An ecological environment pollution risk assessment method and system

Abstract

The application discloses an ecological environment pollution risk assessment method and system, comprising the following steps: collecting radioactive monitoring data of a target area and performing pretreatment; constructing a radioactive pollution space-time distribution observation field of the target area and generating a pollution observation field matrix; coupling a nuclide decay process with a diffusion migration process to form a radioactive decay-diffusion coupling evolution model; constructing a pollution digital twin evolution field, simulating and calculating the pollution diffusion state of the target area under a continuous time sequence; constructing a pollution source assumption set and forming a pollution source parameter vector set; iteratively updating and optimizing the pollution source parameter vector set; calculating the radioactive risk value of each spatial unit of the target area, dividing the risk level, and generating an ecological environment pollution risk assessment result. The application realizes radioactive pollution risk assessment by using the decay diffusion coupling and digital twin inversion method, and has the advantages of high accuracy and strong space-time characterization.

Description

Technical Field

[0001] This invention relates to the field of pollution risk assessment, and in particular to a method and system for assessing ecological and environmental pollution risks. Background Technology

[0002] With increasing public concern about environmental pollution, radioactive monitoring of the environment is continuously advancing. The collection and analysis of nucleoside activity concentrations and dose rates in target areas has led to the development of technical approaches for pollution identification, diffusion assessment, and risk evaluation based on monitoring data. Existing technologies typically analyze the regional radioactive pollution status through statistical analysis of monitoring point data, pollution diffusion model simulations, or risk indicator calculations, providing a basis for pollution early warning, emergency response, and environmental management.

[0003] However, existing technologies mostly focus on the static interpretation of monitoring results or the analysis of a single diffusion process, making it difficult to construct an observation field matrix, decay-diffusion coupled evolution model, and pollution digital twin evolution field around a unified spatiotemporal unit. This results in insufficient ability to invert the location of pollution sources, release intensity, and release time, and makes it difficult to accurately reconstruct pollution propagation paths and pollution evolution time series, thereby affecting the accuracy of calculating radioactive risk values ​​and classifying risk levels for each spatial unit. Summary of the Invention

[0004] One objective of this invention is to propose a method for assessing ecological and environmental pollution risks. This invention utilizes decay-diffusion coupling and digital twin inversion methods to achieve radioactive pollution risk assessment, which has the advantages of high accuracy and strong spatiotemporal characterization.

[0005] An ecological environment pollution risk assessment method according to an embodiment of the present invention includes the following steps:

[0006] Collect radioactive monitoring data from the target area, perform preprocessing, and generate a standardized pollution observation dataset.

[0007] Based on a standardized pollution observation dataset, a spatiotemporal distribution observation field of radioactive pollution in the target area is constructed, and a pollution observation field matrix is ​​generated.

[0008] Based on the pollution observation field matrix, a radionuclide decay time evolution model and an environmental medium diffusion and migration model are established, and the radionuclide decay process and diffusion and migration process are coupled to form a radioactive decay-diffusion coupled evolution model.

[0009] Based on the radioactive decay-diffusion coupled evolution model, a pollution digital twin evolution field is constructed to simulate and calculate the pollution diffusion state of the target area in a continuous time series, generating a virtual pollution spatiotemporal distribution sequence.

[0010] Based on the pollution observation field matrix and the virtual pollution spatiotemporal distribution sequence, a set of pollution source hypotheses is constructed, and a set of pollution source parameter vectors is formed.

[0011] The set of pollution source parameter vectors is iteratively updated and optimized to generate the optimal pollution source parameter results.

[0012] Based on the optimal pollution source parameters, the pollution propagation path and pollution evolution time series of the target area are generated, the radioactivity risk value of each spatial unit in the target area is calculated, the risk level is classified, and the ecological and environmental pollution risk assessment results are generated.

[0013] Optionally, the radioactivity monitoring data includes nuclide activity concentration and dose rate.

[0014] Optionally, the preprocessing includes time synchronization, missing data completion, anomaly removal, and data normalization.

[0015] Optionally, the generation of the pollution observation field matrix specifically includes:

[0016] Read the standardized pollution observation dataset, divide the target area into spatial units according to the spatial coordinates of each monitoring point in the target area, obtain a set of spatial units, and establish a corresponding time index based on the sampling time;

[0017] The radionuclide activity concentration data and dose rate data of each monitoring point at each sampling time are mapped to the corresponding spatial unit and the corresponding time index, respectively, to form the radioactive observation data entries of each spatial unit under each time index;

[0018] Extract all radioactive observation data entries under the same time index for the same spatial unit, and count the number of radioactive observation data entries.

[0019] When the number of radioactive observation data entries is greater than 1, the nuclide activity concentration data and dose rate data in all radioactive observation data entries are aggregated to obtain the nuclide activity concentration aggregated value and dose rate aggregated value, and the radioactivity intensity characterization value of the corresponding spatial unit under the corresponding time index is generated.

[0020] When the number of radioactive observation data entries is equal to 1, the radionuclide activity concentration data and dose rate data in the radioactive observation data entry are read, and the radioactivity intensity characterization value of the corresponding spatial unit under the corresponding time index is generated.

[0021] The radioactivity intensity characterization values ​​corresponding to each spatial unit under each time index are organized according to the correspondence between spatial units and time indices to construct a spatiotemporal distribution observation field for radioactive contamination.

[0022] Using spatial units as row indices and time indices as column indices, the radioactivity intensity characterization values ​​of each spatial unit in the spatiotemporal distribution observation field of radioactive contamination under each time index are written into the corresponding matrix positions to generate the contamination observation field matrix.

[0023] Optionally, the construction of the radioactive decay-diffusion coupled evolution model specifically includes:

[0024] Read the radioactivity intensity characterization values ​​of each spatial unit in the pollution observation field matrix under each time index, extract the radioactivity intensity change sequence corresponding to each spatial unit under continuous time index, and determine the nuclide decay trend value corresponding to each spatial unit;

[0025] Based on the nuclide decay trend value corresponding to each spatial unit, the nuclide decay transfer relationship between each time index is established, a nuclide decay time evolution model for continuous time index is generated, and the nuclide decay evolution value corresponding to each spatial unit under each time index is obtained.

[0026] By combining the spatial distribution relationship of each spatial unit in the pollution observation field matrix, the diffusion and migration correlation between adjacent spatial units is established. Based on the radioactivity intensity change sequence of each spatial unit under continuous time index, spatial migration and transfer paths are generated to form an environmental medium diffusion and migration model, and the corresponding diffusion and migration values ​​between each spatial unit are obtained.

[0027] The nuclide decay evolution values ​​in the nuclide decay time evolution model and the diffusion migration values ​​in the environmental medium diffusion migration model are correlated and mapped according to the same time index and the same spatial unit to form the joint decay migration evolution value corresponding to each spatial unit under each time index.

[0028] A radioactive decay-diffusion coupled evolution model is constructed based on the decay-migration joint evolution values ​​of each spatial unit at each time index.

[0029] Optionally, the generation of the virtual pollution spatiotemporal distribution sequence specifically includes:

[0030] Read the decay migration joint evolution values ​​of each spatial unit in the radioactive decay-diffusion coupled evolution model at each time index, and arrange them according to the correspondence between spatial units and time indices to form a joint evolution value sequence corresponding to the target region;

[0031] Based on the joint evolution value sequence, virtual pollution mapping units corresponding to the target area are constructed according to the spatial coordinates and time indices of the spatial units. The decay and migration joint evolution values ​​of each spatial unit under each time index are written into the corresponding virtual pollution mapping unit to obtain the pollution digital twin basic unit set.

[0032] The values ​​of each virtual pollution mapping unit in the basic unit set of pollution digital twin are continuously correlated in time to form a virtual pollution evolution chain corresponding to each spatial unit under continuous time index, and spatial adjacency correlation is performed on each virtual pollution evolution chain to generate a virtual pollution evolution correlation network.

[0033] Based on the virtual pollution evolution correlation network, the pollution diffusion state of each spatial unit under each time index is recursively simulated to obtain the virtual pollution diffusion value corresponding to each spatial unit under each time index. The values ​​are then organized according to the time index order to form the virtual pollution time series distribution results corresponding to each spatial unit.

[0034] The virtual pollution time series distribution results corresponding to each spatial unit are integrated according to the correspondence between spatial units and time indices to construct a pollution digital twin evolution field;

[0035] The virtual pollution diffusion values ​​corresponding to each spatial unit under each time index are extracted from the pollution digital twin evolution field and arranged according to the time index order and the spatial unit distribution order to generate a virtual pollution spatiotemporal distribution sequence.

