Nuclear power station accident emergency evacuation path generation method

Abstract

The invention provides a nuclear power station accident emergency evacuation path generation method, and relates to the technical field of path generation, and the method comprises the steps: collecting topographic feature information, building distribution information and traffic road data within a calibration radius range with a nuclear power station as the center, and constructing a calibration range scene model; obtaining an accident type and an accident source position in response to the accident alarm signal; acquiring an article position list and an article type list through a monitoring camera; performing spreading analysis to obtain a spreadable position set; performing spreading path prediction to obtain an accident prediction spreading path; and according to the accident prediction spreading path, taking the area outside the calibrated radius range as an evacuation target area, carrying out shortest path planning, obtaining a recommended evacuation path, and sending the recommended evacuation path to an evacuation management end. The technical problem that the evacuation path planning method in the prior art usually carries out path planning according to the state of the happened accident, so that the path planning is poor in advance, and the safety of evacuated crowds cannot be guaranteed is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of path generation, and in particular to a method for generating an emergency evacuation path for a nuclear power plant accident. Background Art

[0002] Nuclear power plant accidents usually involve serious radiation leaks, fires, explosions and other emergencies. These accidents not only damage the facilities inside the nuclear power plant, but may also cause great harm to the surrounding environment and people. Therefore, emergency evacuation planning of nuclear power plants has become an important part of ensuring public safety. However, existing evacuation route planning often takes the total evacuation time as the goal, ignores other important safety factors, fails to provide the best evacuation plan, or relies on fixed fire-fighting facilities and preset safe places, lacking flexibility. For example, traditional methods may plan evacuation routes close to fire-fighting equipment or shelters, but these routes do not take into account the dynamic spread of accidents, which may cause some evacuation routes to become unsafe when accidents occur.

[0003] All of the above evacuation route planning methods are based on the actual status after the accident occurs, and lack advance prediction of the spread of the accident. This lag may cause the evacuation route to pass through or approach high-risk areas, increasing the safety risk of evacuees. Summary of the invention

[0004] The present application provides a method for generating an emergency evacuation route for a nuclear power plant accident, aiming to solve the technical problem that the evacuation route planning method in the prior art usually plans the route according to the accident status that has occurred, resulting in poor predictability of the route planning and failure to ensure the safety of the evacuated people.

[0005] The present application discloses a method for generating an emergency evacuation path for a nuclear power plant accident, the method comprising: collecting terrain feature information, building distribution information and traffic road data within a calibrated radius range centered on the nuclear power plant, and constructing a calibrated range scene model of the nuclear power plant; obtaining the type of nuclear power plant accident and the location of the nuclear power plant accident source in response to a nuclear power plant accident alarm signal sent by a nuclear power plant alarm; collecting an object location list and an object type list within a calibrated radius range centered on the nuclear power plant through a monitoring camera, wherein the objects are movable objects and the buildings are fixed facilities; traversing the building distribution information, the object location list and the object type list, performing a spreading analysis based on the nuclear power plant accident type, and obtaining a set of spreadable locations; performing a spreading path prediction based on the set of spreadable locations according to the location of the nuclear power plant accident source, and obtaining a predicted spreading path of the nuclear power plant accident; performing shortest path planning in the calibrated range scene model of the nuclear power plant with the area outside the calibrated radius range as the evacuation target area according to the predicted spreading path of the nuclear power plant accident, obtaining a recommended evacuation path, and sending it to an evacuation management terminal.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The terrain feature information, building distribution information and traffic road data within the calibrated radius centered on the nuclear power plant are collected to construct a nuclear power plant calibrated range scenario model. The radiation range is used as the basis for building the scenario model, which can avoid nuclear radiation harm to the public due to emergency accidents and improve the safety of evacuation. In response to the nuclear power plant accident alarm signal sent by the nuclear power plant alarm, the type of nuclear power plant accident and the location of the nuclear power plant accident source are obtained, which can quickly respond to the accident in the early stage and provide an accurate starting point for spread analysis and evacuation route planning. Compared with traditional route planning methods, this real-time response enables the evacuation plan to be dynamically adjusted based on the actual situation of the accident development, avoiding the lag of subsequent planning. The list of item locations and item types within the calibrated radius centered on the nuclear power plant is collected through monitoring cameras, and combined with the building distribution information, based on the type of nuclear power plant accident, the evacuation plan is adjusted dynamically based on the actual situation of the accident development. Spread analysis can deeply analyze the spread risk of each location. The obtained set of spreadable locations shows which areas are most vulnerable to accidents, thereby greatly improving the predictability of path planning; according to the location of the nuclear power plant accident source, the spread path is predicted based on the set of spreadable locations, and the predicted spread path of the nuclear power plant accident is obtained. By identifying the actual and potential risk areas for the spread of the accident and avoiding these areas when planning the evacuation path, it is ensured that the evacuation route is far away from the impact range of the accident, thereby improving the safety of the evacuated personnel; according to the predicted spread path of the accident, the area outside the calibrated radius is used as the evacuation target, and the shortest path planning is carried out within the calibrated range of the nuclear power plant to ensure that the planned evacuation route can not only evacuate personnel quickly, but also avoid high-risk areas, avoiding the simple shortest time goal, improving the accuracy and practicality of path planning, and effectively ensuring the safety of the evacuated people.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flowchart of a method for generating an emergency evacuation route for a nuclear power plant accident is provided for an embodiment of the present application.

