Regional environment radiation prediction method and system based on multi-source data fusion

Abstract

The invention discloses a regional environment radiation prediction method and system based on multi-source data fusion, and belongs to the technical field of radiation prediction. The method comprises the steps that after a prediction instruction is received, a system determines a prediction time period and a to-be-detected area, multi-source environment monitoring data such as radiation, meteorology and operation activities of the area within a preset time period are acquired, and preprocessing is carried out to form a multi-source environment data set; and generating fusion adjustment condition data based on a preset sampling time point and a corresponding abnormal event detection result, then calculating a fusion weight of each piece of monitoring data and performing weighted fusion to obtain fusion radiation feature data, and finally inputting the fusion radiation feature data and a prediction time period into the radiation prediction model to obtain an environmental radiation prediction result of the to-be-detected region in the prediction time period. According to the scheme, the radiation monitoring data, the meteorological observation data and the regional operation activity data are fused, the abnormal event information is combined, self-adaptive weighted fusion is carried out on the multi-source environment data, and the accuracy and reliability of environment radiation prediction are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of radiation prediction, and particularly relates to a regional environment radiation prediction method and system based on multi-source data fusion. BACKGROUND

[0002] In the treatment of radioactive substances, accurate prediction of the environmental radiation level of the to-be-tested region is crucial to safety. Environmental radiation is affected by multiple factors and changes dynamically and complexly. Traditional single or simple combined analysis methods cannot fully capture this complexity, have large prediction errors, and cannot meet the high-precision requirements. Therefore, it is urgent to develop an accurate prediction method that comprehensively integrates multi-source data and adaptively fuses.

[0003] When environmental radiation prediction is carried out today, personnel are arranged to hold simple instruments at designated locations to collect radiation, meteorological and other data. Because the sensor points are sparse, only local and scattered information can be obtained. Then, the collected data is preliminarily screened to eliminate obviously unreasonable data without detailed time sequence alignment during the process. Subsequently, the screened data of various types are simply listed and summarized. Finally, based on the summarized data, the environmental radiation prediction of the to-be-tested region is directly inferred by experience.

[0004] However, manual collection of data at sparse points using simple instruments is scattered and incomplete. In addition, the time sequence is not aligned in detail during preliminary data screening, which destroys the data association. After simply listing and summarizing the data, the prediction result is inferred only by experience, which lacks scientificity and accuracy, is difficult to adapt to environmental dynamic changes, and results in low reliability of the prediction result. SUMMARY

[0005] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the technical problems of the prior art, that is, the information is scattered and incomplete by manually collecting data at sparse points using simple instruments. In addition, the time sequence is not aligned in detail during preliminary data screening, which destroys the data association. After simply listing and summarizing the data, the prediction result is inferred only by experience, which lacks scientificity and accuracy, is difficult to adapt to environmental dynamic changes, and results in low reliability of the prediction result.

[0006] In a first aspect, the present application provides a regional environment radiation prediction method based on multi-source data fusion, comprising: If a prediction instruction is received, the prediction period and the to-be-tested region are determined based on the prediction instruction, the multi-source environmental monitoring data of the to-be-tested region within a preset time length is acquired, the multi-source environmental monitoring data is preprocessed, and a multi-source environmental data set is obtained. The multi-source environmental monitoring data includes radiation monitoring data, meteorological observation data, and regional operation activity data. Extract each preset sampling time point within a preset time period, and obtain the abnormal event detection results corresponding to each preset sampling time point in the area to be tested. Generate fusion adjustment condition data for the area to be tested based on each preset sampling time point and the abnormal event detection results. Based on the multi-source environmental dataset and the fusion adjustment condition data, the fusion weight is calculated for each multi-source environmental monitoring data in the multi-source environmental dataset. Based on the fusion weight, the multi-source environmental monitoring data in the multi-source environmental dataset are weighted and fused to obtain fused radiation characteristic data. The fused radiation characteristic data and the prediction time period are input into the preset radiation prediction model to obtain the environmental radiation prediction result of the area to be measured during the prediction time period.

[0007] In a second aspect, the present invention provides a regional environmental radiation prediction system based on multi-source data fusion, the system comprising: The data acquisition module is used to, upon receiving a prediction instruction, determine the prediction period and the area to be measured based on the prediction instruction, acquire multi-source environmental monitoring data of the area to be measured within a preset time period, preprocess the multi-source environmental monitoring data, and obtain a multi-source environmental dataset; wherein, the multi-source environmental monitoring data includes radiation monitoring data, meteorological observation data, and regional operation activity data; The fusion adjustment condition generation module is used to extract each preset sampling time point within a preset time period, and to obtain the abnormal event detection results corresponding to each preset sampling time point in the area to be tested, and to generate fusion adjustment condition data for the area to be tested based on each preset sampling time point and the abnormal event detection results. The fusion weight calculation module is used to calculate the fusion weight of each multi-source environmental monitoring data in the multi-source environmental dataset based on the multi-source environmental dataset and the fusion adjustment condition data. The prediction module is used to perform weighted fusion of the multi-source environmental monitoring data in the multi-source environmental dataset based on the fusion weight to obtain fused radiation characteristic data. The fused radiation characteristic data and the prediction time period are input into the preset radiation prediction model to obtain the environmental radiation prediction result of the area to be measured during the prediction time period.

[0008] In a third aspect, an electronic device is provided, comprising a processor, a memory, and a program or instructions stored in the memory and executable on the processor, the program or instructions being loaded and run by the processor to perform the steps of the aforementioned method for predicting regional environmental radiation based on multi-source data fusion.

[0009] In a fourth aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored therein, the program codes being adapted to be loaded and run by a processor to perform the steps of the above-described method for predicting regional environmental radiation based on multi-source data fusion.