[0036] Optionally, the generation of the pollution source parameter vector set specifically includes:

[0037] Read the radioactivity intensity characterization value corresponding to each spatial unit in the pollution observation field matrix under each time index, and read the virtual pollution diffusion value corresponding to each spatial unit in the virtual pollution spatiotemporal distribution sequence under each time index, and align them to form the observation simulation corresponding data group;

[0038] For each set of observation simulation data, the difference between the radioactivity intensity characterization value and the virtual pollution diffusion value is calculated, and the data is organized according to the spatial unit and time index order to generate a set of observation simulation difference values.

[0039] Based on the variation of the difference of each spatial unit in the observation simulation difference set under the continuous time index, the feature region of continuous concentration of difference is extracted, and combined with the diffusion direction distribution results of the corresponding spatial unit in the virtual pollution spatiotemporal distribution sequence, the set of candidate pollution source spatial locations is determined.

[0040] For each candidate pollution source spatial location in the candidate pollution source spatial location set, the radioactivity intensity characterization value and virtual pollution diffusion value of its corresponding spatial unit under each time index are read to generate the corresponding source intensity change sequence, and the release intensity parameter and release time parameter corresponding to each candidate pollution source spatial location are determined based on the source intensity change sequence.

[0041] The spatial location parameters, release intensity parameters, and release time parameters corresponding to the spatial locations of each candidate pollution source are combined to form a set of pollution source hypotheses. Each pollution source hypothesis is then sequentially numbered to generate a pollution source parameter vector.

[0042] All pollution source parameter vectors are organized according to the pollution source hypothesis numbering order to form a pollution source parameter vector set.

[0043] Optionally, the generation of the optimal pollution source parameter results specifically includes:

[0044] Read the set of pollution source parameter vectors and extract the corresponding spatial location parameters, release intensity parameters and release time parameters to form a set of candidate pollution source parameter groups;

[0045] Based on the candidate groups of each pollution source parameter, the virtual pollution spatiotemporal distribution sequence is iteratively adjusted to obtain the updated virtual pollution distribution results corresponding to each candidate group of pollution source parameter;

[0046] The updated virtual pollution distribution results are compared with the pollution observation field matrix. The observation simulation deviation value of each spatial unit under each time index is calculated, and the total deviation value corresponding to each pollution source parameter candidate group is generated.

[0047] The total deviation values ​​corresponding to each pollution source parameter candidate group are sorted, and the pollution source parameter candidate group with the smallest total deviation value is determined as the current optimal pollution source parameter candidate group, and the corresponding total deviation value is determined as the current optimal deviation value.

[0048] Based on the spatial location parameters, release intensity parameters, and release time parameters in the current optimal pollution source parameter candidate group, incremental and decremental adjustments are made to each parameter to generate a new set of pollution source parameter candidate groups. The virtual pollution spatiotemporal distribution sequence iterative adjustment and total deviation value calculation are repeated to obtain the parameter optimization results corresponding to each pollution source parameter candidate group in the new round.

[0049] The parameter optimization results corresponding to each pollution source parameter candidate group in the new round are compared. When the minimum total deviation value in the new round of parameter optimization results is less than the current optimal deviation value, the pollution source parameter candidate group corresponding to the minimum total deviation value is determined as the updated optimal pollution source parameter candidate group, and the minimum total deviation value is updated to the current optimal deviation value.

[0050] Repeat the process until the current optimal deviation value is less than the preset deviation threshold, then stop the iteration and determine the updated optimal pollution source parameter candidate group at the time of stopping the iteration as the optimal pollution source parameter result.

[0051] Optionally, the generation of the ecological and environmental pollution risk assessment results specifically includes:

[0052] Read the spatial location parameters and release intensity parameters from the optimal pollution source parameter results, determine the initial spatial unit and the initial time index of the pollution source, and form a pollution source initial parameter group;

[0053] Based on the pollution source initial parameter group, the pollution propagation sequence and direction between spatial units are determined according to the time index, and the pollution propagation path corresponding to the target area is generated.

[0054] Based on the transmission order of each spatial unit in the pollution propagation path, the pollution arrival time index and pollution duration index corresponding to each spatial unit are extracted to generate the pollution evolution time series corresponding to the target area.

[0055] Pollution transmission paths and pollution evolution time series are organized according to the correspondence between spatial units and time indices to generate pollution transmission process data;

[0056] Read the pollution arrival time index and pollution duration index corresponding to each spatial unit in the pollution propagation process data, and read the current adjusted pollution value corresponding to each spatial unit under each time index in the updated virtual pollution distribution result. Extract the maximum current adjusted pollution value corresponding to each spatial unit as the pollution intensity result, and calculate the radioactivity risk value corresponding to each spatial unit.

[0057] Based on the correspondence between the radioactivity risk value of each space unit and the preset risk level classification threshold, the risk level of each space unit is classified, and the risk level result of each space unit is generated.

[0058] The radioactivity risk values ​​and risk levels corresponding to each spatial unit are organized according to the spatial unit distribution order to generate ecological and environmental pollution risk assessment results.

[0059] An ecological environment pollution risk assessment system according to an embodiment of the present invention includes:

[0060] The data acquisition module is used to collect radioactivity monitoring data of the target area, perform preprocessing, and form a standardized pollution observation dataset;

[0061] The pollution observation field construction module is used to construct a spatiotemporal distribution observation field of radioactive pollution in the target area and generate a pollution observation field matrix.

[0062] The coupled evolution model construction module is used to establish a radionuclide decay time evolution model and an environmental medium diffusion and migration model based on the pollution observation field matrix, forming a radioactive decay-diffusion coupled evolution model;

[0063] The pollution digital twin evolution module is used to construct a pollution digital twin evolution field based on the radioactive decay-diffusion coupled evolution model, perform simulation calculations, and generate a virtual pollution spatiotemporal distribution sequence.

[0064] The pollution source hypothesis construction module is used to construct a set of pollution source hypotheses based on the pollution observation field matrix and the virtual pollution spatiotemporal distribution sequence, and to form a set of pollution source parameter vectors.

[0065] The pollution source parameter optimization module is used to iteratively update and optimize the set of pollution source parameter vectors to generate the optimal pollution source parameter results.

[0066] The risk assessment results generation module is used to generate pollution propagation paths and pollution evolution time series in the target area, calculate the radioactivity risk value of each spatial unit in the target area, classify the risk level, and generate ecological and environmental pollution risk assessment results.

[0067] The beneficial effects of this invention are:

[0068] Compared to existing technologies, this invention can construct a unified spatiotemporal distribution observation field of radioactive pollution based on radioactive monitoring data of the target area, and further form a pollution observation field matrix. On this basis, the radionuclide decay process is uniformly coupled with the diffusion and migration process of the environmental medium to establish a radioactive decay-diffusion coupled evolution model. This allows for an integrated characterization of the evolutionary relationship of radioactive pollution in continuous temporal and spatial diffusion processes. Compared to methods that rely solely on static interpretation of monitoring results or analyze only a single diffusion process, this invention can more comprehensively reflect the spatiotemporal variation characteristics of radioactive pollution within the target area, improve the ability to characterize the pollution evolution state, and provide a more stable data and evolutionary foundation for subsequent pollution identification and risk assessment.

[0069] Meanwhile, this invention constructs a digital twin evolution field of pollution based on a radioactive decay-diffusion coupled evolution model, generating a virtual spatiotemporal distribution sequence of pollution. It then performs a corresponding analysis between the pollution observation field matrix and the virtual spatiotemporal distribution sequence, thereby constructing a set of pollution source hypotheses and a set of pollution source parameter vectors. Through iterative updates and optimization, the optimal pollution source parameters are obtained, enabling reverse identification of the spatial location, release intensity, and release time of pollution sources. Compared to existing technologies with insufficient pollution source identification capabilities and weak correlation between pollution diffusion mechanisms and observation results, this invention integrates observation data, coupled evolution, and parameter inversion, enhancing the consistency between pollution source inversion results and actual pollution states, and improving the accuracy and reliability of pollution source identification.