[0009] Figure 2 A schematic flow chart of constructing a nuclear power plant calibration range scenario model in a method for generating an emergency evacuation path for a nuclear power plant accident is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0010] The embodiment of the present application provides a method for generating an emergency evacuation route for a nuclear power plant accident, thereby solving the technical problem that the evacuation route planning method in the prior art usually plans the route according to the accident status that has occurred, resulting in poor predictability of the route planning and failure to ensure the safety of the evacuated people.

[0011] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0012] like Figure 1 As shown, an embodiment of the present application provides a method for generating an emergency evacuation path for a nuclear power plant accident, the method comprising: Collect terrain feature information, building distribution information and traffic road data within the calibration radius centered on the nuclear power plant to build a nuclear power plant calibration range scenario model.

[0013] The calibrated radius range of a nuclear power plant is a circular area generated based on the radiation range of the nuclear power plant. Nuclear power plant accidents may cause nuclear leakage and lead to nuclear radiation. Therefore, using the radiation range as the basis for building a scenario model can avoid nuclear radiation harm to people caused by emergency accidents. It is based on the installed capacity of the nuclear power plant, the number of reactors, the fuel type, the reactor type and other information. By analyzing the historical data of other nuclear power plants with similar parameters to the nuclear power plant, the maximum radiation radius data under similar situations is found to obtain the calibrated radius range.

[0014] Terrain feature information can be obtained through geographic information systems and satellite remote sensing technology. The data include terrain features such as elevation, slope, and water distribution that affect evacuation route planning; building distribution information can be obtained through building planning drawings and satellite images. Building data includes building types, such as residential, commercial buildings, industrial facilities, etc., as well as location coordinates, etc.; traffic road data can be obtained through traffic monitoring systems. The data includes road grade, road width, and traffic capacity, etc.

[0015] Determine the boundary of the nuclear power plant and generate the edge coordinates of the nuclear power plant. Extend vertically from the edge coordinates of the nuclear power plant to the calibration radius range to generate a set of extended edge coordinates. These coordinates are used to define the spatial range of the model. All collected extended edge coordinates are connected in sequence to construct the spatial framework of the scene model. The terrain features, building distribution and transportation network within the framework constitute a complete nuclear power plant calibration range scene model.

[0016] In response to a nuclear power plant accident alarm signal sent by a nuclear power plant alarm, the type of nuclear power plant accident and the location of the nuclear power plant accident source are obtained.

[0017] Various types of sensors and monitoring equipment are installed inside nuclear power plants, such as radiation detectors, temperature sensors, gas detectors, etc. These sensors monitor the operating status of the nuclear power plant in real time. When certain key parameters exceed the normal range, such as radiation leakage, abnormal reactor temperature, etc., the alarm will send out an accident alarm signal, indicating that an accident may have occurred.