[0010] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects: In implementing the technical solution of this invention, by integrating radiation monitoring data, meteorological observation data, and regional operation activity data, and combining abnormal event information, adaptive weighted fusion of multi-source environmental data is performed, thereby improving the accuracy and reliability of environmental radiation prediction. Attached Figure Description

[0011] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Furthermore, similar numbers in the drawings are used to denote similar components, wherein: Figure 1 This is a schematic diagram of the main steps of a regional environmental radiation prediction method based on multi-source data fusion according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the main steps of a regional environmental radiation prediction method based on multi-source data fusion according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the main structure of a regional environmental radiation prediction system based on multi-source data fusion according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0012] Some embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0013] In the description of this invention, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and may also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.

[0014] See appendix Figure 1 , Figure 1 This is a schematic flowchart illustrating the main steps of a regional environmental radiation prediction method based on multi-source data fusion according to an embodiment of the present invention. Figure 1 As shown, a regional environmental radiation prediction method based on multi-source data fusion in an embodiment of the present invention mainly includes the following steps S101-S104.

[0015] Step S101: If a prediction instruction is received, the prediction period and the area to be measured are determined based on the prediction instruction. Multi-source environmental monitoring data of the area to be measured within a preset time period are obtained. The multi-source environmental monitoring data are preprocessed to obtain a multi-source environmental dataset. The multi-source environmental monitoring data includes radiation monitoring data, meteorological observation data, and regional operation activity data.

[0016] The prediction command is a control signal or command used to initiate the environmental radiation prediction process. Its core function is to officially start the prediction work.

[0017] The forecast period is the future time interval covered by the forecast results. Its main function is to define the start and end time range of the model's forecast, ensuring that the environmental radiation level data within that specific time period can be accurately output. Common settings include 1 hour or 2 hours in the future.

[0018] The area to be measured is the geographical area or monitoring area where radiation prediction needs to be carried out. Its role is to define the scope of data processing and prediction calculation for the system.

[0019] The preset duration is the length of the time window used for real-time collection of multi-source environmental monitoring data after receiving the prediction command. For example, data collected within 30 minutes can be selected as the input.

[0020] Multi-source environmental monitoring data are various types of raw environmental data used for radiation prediction.

[0021] A multi-source environmental dataset is a collection of data formed by integrating multi-source environmental monitoring data after a series of preprocessing steps.

[0022] Radiation monitoring data is environmental radiation intensity data collected in real time or stored historically by monitoring equipment. Its function is to directly reflect the radiation level in the target area.

[0023] Meteorological observation data refers to various meteorological information collected for a target area, including multiple indicators such as wind speed, wind direction, temperature, and humidity. Its role is to assist in the analysis of the diffusion patterns of radiation in the environment and the impact of environmental factors on radiation.

[0024] Regional operational activity data records information related to operations or activities that may affect environmental radiation within a target area. Its purpose is to help predictive models fully consider the impact of human activities on radiation changes. For example, activities such as the handling of radioactive materials within the area and the operating status of related equipment are recorded.

[0025] Upon receiving the prediction instruction, the system first parses the instruction to determine the target area and prediction time period. Then, within a preset timeframe, it completes the multi-source environmental monitoring data collection for the area to be monitored. For radiation monitoring data, each radiation sensor node continuously measures the radiation intensity within the area at a fixed sampling frequency, while simultaneously recording timestamps. Considering that the raw radiation signals may contain noise or outliers, the data undergoes filtering processing, such as using moving averages or median filtering to remove transient interference. Outliers are also identified and corrected to ensure the continuity and accuracy of the radiation monitoring data.

[0026] For meteorological observation data, after acquiring environmental parameters such as wind speed, wind direction, temperature, and humidity from meteorological sensors, the timestamps of the meteorological information at each sampling point are aligned with those of the radiation monitoring data to achieve time synchronization. If the data collected by the sensors is missing or abnormal, it is supplemented or corrected through interpolation or regression methods to ensure the integrity and consistency of the meteorological observation data.

[0027] Regarding the collection of regional operational activity data, after acquiring operational plans, equipment operating status, and operational event records within the region, the time, location, and type of operational activities are standardized and coded to facilitate subsequent matching with environmental data. If there are time delays or missing data in the operational data, time series alignment and missing value imputation techniques will be used to ensure that the regional operational activity data corresponds accurately in time with radiation monitoring data and meteorological observation data. After completing the cleaning, anomaly handling, time synchronization, and standardized coding of various types of data, radiation monitoring data, meteorological observation data, and regional operational activity data are integrated in a unified format to form a multi-source environmental dataset, ensuring that each record contains radiation values, meteorological parameters, and operational activity information at the same point in time.

[0028] Step S102: Extract each preset sampling time point within a preset time period, and obtain the abnormal event detection results corresponding to each preset sampling time point in the area to be tested. Generate fusion adjustment condition data for the area to be tested based on each preset sampling time point and the abnormal event detection results.

[0029] The preset sampling time point is a time node selected within a preset duration according to a fixed frequency or preset rules, to ensure that the environmental state can be sampled regularly throughout the entire prediction time window.

[0030] Abnormal event detection results are the results of detection of non-radiation related emergencies or abnormal situations. They can determine whether the environment or operation is in an abnormal state. Specifically, this includes equipment failure or abnormal operation, abnormal operation process such as unplanned operation, operation exceeding the limit in the area, and sudden weather changes such as rainstorms and strong winds.

[0031] The fusion adjustment condition data is a dataset generated based on each preset sampling time point and the corresponding anomaly event detection results. It is used to guide the adaptive calculation of subsequent multi-source data fusion weights, ensuring that the contribution of each data source can be dynamically adjusted according to anomaly events during multi-source data fusion.

[0032] After determining the preset duration before forecasting, a continuous time series index is first established within this time period, and then multiple sampling time points are divided according to the preset sampling frequency. The setting of the sampling frequency takes into account both the environmental data update rate and the signal stability period. For example, when the update period for meteorological and operational activity data is 5 minutes, the sampling frequency can be set to 1 minute to ensure that the time resolution is sufficient to capture local abrupt changes. The system will automatically generate a series of equally spaced sampling time point sets within the preset duration according to this sampling frequency. These time points will serve as time anchors for subsequent anomaly detection and fusion adjustment.