[0070] Furthermore, after obtaining the optimal pollution source parameters, this invention can generate pollution propagation paths and pollution evolution time series for the target area. It also calculates radioactive risk values ​​by combining the pollution arrival time, pollution duration, and pollution intensity of each spatial unit, completing the risk level classification and ultimately forming an ecological and environmental pollution risk assessment result. Therefore, it not only enables the identification of the pollution status of the target area but also achieves a systematic output of the pollution propagation process and risk distribution results. This makes the risk assessment results more spatially and temporally specific and valuable, improving the accuracy, completeness, and practicality of ecological and environmental radioactive pollution assessment, and providing more effective technical support for pollution early warning, emergency response, and environmental governance. Attached Figure Description

[0071] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0072] Figure 1 This is a flowchart of an ecological environment pollution risk assessment method proposed in this invention;

[0073] Figure 2 This is a schematic diagram illustrating the construction process of a radioactive decay-diffusion coupled evolution model for an ecological and environmental pollution risk assessment method proposed in this invention. Detailed Implementation

[0074] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0075] refer to Figure 1 and Figure 2 An ecological and environmental pollution risk assessment method includes the following steps:

[0076] Collect radioactive monitoring data from the target area, perform preprocessing, and generate a standardized pollution observation dataset.

[0077] Based on a standardized pollution observation dataset, a spatiotemporal distribution observation field of radioactive pollution in the target area is constructed, and a pollution observation field matrix is ​​generated.

[0078] Based on the pollution observation field matrix, a radionuclide decay time evolution model and an environmental medium diffusion and migration model are established, and the radionuclide decay process and diffusion and migration process are coupled to form a radioactive decay-diffusion coupled evolution model.

[0079] Based on the radioactive decay-diffusion coupled evolution model, a pollution digital twin evolution field is constructed to simulate and calculate the pollution diffusion state of the target area in a continuous time series, generating a virtual pollution spatiotemporal distribution sequence.

[0080] Based on the pollution observation field matrix and the virtual pollution spatiotemporal distribution sequence, a set of pollution source hypotheses is constructed, and a set of pollution source parameter vectors is formed.

[0081] The set of pollution source parameter vectors is iteratively updated and optimized to generate the optimal pollution source parameter results.

[0082] Based on the optimal pollution source parameters, the pollution propagation path and pollution evolution time series of the target area are generated, the radioactivity risk value of each spatial unit in the target area is calculated, the risk level is classified, and the ecological and environmental pollution risk assessment results are generated.

[0083] In this embodiment, the radioactivity monitoring data includes nuclide activity concentration and dose rate.

[0084] In this embodiment, preprocessing includes time synchronization, missing data completion, anomaly removal, and data normalization.

[0085] In this embodiment, the generation of the pollution observation field matrix specifically includes:

[0086] Read the standardized pollution observation dataset, divide the target area into spatial units according to the spatial coordinates of each monitoring point in the target area, obtain a set of spatial units, and establish a corresponding time index based on the sampling time;

[0087] The radionuclide activity concentration data and dose rate data of each monitoring point at each sampling time are mapped to the corresponding spatial unit and the corresponding time index, respectively, to form the radioactive observation data entries of each spatial unit under each time index;

[0088] Extract all radioactive observation data entries under the same time index for the same spatial unit, and count the number of radioactive observation data entries.

[0089] When the number of radioactive observation data entries is greater than 1, the nuclide activity concentration data and dose rate data in all radioactive observation data entries are aggregated to obtain the nuclide activity concentration aggregated value and dose rate aggregated value, and the radioactivity intensity characterization value of the corresponding spatial unit under the corresponding time index is generated.

[0090] The generation of radioactivity intensity characterization values ​​specifically includes: averaging and pooling the nuclide activity concentration data from all radioactivity observation data entries under the same time index for the same spatial unit to obtain a pooled nuclide activity concentration value; averaging and pooling the dose rate data from all radioactivity observation data entries to obtain a pooled dose rate value; standardizing the pooled nuclide activity concentration value and the pooled dose rate value to obtain standardized nuclide activity concentration values ​​and standardized dose rate values; and weighted summing the standardized nuclide activity concentration values ​​and standardized dose rate values ​​to generate the radioactivity intensity characterization value for the corresponding spatial unit under the corresponding time index.

[0091] When the number of radioactive observation data entries is equal to 1, the radionuclide activity concentration data and dose rate data in the radioactive observation data entry are read, and the radioactivity intensity characterization value of the corresponding spatial unit under the corresponding time index is generated.

[0092] The radioactivity intensity characterization values ​​corresponding to each spatial unit under each time index are organized according to the correspondence between spatial units and time indices to construct a spatiotemporal distribution observation field for radioactive contamination.

[0093] Using spatial units as row indices and time indices as column indices, the radioactivity intensity characterization values ​​of each spatial unit in the spatiotemporal distribution observation field of radioactive contamination under each time index are written into the corresponding matrix positions to generate the contamination observation field matrix.

[0094] In this embodiment, the construction of the radioactive decay-diffusion coupled evolution model specifically includes:

[0095] Read the radioactivity intensity characterization values ​​of each spatial unit in the pollution observation field matrix under each time index, extract the radioactivity intensity change sequence corresponding to each spatial unit under continuous time index, and determine the nuclide decay trend value corresponding to each spatial unit;

[0096] The generation of the nuclide decay trend value specifically includes: reading the radioactivity intensity characterization value corresponding to a single spatial unit under all time indices from the contamination observation field matrix row by row, forming a set of radioactivity intensity time series values ​​corresponding to the spatial unit; arranging the set of radioactivity intensity time series values ​​according to the order of the time indices to obtain the radioactivity intensity change sequence corresponding to the spatial unit; performing linear fitting on the radioactivity intensity change sequence with the time index as the input sequence and the radioactivity intensity change sequence as the output sequence to obtain the fitted decay curve corresponding to the spatial unit; reading the first fitted result value under the first time index and the last fitted result value under the last time index of the fitted decay curve, calculating the decay difference between the first fitted result value and the last fitted result value to obtain the total decay value corresponding to the spatial unit; counting the number of time index intervals between the first time index and the last time index, and dividing the total decay value by the number of time index intervals to obtain the unit time index decay value corresponding to the spatial unit; and determining the unit time index decay value as the nuclide decay trend value corresponding to the spatial unit.

[0097] Based on the nuclide decay trend value corresponding to each spatial unit, the nuclide decay transfer relationship between each time index is established, a nuclide decay time evolution model for continuous time index is generated, and the nuclide decay evolution value corresponding to each spatial unit under each time index is obtained.

[0098] The generation of nuclide decay evolution values ​​specifically includes: reading the nuclide decay trend values ​​corresponding to each spatial unit to form a set of nuclide decay trend values ​​corresponding one-to-one with each spatial unit; then reading the radioactivity intensity characterization value corresponding to each spatial unit under the first time index, and determining the radioactivity intensity characterization value as the initial nuclide decay evolution value of each spatial unit under the first time index, thus obtaining the initial set of nuclide decay evolution values; subsequently, for each spatial unit, according to the chronological order of the time index, reading the nuclide decay trend value corresponding to that spatial unit, and setting the nuclide decay trend value under the previous time index as the initial set of nuclide decay evolution values. Subtracting the nuclide decay trend value from the decay evolution value yields the nuclide decay evolution value for the next time index. This process is repeated to generate a sequence of nuclide decay evolution values ​​for the spatial unit across all time indices. The sequence of nuclide decay evolution values ​​for all spatial units is then organized according to the correspondence between spatial units and time indices to generate a nuclide decay time evolution model oriented towards continuous time indices. The numerical results corresponding to each spatial unit under each time index are extracted from the nuclide decay time evolution model to obtain the nuclide decay evolution value for each spatial unit under each time index.

[0099] By combining the spatial distribution relationship of each spatial unit in the pollution observation field matrix, the diffusion and migration correlation between adjacent spatial units is established. Based on the radioactivity intensity change sequence of each spatial unit under continuous time index, spatial migration and transfer paths are generated to form an environmental medium diffusion and migration model, and the corresponding diffusion and migration values ​​between each spatial unit are obtained.