[0018] After the alarm is triggered, relevant data is collected from multiple sensors and monitoring equipment. These data include but are not limited to radiation intensity, temperature changes, pressure fluctuations, gas detection, etc. The type of nuclear power plant accident is determined by analyzing the real-time data of these sensors. For example, if the readings of the radiation detector increase abnormally, it may mean that a radiation leak has occurred; if the temperature and pressure rise sharply, it may indicate problems such as reactor overheating. According to the trigger time and readings of different sensors, multi-point positioning is performed through data fusion technology to further narrow the location range of the accident source. For example, if sensors in multiple areas detect anomalies at the same time, the intersection of these abnormal areas can be calculated, and the approximate location of the accident source can be inferred to obtain the location of the nuclear power plant accident source. The location of the nuclear power plant accident source can be a reactor, storage area, pipeline leakage point or other specific facilities where the accident occurred in the nuclear power plant.

[0019] Through monitoring cameras, a list of object locations and a list of object types within a calibrated radius centered on the nuclear power plant are collected, where objects are movable objects and buildings are fixed facilities.

[0020] Surveillance cameras are arranged in various important areas within the calibrated radius of the nuclear power plant to ensure coverage of key scenes around the nuclear power plant, including major roads, squares, parking lots and other areas where movable objects may appear. The surveillance cameras have real-time video acquisition capabilities, which can clearly capture dynamic images of the areas around the nuclear power plant and track the trajectory and position changes of objects.

[0021] The items are movable items, which means that they have relatively flexible mobility. Unlike fixed buildings, movable items may include equipment, vehicles, etc.; fixed facilities are mainly buildings, fixed equipment, street facilities, etc., and these items will not change location in a short period of time.

[0022] The video and images collected by the surveillance camera are analyzed based on the image recognition algorithm to automatically distinguish the types of objects. For example, different objects in the picture are automatically identified based on the convolutional neural network, and they are classified into different types of objects, such as vehicles, pedestrians, equipment, etc. At the same time, fixed facilities such as buildings and movable objects are distinguished from each other based on visual features. The fixed position and morphological features of the building make it easier to distinguish it from movable objects. Through object type analysis, a list of object locations is established.

[0023] In the captured video or image, the specific location of movable objects is identified based on the position of the monitoring camera and recorded in the form of two-dimensional or three-dimensional coordinates to generate an object location list. For moving objects, the movement trajectory of the objects is tracked in real time through the target tracking algorithm, so as to clarify their dynamic position changes during the evacuation planning process.

[0024] The building distribution information, the item location list and the item type list are traversed, and a spread analysis is performed based on the type of nuclear power plant accident to obtain a set of possible spread locations.

[0025] Building distribution information also includes a list of building types and a list of building locations. Different building types behave differently in accidents. Based on the different characteristics of building types, their sensitivity and impact on the spread of accidents are analyzed. For example, in a fire, wooden buildings are more flammable than steel-structured buildings, so the possibility of spread is greater. The location of each building is traversed, the spatial distance between it and the accident source is calculated, and the probability of being affected is evaluated in combination with the accident type, such as radiation, fire, explosion, etc. In this way, the locations of buildings that may become spread targets in an accident are screened out, and a set of spreadable locations of buildings is obtained.

[0026] Similarly, different types of movable objects are analyzed based on their response characteristics in accidents. For example, the fuel type and storage volume of a vehicle may affect the explosion or spread of fire in an accident; and the availability of emergency equipment, such as fire-fighting equipment, will affect the selection and safety of evacuation routes. Through analysis, a set of spreadable locations for movable objects is obtained.

[0027] According to the location of the nuclear power plant accident source, a spread path prediction is performed based on the set of possible spread locations to obtain a predicted spread path of the nuclear power plant accident.

[0028] The specific location of the accident is determined according to the location of the nuclear power plant accident source. This is the starting point for the spread path prediction. Based on this location, the spread analysis is performed, and each target in the set of spreadable locations is traversed. Combined with the historical nuclear power plant accident monitoring log analysis, the spread distance threshold of each target is determined.