[0033] After determining each sampling time point, multi-source environmental state data corresponding to that time moment is extracted from historical records for each time point, while simultaneously loading the anomaly event detection results occurring at that moment. The anomaly event detection results are generated from the historical monitoring stream by an independent anomaly identification algorithm. The content does not include any radiation-related indicators and is specifically designed to identify external interference factors affecting environmental stability. These anomalies include communication interruptions of monitoring equipment, sensor drift, sampling delays, sudden construction or traffic activities within the area, and short-term extreme weather phenomena such as localized strong winds or rainfall. The anomaly detection algorithm combines a multi-threshold discrimination method based on time-series features with a pattern matching strategy. For example, if a certain type of meteorological factor experiences a sudden change exceeding a multiple of the historical standard deviation within a very short period, it is determined to be a sudden meteorological anomaly; if the equipment output signal exhibits periodic drift and its similarity to the standard signal decreases sharply, it is determined to be an equipment anomaly.

[0034] After extracting abnormal events, the system matches the anomaly detection results at each sampling time point with the corresponding environmental state parameters to form a time-event alignment table. For each record, the system calculates the impact intensity coefficient of the abnormal event, based on factors including the type of abnormal event, its duration, and the spatial coverage of its impact area. For example, a local equipment failure affecting only a single data point results in a lower impact intensity, while a regional operational activity affecting multiple observation points results in a higher impact intensity. The impact intensity is then mapped to an anomaly response coefficient to characterize the degree to which the abnormal event weakens the reliability of the environmental data.

[0035] After calculating the anomaly response coefficient, fluctuation characteristic analysis is performed on environmental data from all sampling time points within a preset time period. By applying a sliding window to the time-series variation curves of each parameter, such as temperature, humidity, wind speed, and wind direction, transient change rate, data stability index, and trend consistency score are calculated. Time periods with higher stability and smaller change rates correspond to larger dynamic weight indices, indicating higher reliability of the monitoring data for that period; conversely, if data fluctuations are frequent or affected by external events, the dynamic weight indices will decrease.

[0036] The dynamic weight indexes and anomaly response coefficients at each sampling time point are fused over time using a fusion algorithm based on weighted moving average and correlation correction. This approach avoids over-amplification of the sudden impact of anomalies within a short period while preserving their actual impact on data confidence. Finally, a fusion adjustment record containing core parameters is generated at each sampling time point, specifically including a time stamp, dynamic weight index, and anomaly response coefficient. All fusion adjustment records from all sampling time points are then integrated chronologically to construct complete fusion adjustment condition data.

[0037] Based on the above technical solution, optionally, fusion adjustment condition data for the area to be tested can be generated based on each preset sampling time point, regional environmental state information, and abnormal event detection results, including: Based on each preset sampling time point and the corresponding abnormal event detection results, generate each multi-source monitoring data unit; The dynamic weight index of each multi-source monitoring data unit in the time dimension is determined based on each preset sampling time point; The abnormal response coefficient of each multi-source monitoring data unit is determined based on the abnormal event detection results; Fusion adjustment condition data for the test region is generated based on dynamic weight index and abnormal response coefficient.

[0038] In this scheme, the multi-source monitoring data unit is a data unit formed by combining the detection results of abnormal events corresponding to each preset sampling time point with the time node.

[0039] The dynamic weight index is a weight index assigned to each multi-source monitoring data unit based on the preset sampling time point and time series characteristics.

[0040] The anomaly response coefficient is a numerical value calculated based on the anomaly event detection results corresponding to the multi-source monitoring data unit. It is used to reflect the degree of influence of anomalies on the weight of samples in the fusion calculation.

[0041] Fusion conditioning data is a dataset that integrates the dynamic weight index and abnormal response coefficients of samples to guide the adaptive weighted fusion of multi-source environmental data.

[0042] Each sampling time point and its corresponding anomaly detection result are mapped to a sample, i.e., a multi-source monitoring data unit. This unit contains a time stamp and anomaly event information. Based on this, a dynamic weight index is determined along the time dimension for each multi-source monitoring data unit according to its position in the time series. The dynamic weight index calculates the importance weight of each time point relative to the entire sampling interval by standardizing the time series. For example, higher weights are assigned to the most recent time point, and decreasing weights are assigned to historical time points, thus reflecting the degree to which samples change over time and influence subsequent data fusion.

[0043] For each multi-source monitoring data unit, the corresponding abnormal event detection results are analyzed, and anomaly response coefficients are calculated. These coefficients are quantified based on the type, severity, and duration of the abnormal event. For example, higher response coefficients are assigned to major equipment failures or sudden weather events, while lower coefficients are assigned to minor operational deviations. This information can be used to adjust the sample's impact on the fusion weights during data fusion. Finally, the dynamic weight index and anomaly response coefficients of each multi-source monitoring data unit are integrated to form fusion adjustment condition data. This dataset stores the dynamic weights and anomaly response information for each sampling time point in a structured format, providing a basis for subsequent adaptive weighted fusion of multi-source environmental datasets.

[0044] In this scheme, by combining sampling time points and abnormal event information to generate fused adjustment condition data, the weights of multi-source monitoring data can be adaptively adjusted during fusion, thereby improving the accuracy and reliability of environmental radiation prediction.

[0045] Step S103: Based on the multi-source environmental dataset and the fusion adjustment condition data, calculate the fusion weight for each multi-source environmental monitoring data in the multi-source environmental dataset.

[0046] Fusion weight is a measure of the importance and reliability of environmental monitoring data from different sources in the overall feature fusion.