[0100] The generation of diffusion migration values ​​specifically includes: reading the spatial coordinates of each spatial unit in the pollution observation field matrix; determining the set of adjacent spatial units corresponding to each spatial unit according to the adjacency relationship of the spatial coordinates to obtain an adjacent spatial unit association table; then, based on the adjacent spatial unit association table, establishing a one-to-one diffusion migration association relationship between each spatial unit and its corresponding adjacent spatial units to obtain a spatial diffusion association unit set; reading the radioactivity intensity change sequence of each spatial unit under continuous time index; calculating the difference between the radioactivity intensity characterization values ​​of each spatial unit and each adjacent spatial unit under the same time index to obtain the intensity difference sequence of each pair of adjacent spatial units under each time index; determining the direction of the intensity difference sequence of each pair of adjacent spatial units according to the order of the time index; when the radioactivity intensity characterization value of a certain spatial unit is greater than that of its adjacent units under a preset number of continuous time indices, the direction of the difference is determined. When characterizing the radioactivity intensity of a spatial unit, the spatial unit is designated as the originating spatial unit and its adjacent spatial units as the destination spatial units, thus obtaining the corresponding spatial migration and transmission direction judgment conditions. Next, the number of times each pair of adjacent spatial units satisfies the transmission direction judgment conditions across all time indices is counted, yielding the transmission count value. The intensity differences under the corresponding time indices are then accumulated to obtain the cumulative difference. This cumulative difference is divided by the transmission count value to obtain the average migration intensity value for the corresponding pair of adjacent spatial units. The originating spatial unit, destination spatial unit, and average migration intensity value are then correlated to obtain the spatial migration and transmission path. All spatial migration and transmission paths are summarized according to the adjacency relationships between spatial units to generate an environmental medium diffusion and migration model. The average migration intensity value corresponding to each spatial migration and transmission path is then determined as the corresponding diffusion and migration value between each spatial unit.

[0101] The nuclide decay evolution values ​​in the nuclide decay time evolution model and the diffusion migration values ​​in the environmental medium diffusion migration model are correlated and mapped according to the same time index and the same spatial unit to form the joint decay migration evolution value corresponding to each spatial unit under each time index.

[0102] The generation of decay-migration joint evolution values ​​specifically includes: first, reading the nuclide decay evolution values ​​corresponding to each spatial unit in the nuclide decay time evolution model at each time index, forming a set of nuclide decay evolution values; then, reading the diffusion migration values ​​corresponding to each spatial unit and its adjacent spatial units in the environmental medium diffusion migration model, and organizing them according to the correspondence between the transfer start spatial unit, the transfer end spatial unit, and the time index, forming a set of diffusion migration values; for each spatial unit at each time index, extracting all diffusion migration values ​​with that spatial unit as the transfer end spatial unit, obtaining the set of migration-in diffusion values ​​for that spatial unit at that time index, and simultaneously extracting all diffusion migration values ​​with that spatial unit as the transfer start spatial unit, obtaining... The set of out-diffusion values ​​for the spatial unit at the given time index is obtained; the diffusion migration values ​​in the set of in-diffusion values ​​are summed to obtain the total in-diffusion value, and the diffusion migration values ​​in the set of out-diffusion values ​​are summed to obtain the total out-diffusion value; the nuclide decay evolution value of the spatial unit at the given time index is added to the total in-diffusion value to obtain the intermediate evolution value, and then the total out-diffusion value is subtracted from the intermediate evolution value to obtain the joint decay migration evolution value of the spatial unit at the given time index; the joint decay migration evolution values ​​corresponding to each spatial unit at each time index are organized according to the correspondence between spatial units and time indices to obtain a set of joint decay migration evolution values ​​covering the entire spatial range and the entire time series range of the target region;

[0103] Based on the decay-migration joint evolution values ​​of each spatial unit under each time index, a radioactive decay-diffusion coupled evolution model is constructed to characterize the joint evolution process of radioactive pollution in the target area under continuous time index and continuous spatial unit, which is simultaneously affected by the decay of radionuclides and the diffusion and migration of environmental media.

[0104] In this embodiment, the generation of the virtual spatiotemporal distribution sequence of pollution specifically includes:

[0105] Read the decay migration joint evolution values ​​of each spatial unit in the radioactive decay-diffusion coupled evolution model at each time index, and arrange them according to the correspondence between spatial units and time indices to form a joint evolution value sequence corresponding to the target region;

[0106] Based on the joint evolution value sequence, virtual pollution mapping units corresponding to the target area are constructed according to the spatial coordinates and time indices of the spatial units. The decay and migration joint evolution values ​​of each spatial unit under each time index are written into the corresponding virtual pollution mapping unit to obtain the pollution digital twin basic unit set.

[0107] A virtual contamination mapping unit is a virtual computing unit that corresponds one-to-one with a spatial unit and a time index within a target area. It is used to carry the decay and migration joint evolution value corresponding to the spatial unit under the time index, thereby characterizing the virtual contamination distribution state of radioactive contamination at a specific spatial location and at a specific time. The contamination digital twin basic unit set is a set of units composed of virtual contamination mapping units corresponding to all spatial units in the target area under all time indices. It is used to characterize the basic structure of the virtual evolution of radioactive contamination in the target area across the entire spatial range and the entire time series.

[0108] The values ​​of each virtual pollution mapping unit in the basic unit set of pollution digital twin are continuously correlated in time to form a virtual pollution evolution chain corresponding to each spatial unit under continuous time index, and spatial adjacency correlation is performed on each virtual pollution evolution chain to generate a virtual pollution evolution correlation network.

[0109] The generation of the virtual pollution evolution correlation network specifically includes: extracting all virtual pollution mapping units belonging to the same spatial unit from the pollution digital twin basic unit set, arranging them according to the chronological order of time indices to obtain the virtual pollution temporal unit sequence corresponding to the spatial unit; sequentially reading the decay-migration joint evolution values ​​corresponding to each virtual pollution mapping unit in the virtual pollution temporal unit sequence, and establishing temporal connection relationships between adjacent virtual pollution mapping units according to the order of the previous time index pointing to the next time index to obtain the temporal correlation unit chain corresponding to the spatial unit; organizing each virtual pollution mapping unit and its corresponding temporal connection relationship in the temporal correlation unit chain to generate the... The virtual pollution evolution chain corresponding to a spatial unit under a continuous time index is defined. For each time index, all virtual pollution mapping units corresponding to that spatial unit are extracted. Based on the spatial adjacency relationship between spatial units, spatial adjacency connection relationships are established between each virtual pollution mapping unit to obtain the spatial adjacency unit group corresponding to that time index. The spatial adjacency unit groups corresponding to each time index are associated and integrated with the virtual pollution evolution chains corresponding to each spatial unit to obtain a set of inter-chain connection relationships covering the entire spatial range and the entire time series range of the target area. The set of inter-chain connection relationships is then uniformly organized with all virtual pollution evolution chains to generate a virtual pollution evolution association network.

[0110] Based on the virtual pollution evolution correlation network, the pollution diffusion state of each spatial unit under each time index is recursively simulated to obtain the virtual pollution diffusion value corresponding to each spatial unit under each time index. The values ​​are then organized according to the time index order to form the virtual pollution time series distribution results corresponding to each spatial unit.

[0111] The generation of the virtual pollution time-series distribution results specifically includes: first, reading the virtual pollution mapping unit values ​​corresponding to each spatial unit in the virtual pollution evolution correlation network under the first time index, and determining the virtual pollution mapping unit values ​​as the initial virtual pollution diffusion values ​​of each spatial unit under the first time index, thus obtaining the initial virtual pollution diffusion value set; then, for each spatial unit, reading the virtual pollution mapping unit values ​​corresponding to that spatial unit under the current time index, thus obtaining the current unit evolution value, and simultaneously reading the virtual pollution mapping unit values ​​corresponding to each adjacent spatial unit with spatial adjacency connection under the current time index, thus obtaining the adjacent unit evolution value set; subsequently, summing the virtual pollution mapping unit values ​​of each adjacent spatial unit in the adjacent unit evolution value set to obtain the total diffusion value of adjacent units, and statistically analyzing the adjacent unit evolution values. The number of adjacent spatial units in the evolution value set is used to obtain the number of adjacent units. The total diffusion value of adjacent units is divided by the number of adjacent units to obtain the average diffusion value of adjacent units. The current unit evolution value is then arithmetically averaged with the average diffusion value of adjacent units to obtain the recursive diffusion value of the spatial unit under the current time index. According to the chronological order of the time indices, the virtual pollution diffusion value of the spatial unit under the previous time index is arithmetically averaged with the recursive diffusion value under the current time index to obtain the virtual pollution diffusion value of the spatial unit under the current time index. This process is repeated to generate a sequence of virtual pollution diffusion values ​​for the spatial unit under all time indices. The sequence of virtual pollution diffusion values ​​for each spatial unit under all time indices is then organized according to the time index order to form the virtual pollution time series distribution result for each spatial unit.