[0029] According to the spreading distance threshold, the location set closest to the accident source is selected from the set of spreadable locations as the first-level spreading location set. These locations are the first batch of targets that may be affected by the spread of the accident. Based on the first-level spreading location set, the impact of these locations on the surrounding areas is further analyzed, and the targets that are closer to the first-level spreading location and have a higher probability of spreading are selected as the second-level spreading location set. Continue to analyze the targets that may be affected until all spreadable locations are traversed. This process forms a complete spreading path. Each node in the path represents an affected target, and the path represents the spread direction and order of the accident, predicting how the accident gradually affects different buildings and objects.

[0030] According to the predicted spread path of the nuclear power plant accident, the area outside the calibrated radius is taken as the evacuation target area, the shortest path planning is performed in the calibrated range scenario model of the nuclear power plant, and the recommended evacuation path is obtained and sent to the evacuation management terminal.

[0031] The calibrated radius range is the influence range set with the nuclear power plant as the center. The evacuation target area is set outside the calibrated radius range, which ensures that the evacuation path can guide people to a safe area.

[0032] The shortest path planning is performed based on the nuclear power plant calibration range scenario model. The shortest path planning algorithm can adopt Dijkstra algorithm, A* algorithm, etc. These algorithms can calculate the shortest path from the accident site to the evacuation target area. In addition to the shortest path, other factors need to be considered, such as road capacity, current traffic conditions, etc. Combined with multi-objective optimization algorithms, dangerous areas can be avoided while ensuring the shortest path.

[0033] In the process of path planning, the predicted spread path of the nuclear power plant accident is integrated into the planning to ensure that evacuees avoid these high-risk areas. For example, all areas that may be affected by the spread are marked, and the evacuation path is prohibited from passing through these areas. The direction of the evacuation path is adjusted to finally obtain the recommended evacuation path, which not only takes into account the shortest driving time, but also comprehensively avoids the accident spread area and the current traffic conditions. The evacuation management end refers to the nuclear power plant emergency response department responsible for emergency evacuation and management. The recommended evacuation path is sent to the evacuation management end to guide relevant personnel to command and dispatch according to the recommended evacuation path.

[0034] Furthermore, terrain feature information, building distribution information and traffic road data within the calibration radius centered on the nuclear power plant are collected to build a nuclear power plant calibration range scenario model, which previously included: Obtain basic information of a nuclear power plant, wherein the basic information of the nuclear power plant includes installed capacity, number of reactors, fuel type and reactor type; collect a data set of records of the maximum radiation radius of samples in the same cluster with the installed capacity, number of reactors, fuel type and reactor type as constraints; perform mode analysis on the data set of records of the maximum radiation radius of samples in the same cluster to obtain the calibrated radius range.

[0035] The basic information of nuclear power plants includes installed capacity, number of reactors, fuel type and reactor type. The installed capacity is the total power generation capacity of the nuclear power plant. The larger the installed capacity, the larger the scale of the reactor, and the wider the potential impact range of the accident. The more reactors there are, the more radiation sources may be caused in the event of an accident. Common fuel types include uranium-235 and plutonium-239. Different fuel types have different radiation effects and consequences in accidents. Reactor types include pressurized water reactors, boiling water reactors, heavy water reactors, etc. Different types of reactors have different safety, heat dissipation performance and risks in the event of accidents. The above basic information of nuclear power plants can come from the design documents, operating manuals, and data provided by regulatory agencies of nuclear power plants. These data provide input parameters for subsequent radiation range analysis.

[0036] Cluster samples refer to other nuclear power plants or nuclear accident cases with similar parameters to the current nuclear power plant. Cluster samples are screened based on the installed capacity, number of reactors, fuel type and reactor type of the nuclear power plant to ensure that the collected samples have sufficient similarity with the current nuclear power plant. The radiation radius record data set of cluster samples is collected from historical data or simulation experiments, and the maximum radius records are screened out to form the maximum radiation radius record data set of cluster samples.