[0047] After acquiring the multi-source environmental dataset and the fused conditioning condition data, a unified format standardization and time alignment process is performed on various monitoring data in the multi-source environmental dataset, transforming records with different sampling frequencies and times into a unified time series format. For radiation monitoring data, meteorological observation data, and regional operational activity data, fluctuation amplitude, trend, and data continuity indicators are calculated within a preset time period to characterize the temporal stability and sensitivity to change of each data source. Dynamic weight indexes and anomaly response coefficients are extracted from the fused conditioning condition data; the former reflects the dynamic importance of each monitoring data point in time-series changes, while the latter reflects the reliability of each data source when affected by abnormal events.

[0048] Based on a dynamic weighted index, basic weights are calculated for each monitoring data point in the multi-source environmental dataset. Specifically, data with higher dynamic weight indices indicate stable signals and a strong correlation with environmental changes during that period, resulting in a higher basic weight; conversely, data with drastic fluctuations or unstable trends will have a lower basic weight. Subsequently, the anomaly response coefficient is incorporated into the basic weight correction process, adjusting the weights of each monitoring data point according to the degree of impact of abnormal events: a low anomaly response coefficient indicates poor reliability of the data under abnormal conditions, leading to a lower weight; a coefficient close to 1 indicates that the data has not been significantly disturbed by abnormal events, and the weight remains stable.

[0049] After comprehensive adjustment using dynamic weight indexing and anomaly response coefficients, the weighted results of each monitoring data are normalized to ensure that the sum of fused weights from different sources is on a uniform scale. Then, a time-sliding window and adaptive smoothing algorithm are used to dynamically smooth the weight change sequence, reducing weight fluctuations caused by instantaneous disturbances and maintaining the continuity and stability of the fused weights. Finally, a set of fused weight sequences corresponding to different monitoring data is generated, reflecting the real-time reliability and dynamic contribution of each monitoring data point in the current time period.

[0050] Step S104: Based on the fusion weight, perform weighted fusion of the multi-source environmental monitoring data in the multi-source environmental dataset to obtain fused radiation characteristic data. Input the fused radiation characteristic data and the prediction time period into the preset radiation prediction model to obtain the environmental radiation prediction result of the area to be measured during the prediction time period.

[0051] Fusion radiation characteristic data is a unified value formed by weighted fusion of multi-source environmental monitoring data. It represents the comprehensive environmental radiation level or characteristic index of a specific time and region, such as the comprehensive radiation intensity index or environmental radiation risk score.

[0052] The preset radiation prediction model is a computational model that uses historical data and fused radiation characteristic data to predict environmental radiation changes in the future.

[0053] Environmental radiation prediction results are information on the radiation level of the area to be measured in the spatial and temporal dimensions during the prediction period, and are generally presented in the form of raster data or feature vectors.

[0054] Based on the acquired multi-source environmental dataset and its corresponding fusion weights, weighted processing is performed on each type of environmental monitoring data according to its weight. Specifically, at each time point or sampling point, radiation monitoring data, meteorological observation data, and regional operational activity data are multiplied by their respective fusion weights. The weighted results are then accumulated and integrated in both spatial and temporal dimensions to generate unified fused radiation characteristic data, ensuring that data from different sources contribute reasonably to the prediction based on their own importance and reliability. This fused radiation characteristic data, along with the defined prediction period information, is input into a pre-set radiation prediction model. After model processing, the environmental radiation prediction results for the area under test during the prediction period are output, covering the radiation value change trends of each sampling point or region during the prediction period.

[0055] The training process for the pre-defined radiation prediction model includes: Historically collected multi-source environmental data, after cleaning, standardization, and fusion processing, forms a historical fused radiation feature dataset. This feature data includes weighted combinations of radiation monitoring data, meteorological observation data, and operational activity data at different time points, while preserving the spatial and temporal correlation information of each data source. Each sample's input consists of fused radiation feature data and corresponding time information, describing the relationship between historical environmental conditions and temporal evolution; the corresponding label is the historically observed environmental radiation value, i.e., the actual measured radiation level at that time point. During training, the model learns the spatiotemporal coupling patterns, dynamic weight influences, and environmental change patterns among multi-source data by fitting the relationship between the input features and actual observations. This allows the model to output the radiation level change trend of the region during the prediction period, obtained by arranging environmental radiation values ​​in a time series, after inputting the current fused radiation feature data and a specified prediction period into the model during the prediction phase—that is, the environmental radiation prediction result.

[0056] Based on steps S101-S104 above, by integrating radiation monitoring data, meteorological observation data, and regional operational activity data, and combining abnormal event information, adaptive weighted fusion of multi-source environmental data is performed, which improves the accuracy and reliability of environmental radiation prediction.

[0057] Based on the above technical solution, optionally, the multi-source environmental monitoring data in the multi-source environmental dataset are weighted and fused according to the fusion weight to obtain fused radiation characteristic data, including: The multi-source environmental monitoring data in the multi-source environmental dataset are weighted and fused using a preset weighted fusion formula and fusion weights to obtain fused radiation characteristic data; wherein the preset weighted fusion formula is: ; in, To fuse radiation characteristic data; This refers to the total amount of multi-source environmental monitoring data. An index for multi-source environmental monitoring data; Weights for the fusion of multi-source environmental monitoring data; These are data values ​​from multi-source environmental monitoring data.

[0058] In this scheme, the data value is the numerical value of multi-source environmental monitoring data observed or processed at a specific sampling time point. Specifically, for radiation monitoring data, the data value is the measured environmental radiation intensity or dose value; for meteorological observation data, the data value can be indicators such as temperature, humidity, wind speed, and wind direction after numerical processing; for regional operation activity data, the data value is the numerical value after quantifying the operation intensity, activity frequency, or impact degree.

[0059] To address the issue of inconsistent dimensions, different types of multi-source environmental monitoring data can first be normalized or standardized. This involves converting radiation monitoring data, meteorological observation data, and regional operational activity data into uniform dimensionless values ​​or unified indices, making them numerically comparable. Then, based on the fusion weights, these normalized data values ​​are weighted and summed to obtain fused radiation characteristic data—a comprehensive numerical value of multi-source environmental data weighted according to the fusion weights. This ensures that different data types accurately reflect the overall environmental radiation characteristics during the weighted fusion process.