[0112] The virtual pollution time series distribution results corresponding to each spatial unit are integrated according to the correspondence between spatial units and time indices to construct a pollution digital twin evolution field;

[0113] The pollution digital twin evolution field is a dynamic evolution calculation field of radioactive pollution constructed in a virtual environment for the entire spatial range and the entire time series range of the target area. Based on the virtual pollution diffusion value corresponding to each spatial unit under each time index, it forms a virtual pollution distribution structure corresponding to the actual pollution state of the target area, which is used to characterize the overall evolution state of radioactive pollution in the continuous time change and spatial diffusion process.

[0114] The virtual pollution diffusion values ​​corresponding to each spatial unit under each time index are extracted from the pollution digital twin evolution field and arranged according to the time index order and the spatial unit distribution order to generate a virtual pollution spatiotemporal distribution sequence.

[0115] In this embodiment, the generation of the pollution source parameter vector set specifically includes:

[0116] Read the radioactivity intensity characterization value corresponding to each spatial unit in the pollution observation field matrix under each time index, and read the virtual pollution diffusion value corresponding to each spatial unit in the virtual pollution spatiotemporal distribution sequence under each time index, and align them to form the observation simulation corresponding data group;

[0117] For each set of observation simulation data, the difference between the radioactivity intensity characterization value and the virtual pollution diffusion value is calculated, and the data is organized according to the spatial unit and time index order to generate a set of observation simulation difference values.

[0118] Based on the variation of the difference of each spatial unit in the observation simulation difference set under the continuous time index, the feature region of continuous concentration of difference is extracted, and combined with the diffusion direction distribution results of the corresponding spatial unit in the virtual pollution spatiotemporal distribution sequence, the set of candidate pollution source spatial locations is determined.

[0119] The generation of the candidate pollution source spatial location set specifically includes: first, reading the differences corresponding to each spatial unit in the observation simulation difference set under all time indices, and forming a difference change sequence for each spatial unit according to the chronological order of the time indices; then, for each spatial unit, counting the consecutive positive difference segments in its difference change sequence to obtain the set of positive continuous difference segments corresponding to that spatial unit, and calculating the segment length value and the cumulative difference of each segment; from the set of positive continuous difference segments corresponding to each spatial unit, selecting positive continuous difference segments with segment length values ​​greater than a preset continuous length threshold and cumulative difference of segments greater than a preset cumulative difference threshold to obtain the set of continuous difference segments corresponding to each spatial unit; based on the spatial adjacency relationship between spatial units, performing connected region aggregation on spatial units with continuous difference segment sets to obtain a set of characteristic regions with continuous difference concentration; then reading each of the virtual pollution spatiotemporal distribution sequences... The virtual pollution diffusion values ​​of spatial units corresponding to the persistent concentration feature region of difference are calculated at each time index. For each spatial unit and its adjacent spatial units, their virtual pollution diffusion values ​​are compared at the same time index. When the virtual pollution diffusion value of a spatial unit is greater than that of its adjacent spatial units, the spatial unit is determined as the diffusion starting point spatial unit at the corresponding time index, thus obtaining the diffusion starting point determination results at each time index. Then, for each persistent concentration feature region of difference, the number of times each spatial unit in the region is determined as the diffusion starting point spatial unit is calculated, thus obtaining the starting point frequency value for each spatial unit. The spatial unit with the largest starting point frequency value is determined as the candidate pollution source spatial location corresponding to the persistent concentration feature region of difference, thus obtaining the candidate pollution source spatial location results. The candidate pollution source spatial location results corresponding to each persistent concentration feature region of difference are summarized to form a candidate pollution source spatial location set.

[0120] For each candidate pollution source spatial location in the candidate pollution source spatial location set, the radioactivity intensity characterization value and virtual pollution diffusion value of its corresponding spatial unit under each time index are read to generate the corresponding source intensity change sequence, and the release intensity parameter and release time parameter corresponding to each candidate pollution source spatial location are determined based on the source intensity change sequence.

[0121] The generation of release intensity and release time parameters specifically includes: for each candidate pollution source spatial location in the candidate pollution source spatial location set, determining its corresponding spatial unit, reading the radioactivity intensity characterization value and virtual pollution diffusion value corresponding to each spatial unit under each time index, forming a radioactivity intensity time series value set and a virtual pollution diffusion time series value set; then, according to the chronological order of the time indexes, subtracting each radioactivity intensity characterization value in the radioactivity intensity time series value set from each virtual pollution diffusion value in the virtual pollution diffusion time series value set to obtain the source strength difference sequence corresponding to each time index of the candidate pollution source spatial location; and then, analyzing the difference values ​​corresponding to each time index in the source strength difference sequence. The results are arranged sequentially to generate a source intensity variation sequence corresponding to the spatial location of the candidate pollution source. The largest difference result is extracted from the source intensity variation sequence to obtain the peak source intensity difference, which is then determined as the release intensity parameter corresponding to the spatial location of the candidate pollution source. Next, the difference results in the source intensity variation sequence are read in chronological order according to the time index to determine the time index of the first occurrence of the peak source intensity difference, which is then determined as the peak time index and determined as the release time parameter corresponding to the spatial location of the candidate pollution source. The release intensity parameter and release time parameter corresponding to the spatial location of each candidate pollution source are output to obtain a parameter result set that corresponds one-to-one with the spatial location of each candidate pollution source.

[0122] The spatial location parameters, release intensity parameters, and release time parameters corresponding to the spatial locations of each candidate pollution source are combined to form a set of pollution source hypotheses. Each pollution source hypothesis is then sequentially numbered to generate a pollution source parameter vector that corresponds one-to-one with each pollution source hypothesis.

[0123] All pollution source parameter vectors are organized according to the pollution source hypothesis numbering order to form a pollution source parameter vector set.

[0124] In this embodiment, the generation of the optimal pollution source parameter results specifically includes:

[0125] Read the set of pollution source parameter vectors and extract the corresponding spatial location parameters, release intensity parameters and release time parameters to form a set of candidate pollution source parameter groups;

[0126] Based on the candidate groups of each pollution source parameter, the virtual pollution spatiotemporal distribution sequence is iteratively adjusted to obtain the updated virtual pollution distribution results corresponding to each candidate group of pollution source parameter;

[0127] The generation of updated virtual pollution distribution results specifically includes: reading the spatial location parameters, release intensity parameters, and release time parameters from the single pollution source parameter candidate group to obtain the current pollution source parameter group; then reading the virtual pollution diffusion values ​​corresponding to each spatial unit in the virtual pollution spatiotemporal distribution sequence at each time index to obtain the basic virtual pollution distribution results; subsequently determining the corresponding action spatial unit based on the spatial location parameters to obtain the action spatial unit results, and determining the corresponding action time index based on the release time parameters to obtain the action time index results; reading the virtual pollution diffusion values ​​corresponding to the action spatial unit results at the action time index results to obtain the initial virtual pollution diffusion values; normalizing the initial virtual pollution diffusion values ​​to obtain the standardized diffusion values; normalizing the release intensity parameters to obtain the standardized release intensity values; weighting the standardized diffusion values ​​and standardized release intensity values ​​to obtain the initial adjusted pollution values; and finally writing the initial adjusted pollution values ​​into the action spatial unit results. If the result corresponds to the position under the time index, the initial updated pollution value is obtained. Then, according to the order of the time index, the current adjusted pollution value corresponding to each spatial unit under the previous time index is read to obtain the previous pollution value set. Then, the virtual pollution diffusion value corresponding to each spatial unit under the next time index is read to obtain the current basic pollution value set. Each virtual pollution diffusion value in the current basic pollution value set is normalized to obtain the standardized basic pollution value set. Each value in the previous pollution value set is normalized to obtain the standardized previous pollution value set. The corresponding values ​​in the standardized basic pollution value set and the standardized previous pollution value set are weighted and summed to obtain the current adjusted pollution value corresponding to each spatial unit under the next time index. The current adjusted pollution values ​​corresponding to each spatial unit under all time indices are organized according to the correspondence between spatial units and time indices to obtain the updated virtual pollution distribution result corresponding to the candidate group of pollution source parameters.

[0128] The updated virtual pollution distribution results are compared with the pollution observation field matrix. The observation simulation deviation value of each spatial unit under each time index is calculated, and the total deviation value corresponding to each pollution source parameter candidate group is generated.

[0129] The total deviation values ​​corresponding to each pollution source parameter candidate group are sorted, and the pollution source parameter candidate group with the smallest total deviation value is determined as the current optimal pollution source parameter candidate group, and the corresponding total deviation value is determined as the current optimal deviation value.