[0037] Mode analysis is a statistical analysis method used to find the most frequently occurring value in a data set, perform statistical analysis on the maximum radiation radius record data set of samples in the same cluster, summarize the maximum radiation radius in each sample, and count the radiation range with the most occurrences as the calibration radius range, which represents the most common maximum distance that nuclear radiation may spread to when an accident occurs in the target nuclear power plant. The calibration radius is the core input parameter for evacuation path planning for nuclear power plant accidents. By using the radiation range as the basis of the scenario model, the harm caused by nuclear radiation to the population during an accident can be effectively avoided.

[0038] Furthermore, if Figure 2 As shown, terrain feature information, building distribution information and traffic road data within the calibration radius centered on the nuclear power plant are collected to build a nuclear power plant calibration range scenario model, including: Obtain the edge coordinates of the nuclear power plant; based on the calibrated radius range, traverse the edge coordinates of the nuclear power plant and extend vertically outward to obtain an extended edge coordinate set; connect the extended edge coordinate sets in sequence to construct a scene acquisition range; according to the scene acquisition range, collect the terrain feature information, the building distribution information and the traffic road data to construct the nuclear power plant calibration range scene model.

[0039] The edge coordinates of a nuclear power plant are obtained through the design drawings, field measurement data or satellite images of the nuclear power plant. The edge coordinates of a nuclear power plant refer to the coordinates of the outer boundaries of the physical facilities of the nuclear power plant. They can be the edge points of the walls, external facilities or key buildings of the nuclear power plant, expressed in longitude and latitude or other standard coordinate systems.

[0040] For each point of the nuclear power plant edge coordinate, vertical extension is performed based on the calibrated radius. Specifically, starting from each edge point, it is extended outward along the line connecting the point and the center of the nuclear power plant to the distance of the calibrated radius. The generated extended edge coordinate set contains all the extended coordinates extending outward from the edge points of the nuclear power plant.

[0041] The generated set of extended edge coordinates is connected in sequence to form a complete external boundary as the scene acquisition range. This irregular boundary line can surround the nuclear power plant and reflect the complex terrain, building distribution and obstacles around the nuclear power plant. Compared with the traditional regular boundary, the extended edge coordinates generated by vertical extension constitute an irregular external boundary, which can more accurately reflect the complex terrain and obstacles around the nuclear power plant. Due to the diversity of terrain, this method is more practical than a single regular radius and can effectively improve the evacuation accuracy.

[0042] Data collection is performed within the scope of scene collection, where terrain feature information can be obtained through geographic information systems and satellite remote sensing technology, including terrain features such as elevation, slope, and water distribution that affect evacuation route planning; building distribution information can be obtained through building planning drawings and satellite images, and building data includes building types, such as residential, commercial buildings, industrial facilities, etc., as well as location coordinates, etc.; traffic road data can be obtained through traffic monitoring systems, and data includes road grade, road width, traffic capacity, etc. Integrate terrain features, building distribution, traffic road information, etc. into the scene model to obtain a nuclear power plant calibration range scene model, which is used to simulate the environment around the nuclear power plant and provide a reference for subsequent evacuation route planning.

[0043] Furthermore, traversing the building distribution information, the item location list and the item type list, performing a spread analysis based on the type of the nuclear power plant accident, and obtaining a set of possible spread locations includes: The building distribution information also includes a building type list and a building location list; the building type list and the item type list are traversed, and a spread probability analysis is performed based on the nuclear power plant accident type to obtain a building type spread probability list and an item type spread probability list; according to the building type spread probability list and the item type spread probability, the building type list and the item type list are sorted into spreadable building types and spreadable item types that are greater than or equal to a spread probability threshold; according to the building location list and the item location list, the spreadable building types and the spreadable item types are matched to obtain the spreadable location set.

[0044] The building distribution information also includes a list of building types and a list of building locations. The building type list contains the classification information of all buildings around the nuclear power plant. Different types of buildings react differently to the spread of accidents. Common building types include residential buildings, commercial buildings, industrial facilities, etc., which can be collected through field surveys, building design drawings or satellite images; the building location list contains the geographic coordinates of all buildings within the calibration range of the nuclear power plant, which can be obtained through a high-precision geographic information system, indicating the specific location of each building around the nuclear power plant, so as to carry out subsequent spatial analysis and spread path calculation.