[0060] Based on the above technical solution, optionally, after obtaining the environmental radiation prediction results of the area to be measured during the prediction period, the method further includes: The risk level of the environmental radiation prediction results is determined based on the preset risk level standard. If the risk level reaches the preset alarm level, the environmental radiation prediction results are input into the preset anomaly handling model to obtain the anomaly type and the corresponding response strategy. Send the exception type and corresponding response strategy to the control center.

[0061] In this plan, the preset risk level standard is a standard system for judging the safety level of environmental radiation prediction results, which can clearly divide the risk into several levels: low, medium and high.

[0062] The risk level is the specific safety level corresponding to the environmental radiation prediction results. It is calculated based on the preset risk level standards and can directly reflect the magnitude of radiation risk that may exist in the prediction stage.

[0063] The preset alarm level is a critical value set in advance in the risk level system. Once the environmental radiation risk level reaches or exceeds this value, the early warning or emergency response procedure will be activated immediately.

[0064] The pre-defined anomaly handling model is a processing mechanism established by combining historical data accumulated daily with established rules. Its core function is to analyze the potential anomalies behind high-risk radiation prediction results and provide corresponding handling methods.

[0065] Anomaly type refers to the specific reasons that cause abnormalities in environmental radiation, such as sudden accidents, monitoring equipment failures, and violations of on-site operating procedures.

[0066] Response strategies are pre-defined solutions for different causes of anomalies, such as remotely debugging equipment, taking protective measures, and arranging personnel for on-site handling.

[0067] A control center is an organization or system specifically responsible for monitoring, managing, and implementing environmental radiation early warning and response strategies.

[0068] First, based on a preset risk level standard, the environmental radiation prediction results are matched with a specific risk level. This standard clearly defines the correspondence between radiation value ranges and risk levels; for example, values ​​below the safety threshold are considered low risk, values ​​within the medium threshold range are considered medium risk, and values ​​above the high threshold are classified as high risk. The system first reads the radiation values ​​output by the prediction model, and then compares these values ​​with each risk range in the standard to determine the corresponding risk level.

[0069] Next, the obtained risk level is compared with the preset alarm level. The alarm level itself is the threshold for triggering management or emergency response; for example, medium risk and above trigger a normal alarm, while high risk triggers an emergency alarm. If the risk level of the environmental radiation prediction result reaches or exceeds the alarm level, the prediction result is input into the preset anomaly handling model, which directly outputs the anomaly type and corresponding response strategy. Finally, the identified anomaly types and response methods are organized into structured data and sent to the control center. For example, if the environmental radiation prediction result for the area to be measured during the prediction period is 0.12 μSv / h, according to the preset risk level standard, this value corresponds to a medium-high risk level and exceeds the alarm threshold of 0.1 μSv / h. After inputting this prediction result into the anomaly handling model, the model determines that there may be an anomaly in the equipment's scattering shielding, and then generates a response strategy to remotely shut down some radiation sources and notify on-site personnel to check. Subsequently, the anomaly type and response strategy are sent to the control center.

[0070] The training steps for the pre-defined exception handling model include: First, historical data is retrieved from radiation monitoring systems, meteorological databases, and emergency response systems and compiled into a training sample set. Each sample includes environmental radiation predictions for a specific time and location, corresponding actual radiation observation data, the environmental conditions at the time, and subsequent response measures. Next, these samples are labeled against historical event records to identify different anomaly types and corresponding response strategies. For example, radiation anomalies caused by equipment failure, diffusion anomalies caused by meteorological factors, and short-term fluctuations caused by operational activities are all common anomaly types; issuing warnings, restricting operations, increasing monitoring frequency, and activating emergency protective measures are all corresponding response measures. Then, the input feature data is standardized and optimized through feature engineering to extract key variables that reflect the characteristics of the anomaly, such as the deviation between predicted and actual radiation values, the rate of change in radiation values, the duration of the anomaly, and the distribution of risk levels. Finally, machine learning multi-classification algorithms such as random forests and gradient boosting decision trees are used to train the model, allowing it to learn the correlation between input features, anomaly types, and response strategies. Once the model is trained, it can automatically identify the corresponding anomaly type and provide the most appropriate response strategy when it receives new environmental radiation prediction results.

[0071] This solution links the prediction results with risk level, alarm level, and anomaly handling model, realizing an automated closed loop from radiation prediction to anomaly identification and response decision-making. This significantly improves the timeliness and accuracy of anomaly response and reduces the risk of delays and missed reports caused by manual judgment.

[0072] Based on the above technical solution, optionally, after sending the exception type and corresponding response strategy to the control center, the method also includes: If a processing completion instruction is received from the control center, the multi-source environmental monitoring data of the area to be measured within the preset time period is reacquired, and the multi-source environmental monitoring data is preprocessed to obtain a multi-source environmental dataset. Re-extract each preset sampling time point within the preset time period, and obtain the regional environmental status information and abnormal event detection results corresponding to each preset sampling time point in the area to be tested. Generate fusion adjustment condition data for the area to be tested based on each preset sampling time point, regional environmental status information and abnormal event detection results. Based on the aforementioned multi-source environmental dataset and the fusion adjustment condition data, the fusion weights are recalculated for each multi-source environmental monitoring data in the multi-source environmental dataset. The multi-source environmental monitoring data in the multi-source environmental dataset are re-weighted and fused based on the fusion weight to obtain fused radiation characteristic data. The fused radiation characteristic data and the prediction period are then input into a preset radiation prediction model to obtain the environmental radiation prediction result of the area to be measured during the prediction period. The risk level of the environmental radiation prediction results is re-determined based on the preset risk level standard. If the risk level does not reach the preset alarm level, optimization success information is generated based on the re-determined environmental radiation prediction results and the re-determined risk level, and the optimization success information is sent to the control center.

[0073] In this scheme, the completion instruction is a feedback instruction sent by the control center to the system after completing all handling operations according to the exception type and response strategy given by the exception handling model.