[0130] Based on the spatial location parameters, release intensity parameters, and release time parameters in the current optimal pollution source parameter candidate group, incremental and decremental adjustments are made to each parameter to generate a new set of pollution source parameter candidate groups. The virtual pollution spatiotemporal distribution sequence iterative adjustment and total deviation value calculation are repeated to obtain the parameter optimization results corresponding to each pollution source parameter candidate group in the new round.

[0131] Incremental and decremental adjustments to each parameter are made by taking the spatial location parameter, release intensity parameter, and release time parameter in the current optimal pollution source parameter candidate group as the center, and perturbing the parameters in the increasing and decreasing directions according to the preset step size, respectively, to generate a new round of pollution source parameter candidate group set;

[0132] The parameter optimization results corresponding to each pollution source parameter candidate group in the new round are compared. When the minimum total deviation value in the new round of parameter optimization results is less than the current optimal deviation value, the pollution source parameter candidate group corresponding to the minimum total deviation value is determined as the updated optimal pollution source parameter candidate group, and the minimum total deviation value is updated to the current optimal deviation value.

[0133] Repeat the process until the current optimal deviation value is less than the preset deviation threshold, then stop the iteration and determine the updated optimal pollution source parameter candidate group at the time of stopping the iteration as the optimal pollution source parameter result.

[0134] In this embodiment, the generation of ecological and environmental pollution risk assessment results specifically includes:

[0135] Read the spatial location parameters and release intensity parameters from the optimal pollution source parameter results, determine the initial spatial unit and the initial time index of the pollution source, and form a pollution source initial parameter group;

[0136] The generation of the pollution source initiation parameter set specifically includes: reading the spatial location parameters from the optimal pollution source parameter results to obtain the pollution source location parameter values, and matching the pollution source location parameter values ​​with the spatial unit set of the target area to determine the spatial units corresponding to the pollution source location parameter values, thus obtaining the pollution source initiation spatial units; then reading the release time parameters from the optimal pollution source parameter results to obtain the release time parameter values, and matching the release time parameter values ​​with the time index sequence to determine the time index corresponding to the release time parameter values, thus obtaining the pollution source initiation time index; and organizing the pollution source initiation spatial units and pollution source initiation time indices according to the parameter correspondence to generate the pollution source initiation parameter set.

[0137] Based on the pollution source initial parameter group, the pollution propagation sequence and direction between spatial units are determined according to the time index, and the pollution propagation path corresponding to the target area is generated.

[0138] Based on the transmission order of each spatial unit in the pollution propagation path, the pollution arrival time index and pollution duration index corresponding to each spatial unit are extracted to generate the pollution evolution time series corresponding to the target area.

[0139] Pollution transmission paths and pollution evolution time series are organized according to the correspondence between spatial units and time indices to generate pollution transmission process data;

[0140] Read the pollution arrival time index and pollution duration index corresponding to each spatial unit in the pollution propagation process data, and read the current adjusted pollution value corresponding to each spatial unit under each time index in the updated virtual pollution distribution result. Extract the maximum current adjusted pollution value corresponding to each spatial unit as the pollution intensity result, and calculate the radioactivity risk value corresponding to each spatial unit.

[0141] The generation of radioactivity risk values ​​specifically includes: first, reading the pollution arrival time index corresponding to each spatial unit in the pollution propagation process data to obtain the arrival time result for each spatial unit; then, reading the pollution duration index corresponding to each spatial unit to obtain the duration result for each spatial unit; subsequently, reading the current adjusted pollution value corresponding to each spatial unit under each time index in the updated virtual pollution distribution results, extracting the current adjusted pollution value sequence corresponding to each spatial unit under all time indices, and extracting the current adjusted pollution value with the largest value from the current adjusted pollution value sequence corresponding to each spatial unit to obtain the pollution intensity result for each spatial unit; normalizing the arrival time result for each spatial unit to obtain the arrival time risk value for each spatial unit, normalizing the duration result for each spatial unit to obtain the duration risk value for each spatial unit, and normalizing the pollution intensity result for each spatial unit to obtain the pollution intensity risk value for each spatial unit; weighting and summing the arrival time risk value, duration risk value, and pollution intensity risk value corresponding to the same spatial unit to obtain the comprehensive risk value for each spatial unit; and determining the comprehensive risk value for each spatial unit as the radioactivity risk value for each spatial unit.

[0142] Based on the correspondence between the radioactivity risk value of each space unit and the preset risk level classification threshold, the risk level of each space unit is classified, and the risk level result of each space unit is generated.

[0143] The radioactivity risk values ​​and risk levels corresponding to each spatial unit are organized according to the spatial unit distribution order to generate ecological and environmental pollution risk assessment results.

[0144] An ecological and environmental pollution risk assessment system includes:

[0145] The data acquisition module is used to collect radioactivity monitoring data of the target area, perform preprocessing, and form a standardized pollution observation dataset;

[0146] The pollution observation field construction module is used to construct a spatiotemporal distribution observation field of radioactive pollution in the target area and generate a pollution observation field matrix.

[0147] The coupled evolution model construction module is used to establish a radionuclide decay time evolution model and an environmental medium diffusion and migration model based on the pollution observation field matrix, forming a radioactive decay-diffusion coupled evolution model;

[0148] The pollution digital twin evolution module is used to construct a pollution digital twin evolution field based on the radioactive decay-diffusion coupled evolution model, perform simulation calculations, and generate a virtual pollution spatiotemporal distribution sequence.

[0149] The pollution source hypothesis construction module is used to construct a set of pollution source hypotheses based on the pollution observation field matrix and the virtual pollution spatiotemporal distribution sequence, and to form a set of pollution source parameter vectors.

[0150] The pollution source parameter optimization module is used to iteratively update and optimize the set of pollution source parameter vectors to generate the optimal pollution source parameter results.

[0151] The risk assessment results generation module is used to generate pollution propagation paths and pollution evolution time series in the target area, calculate the radioactivity risk value of each spatial unit in the target area, classify the risk level, and generate ecological and environmental pollution risk assessment results.

[0152] Example 1: To verify the feasibility of this invention in practice, it was applied to a radioactive pollution risk assessment scenario in the ecological buffer zone surrounding a riverside chemical industrial park and its downstream wetland connectivity area. This area is equipped with fixed radioactive monitoring stations, portable monitoring equipment, and several historical monitoring points. The monitoring area simultaneously covers the plant boundary, waterfront, agricultural land transition zone, and the edge of residential activity areas. Due to the large spatial area, complex surface media distribution, and significant temporal variations in pollution diffusion, existing manual interpretation methods or analysis based solely on instantaneous data from monitoring points often only reflect the pollution status at a local moment. This makes it difficult to accurately identify the location of pollution sources, the duration of release, and the direction of pollution propagation between different spatial units, leading to a lag in ecological and environmental risk assessment and failing to provide continuous, complete, and reliable evidence for subsequent management. This example addresses these problems by deploying the ecological and environmental pollution risk assessment method described in this invention in the daily monitoring and emergency analysis process of this area, forming a unified and interconnected processing chain for monitoring data, pollution evolution analysis, pollution source inversion, and risk assessment.

[0153] In practical applications, the system first collects radioactive monitoring data from the target area over a continuous monitoring period. This data includes nuclide activity concentration information, dose rate information, spatial coordinates of monitoring points, and corresponding sampling time information. The system performs time synchronization, missing data completion, anomaly removal, and data normalization on the acquired data to form a standardized pollution observation dataset. Subsequently, the target area is divided into spatial units based on the spatial coordinates of the monitoring points, and a time index is established based on continuous sampling times. Monitoring records from each monitoring point at each sampling time are mapped to the corresponding spatial unit and time index, generating a radioactive pollution spatiotemporal distribution observation field and a pollution observation field matrix. Users do not need to directly engage in complex calculations; they only need to import the monitoring records for the corresponding time period into the monitoring platform. The system automatically organizes and calculates the pollution intensity characterization values ​​for each spatial unit and transforms discrete monitoring records into a spatiotemporal matrix structure that continuously reflects the changing state of regional pollution. Based on this matrix, the system further establishes a nuclide decay time evolution model and an environmental media diffusion and migration model, uniformly coupling the natural decay process of nuclides with the migration process of pollution between adjacent spatial units to form a radioactive decay-diffusion coupled evolution model. Based on this, the system constructs a digital twin evolution field of pollution, recursively simulates the pollution diffusion state of the target area within a continuous time range, and generates a virtual spatiotemporal distribution sequence of pollution, enabling staff to view the evolution trend of pollution spreading from the source area to the surrounding area on the same interface.