[0045] Based on the types of nuclear power plant accidents, such as fire, nuclear leakage, explosion, etc., the spread probability of buildings and objects is evaluated. The type of nuclear power plant accident determines the spread mode of the accident. Different types of accidents have different spread risks. For example, the spread path of a fire accident is affected by combustibles, while the spread range of a radiation leakage accident depends on wind speed and the blocking effect of buildings. Based on the historical nuclear power plant accident monitoring logs, the building type list and the object type list are traversed, the spread probability of each building and object under a specific accident is analyzed, and a building type spread probability list and an object type spread probability list are generated.

[0046] A spreading probability threshold is set through statistical analysis, accident experience or simulation experiments, and buildings and objects above this threshold are judged as high-risk targets. According to the building type spreading probability list and the object type spreading probability list, each type is screened, and the targets with spreading probability greater than or equal to the set threshold are retained, and these buildings and objects are regarded as spreading building types and spreading object types.

[0047] According to the building location list and the object location list, the selected spreadable building types and object types are matched. By associating the specific locations with the corresponding high-risk types, it is possible to locate which buildings and objects are most likely to be affected in the accident. Finally, a spreadable location set is generated, which contains the specific locations of all high-risk buildings and objects. These locations serve as an important basis for subsequent spread path analysis and evacuation path planning.

[0048] Furthermore, traversing the building type list and the item type list, performing a spread probability analysis based on the nuclear power plant accident type, and obtaining a building type spread probability list and an item type spread probability list include: Extract the target to be analyzed from the building type list and the object type list; collect the first historical nuclear power plant accident monitoring log of the nuclear power plant accident type, wherein the accident location of any historical nuclear power plant accident in the first historical nuclear power plant accident monitoring log is unobstructed from the target to be analyzed and is less than or equal to a preset distance; obtain the proportion of the spread frequency of the target to be analyzed in the first historical nuclear power plant accident monitoring log, set it as the spread probability of the target to be analyzed, and add it to the building type spread probability list or the object type spread probability list.

[0049] Randomly extract an analysis object from the building type list and the item type list as the target to be analyzed.

[0050] The type of nuclear power plant accident determines the spread pattern of the accident. Different types of accidents have different spread risks. The first historical nuclear power plant accident monitoring log collected is the same as the current nuclear power plant accident type. For example, if the current accident is a radiation leak, the collected historical log also contains similar radiation leak accident records. The monitoring log refers to the data record collected from past nuclear power plant accidents, which records the specific location of the accident, the type of accident, the spread path, the affected buildings and objects, and other information.

[0051] Among them, only those monitoring logs with no significant obstruction between the accident location and the target to be analyzed are collected, which means that there are no obstacles or other building interferences between the accident source in the historical accident and the target to be analyzed, and the distance between the location of the historical accident and the target to be analyzed is less than or equal to the preset distance. This preset distance is set according to the type of accident and the spread mode, so that the impact on the target to be analyzed can be obtained to the greatest extent.

[0052] Analyze the first historical nuclear power plant accident monitoring logs, count the frequency of each target being affected in these accidents, and calculate the spread frequency ratio. The spread frequency ratio is the ratio of the number of times the target being affected in historical accidents to the total number of accidents. This ratio can reflect the probability of the target being affected in the accident. The definition of being affected is the situation in which the building or object is affected in historical accidents. For example, if a certain building is affected in 6 out of 10 related accidents, its spread frequency is 60%. The calculated spread probability of the target to be analyzed is added to the building type spread probability list or object type spread probability list corresponding to the target to be analyzed, as an important basis for subsequent spread path prediction and evacuation planning.

[0053] Further, according to the location of the nuclear power plant accident source, a propagation path prediction is performed based on the set of possible propagation locations to obtain a predicted propagation path of the nuclear power plant accident, including: Traversing the set of spread target types of the set of spreadable locations to calibrate the spread distance and obtain a set of spread distance thresholds; selecting a first-level spread location set from the set of spreadable locations based on the location of the nuclear power plant accident source and the set of spread distance thresholds; traversing the first-level spread location set and the set of spread distance thresholds and selecting a second-level spread location set from the set of spreadable locations; until all the sets of spreadable locations are traversed to obtain an N-level spread location set; connecting the location of the nuclear power plant accident source, the first-level spread location set, the second-level spread location set until the N-level spread location set, to obtain the predicted spread path of the nuclear power plant accident.