[0074] The optimization success message is a feedback message generated by the system when the risk level of the re-predicted environmental radiation result does not reach the preset alarm level. This message indicates that after anomaly handling and model optimization, the radiation level of the area under test has returned to a safe or stable state, demonstrating that this round of optimization and response measures were effective.

[0075] Upon receiving the completion instruction, it indicates that anomaly handling has been performed based on the corresponding strategy and anomaly type. To verify the success of this optimization, the prediction process for the environmental radiation of the area under test during the prediction period needs to be re-executed. After the prediction is completed, the risk level must be re-determined. If the risk level does not reach the preset alarm level, the optimization is considered successful. Then, based on the environmental radiation prediction results and the re-determined risk level, an optimization success message is generated and sent to the control center via wireless communication technology.

[0076] This solution allows for real-time verification of environmental recovery effects after anomaly handling, ensuring that prediction results are based on the latest data and improving the accuracy, reliability, and closed-loop adaptability of radiation monitoring and early warning.

[0077] Based on the above technical solution, optionally, after re-determining the risk level of the environmental radiation prediction results based on a preset risk level standard, the method further includes: If the risk level still reaches the preset alarm level, manual inspection information is generated based on the re-determined environmental radiation prediction results and the re-determined risk level, and the manual inspection information is sent to the control center.

[0078] In this solution, the manual inspection information is a task instruction generated for manual intervention and verification when the risk level still reaches or exceeds the preset alarm threshold after the system executes an abnormal response and makes a re-prediction.

[0079] If, after comparison, the risk level still reaches the preset alarm level, it means that the treatment is ineffective. At this time, a manual inspection process should be triggered. Based on the re-determined environmental radiation prediction results and the re-determined risk level, manual inspection information is generated and then sent to the control center via wireless communication technology, so that the control center can dispatch relevant personnel to conduct the inspection.

[0080] This solution ensures that manual verification is introduced when automatic response fails to eliminate high risks, avoiding the omission of risks due to model misjudgment or sensor malfunction, and improving the safety, reliability and accuracy of anomaly handling of the system.

[0081] See appendix Figure 2 , Figure 2 This is a schematic flowchart illustrating the main steps of a regional environmental radiation prediction method based on multi-source data fusion according to an embodiment of the present invention. Figure 2 As shown, a regional environmental radiation prediction method based on multi-source data fusion in an embodiment of the present invention mainly includes the following steps S201-S206.

[0082] Step S201: If a prediction instruction is received, the prediction period and the area to be measured are determined based on the prediction instruction. Multi-source environmental monitoring data of the area to be measured within a preset time period are obtained. The multi-source environmental monitoring data are preprocessed to obtain a multi-source environmental dataset. The multi-source environmental monitoring data includes radiation monitoring data, meteorological observation data, and regional operation activity data.

[0083] Step S202: Extract each preset sampling time point within a preset time period, and obtain the abnormal event detection results corresponding to each preset sampling time point in the area to be tested. Generate fusion adjustment condition data for the area to be tested based on each preset sampling time point and the abnormal event detection results.

[0084] Step S203: Based on the multi-source environmental dataset and the fusion adjustment condition data, calculate the fusion weight for each multi-source environmental monitoring data in the multi-source environmental dataset.

[0085] Step S204: Based on the fusion weight, perform weighted fusion of the multi-source environmental monitoring data in the multi-source environmental dataset to obtain fused radiation characteristic data. Input the fused radiation characteristic data and the prediction time period into the preset radiation prediction model to obtain the environmental radiation prediction result of the area to be measured during the prediction time period.

[0086] Step S205: Extract wind direction and wind speed information from meteorological observation data, determine the wind direction vector based on the wind direction information, and determine the wind speed vector based on the wind speed information.

[0087] Wind direction information describes the direction of air flow and is usually expressed in degrees or compass orientation. It is mainly used to determine the possible direction of diffusion of pollutants or radioactive substances.

[0088] Wind speed information describes the speed of air movement, usually expressed in meters per second or kilometers per hour, and is mainly used to estimate the rate of material diffusion.

[0089] Wind direction vectors convert wind direction information into a two-dimensional or three-dimensional vector form to represent the direction of wind in space, making it easier to calculate the potential impact path of materials.

[0090] A wind speed vector combines wind speed information with a wind direction vector to form a vector that simultaneously reflects the magnitude and direction of wind, used to accurately measure the possible speed and specific path of matter in space.

[0091] First, meteorological observation data of the area to be measured is collected. This data generally includes air temperature, air pressure, humidity, and raw wind speed and direction readings collected by various observation stations. To extract wind direction information, the raw wind direction measurements must first be decoded and standardized. Different observation stations may record the direction in angles or compass directions. These data must be uniformly converted into degrees, such as 30° north of east or 135° southeast, and must correspond to the standard circumferential angle range of 0° to 360°.

[0092] When processing wind speed information, the instantaneous wind speeds recorded by each observation station must first be filtered to remove abrupt changes and interference values. Then, time series processing methods such as moving averages and exponential smoothing can be used to obtain more stable wind speed data. After obtaining the standardized wind direction information, it is converted into a spatial vector form, namely the wind direction vector. Specifically, trigonometric functions are used to map the wind direction angle to a unit vector in a two- or three-dimensional coordinate system, where the X and Y components, and, if necessary, the Z component, represent the wind direction. The vector length is initially set to a unit length specifically to represent the directional attribute. Next, combined with the previously processed wind speed information, the unit wind direction vector is scaled according to the wind speed magnitude, thus forming the wind speed vector. The length of this vector corresponds to the wind speed magnitude, and its direction is consistent with the wind direction vector, simultaneously representing the wind speed and direction in space.

[0093] Step S206: Obtain information about the neighboring areas of the area to be measured. Based on the wind direction vector, wind speed vector, neighboring area information, and environmental radiation prediction results, determine a list of potential impact areas and send the list of potential impact areas to the control center.