[0154] Furthermore, to address the problem of accurate reverse identification of pollution sources in existing technologies, this embodiment compares the pollution observation field matrix with the virtual pollution spatiotemporal distribution sequence to form a set of differences between observation and simulation. Based on the relationship between the sustained concentration area of ​​the differences and the change in the diffusion direction, the spatial location of candidate pollution sources is determined, thereby generating a set of pollution source parameter vectors. Subsequently, the system continuously iterates and updates the parameters around the candidate spatial location, release intensity, and release time. By comparing the convergence of the deviation between the updated virtual pollution distribution results and the actual observation matrix, the optimal pollution source parameter results are gradually determined.

[0155] To verify the performance of the present invention, it was compared with the traditional method. The comparison results are shown in Table 1.

[0156] Table 1. Comparison of the overall performance of ecological and environmental pollution risk assessment methods

[0157] Method Category Pollution source location error (m) Release time identification error (h) Pollution transmission path matching degree (%) Accuracy rate of risk level classification (%) Recall rate in high-risk areas (%) Overall assessment duration (min) Traditional manual interpretation method 186.4 7.8 61.5 68.2 64.7 96 Methods based on statistical analysis of single monitoring data 142.7 6.1 69.8 74.6 72.3 78 Conventional evaluation methods based on diffusion models 97.3 4.5 81.2 83.4 80.6 64 Method of the present invention 38.6 1.7 93.8 94.9 92.7 41

[0158] As can be seen from Table 1, the method of the present invention is superior to the traditional method in all core indicators.

[0159] Regarding the pollution source location error, the method of this invention controls it to 38.6m, while traditional manual interpretation methods reach 186.4m, methods based on single monitoring data statistics reach 142.7m, and conventional assessment methods based on diffusion models still have an error of 97.3m. This result demonstrates that this invention does not simply rely on local values ​​of monitoring points for judgment, but rather transforms discrete monitoring information into continuous spatiotemporal correlated information through the combined effects of the pollution observation field matrix, the radioactive decay-diffusion coupled evolution model, and the pollution digital twin evolution field. Furthermore, it combines the difference between observed values ​​and the virtual pollution distribution to invert pollution source parameters, thus enabling more accurate location of the pollution source.

[0160] Regarding the error in identifying the release time, the method of this invention has an accuracy of 1.7 hours, significantly lower than the 7.8 hours of traditional manual interpretation methods, the 6.1 hours of methods based on single monitoring data statistics, and the 4.5 hours of conventional assessment methods based on diffusion models. This result corresponds to the technical approach described in the claims of this invention. This invention does not directly determine the time based on static monitoring results. Instead, it first constructs a nuclide decay time evolution model under a continuous time index, then combines this with diffusion and migration relationships to generate a virtual spatiotemporal distribution sequence of pollution. The release time parameters are continuously corrected during the iterative optimization of pollution source parameters, thus providing a more accurate identification of the pollution occurrence time. In other words, traditional methods mostly only show "pollution has increased in a certain area," while this invention can further determine "approximately when the pollution began to be released and how it gradually spread," thus significantly reducing the time identification error.

[0161] In terms of pollution propagation path matching accuracy, risk level classification accuracy, and high-risk area recall rate, the method of this invention achieved 93.8%, 94.9%, and 92.7% respectively, all higher than the comparative methods. In particular, compared with conventional assessment methods based on diffusion models, the pollution propagation path matching accuracy improved from 81.2% to 93.8%, the risk level classification accuracy improved from 83.4% to 94.9%, and the high-risk area recall rate improved from 80.6% to 92.7%. This indicates that the present invention is not only superior in pollution source identification but also has significant advantages in risk assessment. The fundamental reason for this difference lies in the fact that traditional methods often separate "pollution diffusion analysis" and "risk classification processing," lacking a unified data transmission link. In contrast, after obtaining the optimal pollution source parameters, the present invention further generates pollution propagation paths and pollution evolution time series, and calculates a comprehensive risk value by combining pollution arrival time, duration, and pollution intensity. Therefore, the final risk level classification results are more consistent with the actual pollution propagation patterns and have a stronger ability to identify high-risk spatial units.

[0162] In terms of overall assessment time, the method of this invention takes 41 minutes. Although the processing chain is more complete, its overall efficiency is still better than the traditional manual interpretation method (96 minutes), the method based on single monitoring data statistics (78 minutes), and the conventional assessment method based on diffusion models (64 minutes). This result shows that this invention does not sacrifice efficiency for improved accuracy. Instead, it achieves simultaneous improvement in assessment efficiency and accuracy through an integrated process of unified spatiotemporal unit organization, model coupling, parameter inversion, and risk output. This is because this invention incorporates monitoring data preprocessing, pollution observation field construction, coupled evolution calculation, pollution source parameter optimization, and risk assessment result generation into a single processing framework. This reduces the time consumed by multiple manual switching analysis steps, repeated verification, and subsequent corrections in traditional methods, thus resulting in a greater advantage in overall time.

[0163] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for assessing ecological and environmental pollution risks, characterized in that, Includes the following steps: Collect radioactive monitoring data from the target area, perform preprocessing, and generate a standardized pollution observation dataset. Based on a standardized pollution observation dataset, a spatiotemporal distribution observation field of radioactive pollution in the target area is constructed, and a pollution observation field matrix is ​​generated. Based on the pollution observation field matrix, a radionuclide decay time evolution model and an environmental medium diffusion and migration model are established, and the radionuclide decay process and diffusion and migration process are coupled to form a radioactive decay-diffusion coupled evolution model. Based on the radioactive decay-diffusion coupled evolution model, a pollution digital twin evolution field is constructed to simulate and calculate the pollution diffusion state of the target area in a continuous time series, generating a virtual pollution spatiotemporal distribution sequence. Based on the pollution observation field matrix and the virtual pollution spatiotemporal distribution sequence, a set of pollution source hypotheses is constructed, and a set of pollution source parameter vectors is formed. The set of pollution source parameter vectors is iteratively updated and optimized to generate the optimal pollution source parameter results. Based on the optimal pollution source parameters, the pollution propagation path and pollution evolution time series of the target area are generated, the radioactivity risk value of each spatial unit in the target area is calculated, the risk level is classified, and the ecological and environmental pollution risk assessment results are generated. The construction of the radioactive decay-diffusion coupled evolution model specifically includes: Read the radioactivity intensity characterization values ​​of each spatial unit in the pollution observation field matrix under each time index, extract the radioactivity intensity change sequence corresponding to each spatial unit under continuous time index, and determine the nuclide decay trend value corresponding to each spatial unit; Based on the nuclide decay trend value corresponding to each spatial unit, the nuclide decay transfer relationship between each time index is established, a nuclide decay time evolution model for continuous time index is generated, and the nuclide decay evolution value corresponding to each spatial unit under each time index is obtained. By combining the spatial distribution relationship of each spatial unit in the pollution observation field matrix, the diffusion and migration correlation between adjacent spatial units is established. Based on the radioactivity intensity change sequence of each spatial unit under continuous time index, spatial migration and transfer paths are generated to form an environmental medium diffusion and migration model, and the corresponding diffusion and migration values ​​between each spatial unit are obtained. The nuclide decay evolution values ​​in the nuclide decay time evolution model and the diffusion migration values ​​in the environmental medium diffusion migration model are correlated and mapped according to the same time index and the same spatial unit to form the joint decay migration evolution value corresponding to each spatial unit under each time index. A radioactive decay-diffusion coupled evolution model is constructed based on the decay-migration joint evolution values ​​of each spatial unit at each time index.

2. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The radioactivity monitoring data includes nuclide activity concentration and dose rate.

3. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The preprocessing includes time synchronization, missing data completion, anomaly removal, and data normalization.

4. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The generation of the pollution observation field matrix specifically includes: Read the standardized pollution observation dataset, divide the target area into spatial units according to the spatial coordinates of each monitoring point in the target area, obtain a set of spatial units, and establish a corresponding time index based on the sampling time; The radionuclide activity concentration data and dose rate data of each monitoring point at each sampling time are mapped to the corresponding spatial unit and the corresponding time index, respectively, to form the radioactive observation data entries of each spatial unit under each time index; Extract all radioactive observation data entries under the same time index for the same spatial unit, and count the number of radioactive observation data entries. When the number of radioactive observation data entries is greater than 1, the nuclide activity concentration data and dose rate data in all radioactive observation data entries are aggregated to obtain the nuclide activity concentration aggregated value and dose rate aggregated value, and the radioactivity intensity characterization value of the corresponding spatial unit under the corresponding time index is generated. When the number of radioactive observation data entries is equal to 1, the radionuclide activity concentration data and dose rate data in the radioactive observation data entry are read, and the radioactivity intensity characterization value of the corresponding spatial unit under the corresponding time index is generated. The radioactivity intensity characterization values ​​corresponding to each spatial unit under each time index are organized according to the correspondence between spatial units and time indices to construct a spatiotemporal distribution observation field for radioactive contamination. Using spatial units as row indices and time indices as column indices, the radioactivity intensity characterization values ​​of each spatial unit in the spatiotemporal distribution observation field of radioactive contamination under each time index are written into the corresponding matrix positions to generate the contamination observation field matrix.

5. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The generation of the virtual pollution spatiotemporal distribution sequence specifically includes: Read the decay migration joint evolution values ​​of each spatial unit in the radioactive decay-diffusion coupled evolution model at each time index, and arrange them according to the correspondence between spatial units and time indices to form a joint evolution value sequence corresponding to the target region; Based on the joint evolution value sequence, virtual pollution mapping units corresponding to the target area are constructed according to the spatial coordinates and time indices of the spatial units. The decay and migration joint evolution values ​​of each spatial unit under each time index are written into the corresponding virtual pollution mapping unit to obtain the pollution digital twin basic unit set. The values ​​of each virtual pollution mapping unit in the basic unit set of pollution digital twin are continuously correlated in time to form a virtual pollution evolution chain corresponding to each spatial unit under continuous time index, and spatial adjacency correlation is performed on each virtual pollution evolution chain to generate a virtual pollution evolution correlation network. Based on the virtual pollution evolution correlation network, the pollution diffusion state of each spatial unit under each time index is recursively simulated to obtain the virtual pollution diffusion value corresponding to each spatial unit under each time index. The values ​​are then organized according to the time index order to form the virtual pollution time series distribution results corresponding to each spatial unit. The virtual pollution time series distribution results corresponding to each spatial unit are integrated according to the correspondence between spatial units and time indices to construct a pollution digital twin evolution field; The virtual pollution diffusion values ​​corresponding to each spatial unit under each time index are extracted from the pollution digital twin evolution field and arranged according to the time index order and the spatial unit distribution order to generate a virtual pollution spatiotemporal distribution sequence.

6. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The generation of the pollution source parameter vector set specifically includes: Read the radioactivity intensity characterization value corresponding to each spatial unit in the pollution observation field matrix under each time index, and read the virtual pollution diffusion value corresponding to each spatial unit in the virtual pollution spatiotemporal distribution sequence under each time index, and align them to form the observation simulation corresponding data group; For each set of observation simulation data, the difference between the radioactivity intensity characterization value and the virtual pollution diffusion value is calculated, and the data is organized according to the spatial unit and time index order to generate a set of observation simulation difference values. Based on the variation of the difference of each spatial unit in the observation simulation difference set under the continuous time index, the feature region of continuous concentration of difference is extracted, and combined with the diffusion direction distribution results of the corresponding spatial unit in the virtual pollution spatiotemporal distribution sequence, the set of candidate pollution source spatial locations is determined. For each candidate pollution source spatial location in the candidate pollution source spatial location set, the radioactivity intensity characterization value and virtual pollution diffusion value of its corresponding spatial unit under each time index are read to generate the corresponding source intensity change sequence, and the release intensity parameter and release time parameter corresponding to each candidate pollution source spatial location are determined based on the source intensity change sequence. The spatial location parameters, release intensity parameters, and release time parameters corresponding to the spatial locations of each candidate pollution source are combined to form a set of pollution source hypotheses. Each pollution source hypothesis is then sequentially numbered to generate a pollution source parameter vector. All pollution source parameter vectors are organized according to the pollution source hypothesis numbering order to form a pollution source parameter vector set.

7. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The generation of the optimal pollution source parameter results specifically includes: Read the set of pollution source parameter vectors and extract the corresponding spatial location parameters, release intensity parameters and release time parameters to form a set of candidate pollution source parameter groups; Based on the candidate groups of each pollution source parameter, the virtual pollution spatiotemporal distribution sequence is iteratively adjusted to obtain the updated virtual pollution distribution results corresponding to each candidate group of pollution source parameter; The updated virtual pollution distribution results are compared with the pollution observation field matrix. The observation simulation deviation value of each spatial unit under each time index is calculated, and the total deviation value corresponding to each pollution source parameter candidate group is generated. The total deviation values ​​corresponding to each pollution source parameter candidate group are sorted, and the pollution source parameter candidate group with the smallest total deviation value is determined as the current optimal pollution source parameter candidate group, and the corresponding total deviation value is determined as the current optimal deviation value. Based on the spatial location parameters, release intensity parameters, and release time parameters in the current optimal pollution source parameter candidate group, incremental and decremental adjustments are made to each parameter to generate a new set of pollution source parameter candidate groups. The virtual pollution spatiotemporal distribution sequence iterative adjustment and total deviation value calculation are repeated to obtain the parameter optimization results corresponding to each pollution source parameter candidate group in the new round. The parameter optimization results corresponding to each pollution source parameter candidate group in the new round are compared. When the minimum total deviation value in the new round of parameter optimization results is less than the current optimal deviation value, the pollution source parameter candidate group corresponding to the minimum total deviation value is determined as the updated optimal pollution source parameter candidate group, and the minimum total deviation value is updated to the current optimal deviation value. Repeat the process until the current optimal deviation value is less than the preset deviation threshold, then stop the iteration and determine the updated optimal pollution source parameter candidate group at the time of stopping the iteration as the optimal pollution source parameter result.

8. The method for assessing ecological and environmental pollution risk according to claim 1, characterized in that, The generation of the ecological and environmental pollution risk assessment results specifically includes: Read the spatial location parameters and release intensity parameters from the optimal pollution source parameter results, determine the initial spatial unit and the initial time index of the pollution source, and form a pollution source initial parameter group; Based on the pollution source initial parameter group, the pollution propagation sequence and direction between spatial units are determined according to the time index, and the pollution propagation path corresponding to the target area is generated. Based on the transmission order of each spatial unit in the pollution propagation path, the pollution arrival time index and pollution duration index corresponding to each spatial unit are extracted to generate the pollution evolution time series corresponding to the target area. Pollution transmission paths and pollution evolution time series are organized according to the correspondence between spatial units and time indices to generate pollution transmission process data; Read the pollution arrival time index and pollution duration index corresponding to each spatial unit in the pollution propagation process data, and read the current adjusted pollution value corresponding to each spatial unit under each time index in the updated virtual pollution distribution result. Extract the maximum current adjusted pollution value corresponding to each spatial unit as the pollution intensity result, and calculate the radioactivity risk value corresponding to each spatial unit. Based on the correspondence between the radioactivity risk value of each space unit and the preset risk level classification threshold, the risk level of each space unit is classified, and the risk level result of each space unit is generated. The radioactivity risk values ​​and risk levels corresponding to each spatial unit are organized according to the spatial unit distribution order to generate ecological and environmental pollution risk assessment results.

9. An ecological environment pollution risk assessment system, implementing the ecological environment pollution risk assessment method according to any one of claims 1 to 8, characterized in that, include: The data acquisition module is used to collect radioactivity monitoring data of the target area, perform preprocessing, and form a standardized pollution observation dataset; The pollution observation field construction module is used to construct a spatiotemporal distribution observation field of radioactive pollution in the target area and generate a pollution observation field matrix. The coupled evolution model construction module is used to establish a radionuclide decay time evolution model and an environmental medium diffusion and migration model based on the pollution observation field matrix, forming a radioactive decay-diffusion coupled evolution model; The pollution digital twin evolution module is used to construct a pollution digital twin evolution field based on the radioactive decay-diffusion coupled evolution model, perform simulation calculations, and generate a virtual pollution spatiotemporal distribution sequence. The pollution source hypothesis construction module is used to construct a set of pollution source hypotheses based on the pollution observation field matrix and the virtual pollution spatiotemporal distribution sequence, and to form a set of pollution source parameter vectors. The pollution source parameter optimization module is used to iteratively update and optimize the set of pollution source parameter vectors to generate the optimal pollution source parameter results. The risk assessment results generation module is used to generate pollution propagation paths and pollution evolution time series in the target area, calculate the radioactivity risk value of each spatial unit in the target area, classify the risk level, and generate ecological and environmental pollution risk assessment results.

Images