[0054] By analyzing historical data, the spread distance threshold of each spread target type is obtained. The threshold represents the maximum distance that the accident may spread from the source location to the target of this type. All spreadable locations are traversed, the spread distance of each spread target type is calibrated, and a set of spread distance thresholds is generated.

[0055] The location of the accident source is compared with the threshold in the spread distance threshold set, and the spreadable locations that are closest to the accident source and meet the spread threshold are selected to form a first-level spread location set. These first-level locations are the first wave of impact objects of the accident spread. Their distance from the accident source meets the corresponding spread distance threshold, so they are the areas most likely to be affected by the accident first.

[0056] After the first-level spread position is affected by the accident, the spread may continue to spread to other locations around it. The first-level spread position set is traversed, and the distance between each first-level position and other spreadable positions is analyzed. Combined with the spread distance threshold, the spreadable positions that meet the spread conditions of the first-level position are selected to form a second-level spread position set. Each first-level spread position will further affect other buildings or objects. Therefore, the second-level spread position is a further extension of the first-level spread position.

[0057] The set of spreadable locations is traversed step by step, from the first-level spread location set to the second-level, third-level, and so on, until all the sets of spreadable locations are traversed. Each level of spread location continues to spread to the next level, forming a complete spread chain, and finally obtaining an N-level spread location set. The N-level spread location represents the last level of spread, which is usually the farthest boundary area of ​​the accident impact range.

[0058] The accident source location, the first-level spread location, the second-level spread location, and the N-level spread location are gradually connected to form a complete nuclear power plant accident prediction spread path. This path reflects the process of the accident starting from the source location and gradually spreading to different target locations. Each location represents a spread node of the accident, and the path shows the order and direction of the accident spread. The spread path includes not only the connection of physical locations, but also the spread time and impact degree of each location, which provides a reference for subsequent evacuation path planning to ensure that the evacuation route can avoid high-risk areas.

[0059] Furthermore, traversing the set of spreading target types of the set of spreading locations to calibrate the spreading distance to obtain a set of spreading distance thresholds includes: Obtain a first spread target type, collect a second historical nuclear power plant accident monitoring log of the nuclear power plant accident type, wherein the second historical nuclear power plant accident monitoring log includes a distribution distance record value of a fault area and the first spread target type; perform de-extreme value processing on the distribution distance record value to obtain a concentrated distribution distance record value; extract the maximum value of the concentrated distribution distance record value, set it as the spread distance threshold of the first spread target type, and add it into the spread distance threshold set.

[0060] An analysis object is randomly selected from the set of spread target types as the first spread target type. The collected second historical nuclear power plant accident monitoring log is the same as the current nuclear power plant accident type. The log includes the fault area and the distribution distance of the first spread target type relative to the fault area. The distribution distance indicates the distance that the target is affected in the accident.

[0061] Extreme values ​​usually refer to abnormal extreme data points that are far away from the main trend. For accident spread analysis, certain extreme conditions, such as abnormal weather conditions and special terrain conditions, may cause individual spread distances to far exceed the normal range. For distribution distance record values, statistical methods are used to remove these extreme data points to prevent them from affecting the reasonable judgment of spread distance. After de-extreme value processing, concentrated distribution distance record values ​​are obtained. This set of data can better represent the actual distance affected by the first spread target type in most accidents.

[0062] From the obtained concentrated distribution distance record values, the maximum value is extracted as the maximum affected distance of the first spread target type, that is, the spread distance threshold of the first spread target type. This maximum value reflects the maximum distance at which the target type is affected in most accidents. The calculated spread distance threshold of the first spread target type is added to the spread distance threshold set, which includes thresholds for all spread target types, providing a clear distance limit for subsequent spread path analysis, thereby better predicting the scope of accident spread.

