results in charts and reports; Calculating the loss cost includes the following steps: The loss cost includes production loss, ecological loss and processing loss. The designated breeding area is obtained and the production loss L1 is calculated based on the first formula. The first formula is: , for the overlapping area between the breeding area and the target area, N is the number of breeding species in it, is the annual production per unit of the nth farmed species, is the area where the breeding area of ​​the nth breeding species overlaps with the target area, is the farming cost per unit of the nth farmed species, is the correction factor for the nth cultured species; Set a plurality of natural species, and calculate the ecological loss L2 based on a second formula, wherein the second formula is: , for the target area, K is the number of natural species existing therein, is the number of units of the kth natural species existing before pollution, is the number of the kth natural species per unit after pollution, H is the average water depth of the target area, The cost per unit of restoration of the kth natural species; A unit processing cost is set for each pollution phenomenon, and the processing loss L3 is calculated based on the third formula, which is: , where A is the unit treatment cost corresponding to the current pollution phenomenon, and D is the pollution area in the target area.

2. The method according to claim 1, characterized in that Establishing the basic association network includes the following steps: An initial network structure is established, wherein the initial network structure includes a root node, a cause node, a manifestation node, and a result node, and a plurality of search statements are preset. The search statements are searched in the network data to obtain a plurality of first description statements, and a dependency analysis is performed on each of the first description statements to obtain a plurality of keywords and dependency relationships. The keywords are filled into the initial network structure according to the dependency relationships to obtain a plurality of sub-networks, and each of the sub-networks is connected with the root node as a connection point to complete the construction of the basic association network.

3. The method according to claim 2, characterized in that Optimizing the basic association network includes the following steps: Selecting a group of the cause nodes, the performance nodes, and the result nodes having a connection relationship in the basic association network as the first node, the second node, and the third node, respectively; using the first node and the second node to form an extended phrase; using the extended phrase to search the network data to obtain a plurality of second description sentences; generating an extended network based on the second description sentence; generating a network to be optimized according to the first node, the second node, and the third node; screening the extended network having at least one node identical to the network to be optimized, and using it as a network to be processed; Merge the same nodes in the network to be optimized and the network to be processed, and retain the connection relationship between the nodes before the fusion, and continue to merge the fused network with the network to be processed that has not participated in the fusion, until the fusion with all the networks to be processed is completed, and define the network to be optimized at this time as the initial optimized network; In the initial optimization network, two nodes with a connection relationship are selected to form the extended phrase, and the above steps are repeated to continue the expansion until a new initial optimization network can no longer be obtained, and the initial optimization network at this time is used as the optimized association network.

4. The method according to claim 2, characterized in that: Splitting the optimized association network into the pollution causal chain comprises the following steps: A plurality of causal paths are generated starting from the cause node in the optimized association network, the causal path is defined as the contaminated causal chain, the node at the tail end of the contaminated causal chain is defined as the result node, and the node between the cause node and the result node is defined as the performance node.

5. The method according to claim 1, characterized in that The analysis results include the pollution time point when the pollution first occurs, obtaining historical pollution data corresponding to the pollution causal chain, analyzing the historical pollution data to generate the response process of the nearshore marine environment under each pollution cause, the response process includes the change trend of the water quality data within a preset time period after the pollution occurs, selecting the corresponding response process according to the actual pollution cause, and combining the time series and the response process for inversion calculation to obtain the pollution time point.

6. The method according to claim 1, characterized in that The trigger condition includes a first condition and a second condition. A pollution threshold is set for each type of water quality data. The first condition is met when the current water quality data is greater than the pollution threshold, and the second condition is met when the time series of the water quality data shows a continuous upward trend.

7. The method according to claim 1, characterized in that The water quality data include the concentrations of nutrients, organic matter and heavy metals.

8. The method according to claim 1, characterized in that The water quality data collection method includes remote sensing image collection and buoy network collection.

9. A system for evaluating pollution damage in coastal waters, used to implement the method according to any one of claims 1 to 8, characterized in that: include: A knowledge base module is used to construct a basic correlation network between environmental pollution behavior and coastal marine environmental damage based on network data, optimize the basic correlation network to obtain an optimized correlation network, and split the optimized correlation network into a pollution causal chain, wherein the pollution causal chain includes pollution causes, pollution phenomena and pollution results; A collection module, used to set a target area in the nearshore waters, the collection module collects image data and multiple water quality data of the target area at intervals of a predetermined length of time, and constructs a time series based on the water quality data collected at multiple time points; A verification module, used to set a trigger condition. When the trigger condition is met, the verification module analyzes the time series and the image data to obtain the actual pollution phenomenon of the target area, and determines the corresponding actual pollution cause in combination with the pollution cause-effect chain; A quantification module is used to establish a basic database, which includes a variety of calculation formulas and basic parameters for calculating loss costs. The quantification module calculates the loss cost of this pollution according to the actual pollution phenomenon, the time series and the basic database. The loss cost includes production loss, ecological loss and treatment loss. The designated breeding area is obtained and the production loss L1 is calculated based on the first formula. The first formula is: , for the overlapping area between the breeding area and the target area, N is the number of breeding species in it, is the annual production per unit of the nth farmed species, is the area where the breeding area of ​​the nth breeding species overlaps with the target area, is the farming cost per unit of the nth farmed species, is the correction coefficient of the nth farmed species, multiple natural species are set, and the ecological loss L2 is calculated based on the second formula, which is: , for the target area, K is the number of natural species existing therein, is the number of units of the kth natural species existing before pollution, is the number of the kth natural species per unit after pollution, H is the average water depth of the target area, The restoration cost per unit of the k-th natural species is set for each pollution phenomenon, and the processing loss L3 is calculated based on the third formula, which is: , where A is the unit treatment cost corresponding to the current pollution phenomenon, and D is the pollution area in the target area; The display module generates charts and reports based on the actual pollution phenomenon, the actual pollution cause and the loss cost as analysis results and displays them.

Images