[0094] Neighboring area information refers to the geographic or spatial unit information surrounding the area to be measured, covering key information such as the location, boundaries, area, relative distance, and relative direction of each neighboring area to the area to be measured.

[0095] The list of potential affected areas is a summary of areas that may be affected by environmental radiation during the forecast period, calculated by combining wind direction vectors, wind speed vectors, information from neighboring areas, and environmental radiation prediction results. Each entry in the list corresponds to a potentially affected area.

[0096] First, the geographical boundaries and spatial coordinates of the area to be measured must be clearly defined. This information can be retrieved from geographic information system data or regional planning documents. With these coordinates, other areas within a certain radius around the area to be measured can be analyzed, and those with proximity relationships can be selected and compiled into neighboring area information. This information will clearly state the location, boundaries, area, relative distance to the area to be measured, and azimuth angle of each neighboring area.

[0097] Next, the information from neighboring areas is combined with the previously predicted wind direction and speed vectors for analysis. The wind direction vector shows the direction of airflow, while the wind speed vector reflects the strength of the airflow. By extending the wind direction vector to neighboring areas and combining it with the specific values ​​of the wind speed vector, the possible propagation paths and coverage areas of air or environmental substances over a future period can be calculated. Then, the environmental radiation prediction results for the area under test are correlated with the wind direction and speed information to determine which neighboring areas are on the potential propagation paths. Finally, all potentially affected neighboring areas are compiled into a list of potential impact areas, with each entry recorded as an area identifier. This list is sent to the control center in structured data format.

[0098] This scheme combines wind direction and speed with information from neighboring areas to accurately identify areas that may be affected by environmental radiation, providing the control center with a reliable list of potentially affected areas, thereby supporting timely early warning and scientific decision-making.

[0099] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effects of the present invention, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of the present invention.

[0100] Furthermore, the present invention also provides a regional environmental radiation prediction system based on multi-source data fusion.

[0101] See appendix Figure 3 , Figure 3 This is a main structural block diagram of a regional environmental radiation prediction system based on multi-source data fusion according to an embodiment of the present invention. Figure 3 As shown, it specifically includes: The data acquisition module 301 is used to, upon receiving a prediction instruction, determine the prediction period and the area to be measured based on the prediction instruction, acquire multi-source environmental monitoring data of the area to be measured within a preset time period, preprocess the multi-source environmental monitoring data, and obtain a multi-source environmental dataset; wherein, the multi-source environmental monitoring data includes radiation monitoring data, meteorological observation data, and regional operation activity data; The fusion adjustment condition generation module 302 is used to extract each preset sampling time point within a preset time period, and to obtain the abnormal event detection results corresponding to each preset sampling time point in the area to be tested, and to generate fusion adjustment condition data for the area to be tested based on each preset sampling time point and the abnormal event detection results. The fusion weight calculation module 303 is used to calculate the fusion weight of each multi-source environmental monitoring data in the multi-source environmental dataset based on the multi-source environmental dataset and the fusion adjustment condition data. The prediction module 304 is used to perform weighted fusion of the multi-source environmental monitoring data in the multi-source environmental dataset based on the fusion weight to obtain fused radiation characteristic data. The fused radiation characteristic data and the prediction time period are input into the preset radiation prediction model to obtain the environmental radiation prediction result of the area to be measured during the prediction time period.

[0102] This application provides a regional environmental radiation prediction system based on multi-source data fusion, which can achieve... Figure 1 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.

[0103] Those skilled in the art will understand that all or part of the processes in the method of the above embodiment of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable storage medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0104] Furthermore, the present invention also provides an electronic device 400, including a processor 401, a memory 402, and a program or instructions stored in the memory 402 and executable on the processor 401. When the program or instructions are executed by the processor 401, they implement the various processes of the above-described embodiment of a regional environmental radiation prediction method based on multi-source data fusion and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0105] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0106] Furthermore, the present invention also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to the present invention, the computer-readable storage medium can be configured to store a program for executing a regional environmental radiation prediction method based on multi-source data fusion according to the above-described method embodiments. This program can be loaded and run by a processor to implement the above-described regional environmental radiation prediction method based on multi-source data fusion. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The computer-readable storage medium can be a storage device comprising various electronic devices. Optionally, in the embodiments of the present invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0107] Furthermore, it should be understood that since the various modules are only provided to illustrate the functional units of the device of the present invention, the physical devices corresponding to these modules may be the processor itself, or a part of the processor's software, hardware, or a combination of software and hardware. Therefore, the number of modules shown in the figures is merely illustrative.

[0108] Those skilled in the art will understand that the various modules in the device can be adaptively split or combined. Such splitting or combining of specific modules will not cause the technical solution to deviate from the principles of the present invention; therefore, the technical solutions after splitting or combining will fall within the protection scope of the present invention.

[0109] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A regional environmental radiation prediction method based on multi-source data fusion, characterized in that, The method comprises: If a prediction instruction is received, a prediction period and a to-be-tested region are determined based on the prediction instruction, multi-source environmental monitoring data of the to-be-tested region within a preset time length is acquired, the multi-source environmental monitoring data is preprocessed to obtain a multi-source environmental data set; wherein the multi-source environmental monitoring data comprises radiation monitoring data, meteorological observation data and regional operation activity data; Each preset sampling time point within the preset time length is extracted, and abnormal event detection results corresponding to each preset sampling time point in the to-be-tested region are acquired, and fusion adjustment condition data of the to-be-tested region is generated based on each preset sampling time point and the abnormal event detection results; Based on the multi-source environmental data set and the fusion adjustment condition data, the fusion weight of each multi-source environmental monitoring data in the multi-source environmental data set is calculated respectively; Based on the fusion weight, each multi-source environmental monitoring data in the multi-source environmental data set is weighted and fused to obtain fusion radiation feature data, and the fusion radiation feature data and the prediction period are input into a preset radiation prediction model to obtain an environmental radiation prediction result of the to-be-tested region in the prediction period.

2. The regional environmental radiation prediction method based on multi-source data fusion according to claim 1, characterized in that, Wherein, The fusion adjustment condition data of the to-be-tested region is generated based on each preset sampling time point, regional environmental state information and abnormal event detection results, comprising: Based on each preset sampling time point and the corresponding abnormal event detection results, each multi-source monitoring data unit is generated; Based on each preset sampling time point, the dynamic weight index of each multi-source monitoring data unit in the time dimension is determined; Based on the abnormal event detection results, the abnormal response coefficient of each multi-source monitoring data unit is determined; Based on the dynamic weight index and the abnormal response coefficient, the fusion adjustment condition data of the to-be-tested region is generated.

3. The regional environmental radiation prediction method based on multi-source data fusion according to claim 1, characterized in that, Wherein, Based on the fusion weight, each multi-source environmental monitoring data in the multi-source environmental data set is weighted and fused to obtain fusion radiation feature data, comprising: The multi-source environmental monitoring data in the multi-source environmental data set are weighted and fused by a preset weighted fusion formula and a fusion weight to obtain fused radiation feature data; wherein the preset weighted fusion formula is: ; wherein, is a fusion radiation feature data; is a total number of multi-source environmental monitoring data; is an index of multi-source environmental monitoring data; is a fusion weight of multi-source environmental monitoring data; is a data value of multi-source environmental monitoring data.

4. The regional environmental radiation prediction method based on multi-source data fusion according to claim 1, characterized in that, Wherein, After obtaining the environmental radiation prediction result of the to-be-tested region in the prediction period, the method further comprises: Based on the preset risk level standard, the risk level of the environmental radiation prediction result is determined, and if the risk level reaches the preset alarm level, the environmental radiation prediction result is input into a preset abnormal processing model to obtain an abnormal type and a corresponding response strategy; The abnormal type and the corresponding response strategy are sent to the control center.

5. The regional environmental radiation prediction method based on multi-source data fusion according to claim 4, characterized in that, Wherein, After sending the abnormal type and the corresponding response strategy to the control center, the method further comprises: If a processing completed instruction sent by the control center is received, the multi-source environmental monitoring data of the to-be-tested region within the preset time length is re-acquired, the multi-source environmental monitoring data is preprocessed to obtain a multi-source environmental data set; Each preset sampling time point within the preset time length is re-extracted, and regional environmental state information and abnormal event detection results corresponding to each preset sampling time point in the to-be-tested region are acquired, and fusion adjustment condition data of the to-be-tested region is generated based on each preset sampling time point, regional environmental state information and abnormal event detection results; Based on the multi-source environmental data set and the fusion adjustment condition data, the fusion weight of each multi-source environmental monitoring data in the multi-source environmental data set is calculated respectively; The multi-source environmental monitoring data in the multi-source environmental data set is weighted and fused based on the fusion weight again to obtain fusion radiation feature data, and the fusion radiation feature data and a prediction period are input into a preset radiation prediction model to obtain an environmental radiation prediction result of the to-be-measured region in the prediction period. The risk level of the environmental radiation prediction result is determined again based on the preset risk level standard, and if the risk level does not reach a preset alarm level, an optimization success information is generated based on the re-determined environmental radiation prediction result and the re-determined risk level, and the optimization success information is sent to the control center.

6. The regional environmental radiation prediction method based on multi-source data fusion according to claim 5, characterized in that, Wherein, After the risk level of the environmental radiation prediction result is determined again based on the preset risk level standard, the method further comprises: If the risk level still reaches the preset alarm level, an artificial inspection information is generated based on the re-determined environmental radiation prediction result and the re-determined risk level, and the artificial inspection information is sent to the control center.

7. The regional environmental radiation prediction method based on multi-source data fusion according to claim 1, characterized in that, Wherein, After the environmental radiation prediction result of the to-be-measured region in the prediction period is obtained, the method further comprises: Wind direction information and wind speed information are extracted from the meteorological observation data, a wind direction vector is determined according to the wind direction information, and a wind speed vector is determined according to the wind speed information; Adjacent region information of the to-be-measured region is obtained, a potential influence region list is determined based on the wind direction vector, the wind speed vector, the adjacent region information and the environmental radiation prediction result, and the potential influence region list is sent to the control center.

8. A regional environmental radiation prediction system based on multi-source data fusion, characterized in that, The system comprises: A data acquisition module is configured to determine a prediction period and a to-be-measured region based on a prediction instruction if the prediction instruction is received, acquire multi-source environmental monitoring data of the to-be-measured region within a preset time length, and obtain a multi-source environmental data set by preprocessing the multi-source environmental monitoring data; wherein the multi-source environmental monitoring data comprises radiation monitoring data, meteorological observation data and regional operation activity data. A fusion adjustment condition generation module is configured to extract each preset sampling time point within the preset time length, and obtain abnormal event detection results corresponding to each preset sampling time point in the to-be-measured region, and generate fusion adjustment condition data of the to-be-measured region based on each preset sampling time point and the abnormal event detection results. A fusion weight calculation module is configured to calculate a fusion weight for each multi-source environmental monitoring data in the multi-source environmental data set based on the multi-source environmental data set and the fusion adjustment condition data. A prediction module is configured to weight and fuse each multi-source environmental monitoring data in the multi-source environmental data set based on the fusion weight to obtain fusion radiation feature data, and input the fusion radiation feature data and a prediction period into a preset radiation prediction model to obtain an environmental radiation prediction result of the to-be-measured region in the prediction period.

9. An electronic device comprising a processor, a memory, and a program or instructions stored on the memory and executable on the processor, characterized in that, The program or instructions are adapted to be loaded and run by the processor to perform the region environmental radiation prediction method based on multi-source data fusion in any one of claims 1 to 7.

10. A computer readable storage medium having stored therein a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to perform the region environmental radiation prediction method based on multi-source data fusion in any one of claims 1 to 7.

Images