[0063] In summary, the method for generating an emergency evacuation path for a nuclear power plant accident provided by the embodiment of the present application has the following technical effects: The terrain feature information, building distribution information and traffic road data within the calibrated radius centered on the nuclear power plant are collected to construct a nuclear power plant calibrated range scenario model. The radiation range is used as the basis for building the scenario model, which can avoid nuclear radiation harm to the public due to emergency accidents and improve the safety of evacuation. In response to the nuclear power plant accident alarm signal sent by the nuclear power plant alarm, the type of nuclear power plant accident and the location of the nuclear power plant accident source are obtained, which can quickly respond to the accident in the early stage and provide an accurate starting point for spread analysis and evacuation route planning. Compared with traditional route planning methods, this real-time response enables the evacuation plan to be dynamically adjusted based on the actual situation of the accident development, avoiding the lag of subsequent planning. The list of item locations and item types within the calibrated radius centered on the nuclear power plant is collected through monitoring cameras, and combined with the building distribution information, based on the type of nuclear power plant accident, the evacuation plan is adjusted dynamically based on the actual situation of the accident development. Spread analysis can deeply analyze the spread risk of each location. The obtained set of spreadable locations shows which areas are most vulnerable to accidents, thereby greatly improving the predictability of path planning; according to the location of the nuclear power plant accident source, the spread path is predicted based on the set of spreadable locations, and the predicted spread path of the nuclear power plant accident is obtained. By identifying the actual and potential risk areas for the spread of the accident and avoiding these areas when planning the evacuation path, it is ensured that the evacuation route is far away from the impact range of the accident, thereby improving the safety of the evacuated personnel; according to the predicted spread path of the accident, the area outside the calibrated radius is used as the evacuation target, and the shortest path planning is carried out within the calibrated range of the nuclear power plant to ensure that the planned evacuation route can not only evacuate personnel quickly, but also avoid high-risk areas, avoiding the simple shortest time goal, improving the accuracy and practicality of path planning, and effectively ensuring the safety of the evacuated people.

[0064] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating an emergency evacuation route for a nuclear power plant accident, characterized in that: The method comprises: Collect terrain feature information, building distribution information and traffic road data within the calibration radius centered on the nuclear power plant, and build a scenario model of the nuclear power plant calibration range; In response to a nuclear power plant accident alarm signal sent by a nuclear power plant alarm, obtaining the type of nuclear power plant accident and the location of the nuclear power plant accident source; Through surveillance cameras, a list of object locations and a list of object types within a calibrated radius centered on the nuclear power plant are collected, where objects are movable objects and buildings are fixed facilities; Traversing the building distribution information, the item location list and the item type list, performing a spreading analysis based on the type of the nuclear power plant accident, and obtaining a set of possible spreading locations; According to the location of the nuclear power plant accident source, a spread path is predicted based on the set of possible spread locations to obtain a predicted spread path of the nuclear power plant accident; According to the predicted spread path of the nuclear power plant accident, the area outside the calibrated radius is taken as the evacuation target area, the shortest path planning is performed in the calibrated range scenario model of the nuclear power plant, and the recommended evacuation path is obtained and sent to the evacuation management terminal.

2. The method according to claim 1, characterized in that Collect terrain feature information, building distribution information and traffic road data within the calibration radius centered on the nuclear power plant to build a nuclear power plant calibration range scenario model, which previously included: Obtaining basic information of a nuclear power plant, wherein the basic information of a nuclear power plant includes installed capacity, number of reactors, fuel type, and reactor type; Taking the installed capacity, the number of reactors, the fuel type and the reactor type as constraints, collecting a data set of maximum radiation radius records of samples in the same cluster; The mode analysis is performed on the maximum radiation radius record data set of the same cluster samples to obtain the calibrated radius range.

3. The method according to claim 1, characterized in that Collect terrain feature information, building distribution information and traffic road data within the calibration radius centered on the nuclear power plant, and build a nuclear power plant calibration range scenario model, including: Get the edge coordinates of the nuclear power plant; Based on the calibrated radius range, traverse the edge coordinates of the nuclear power plant and extend vertically outward to obtain an extension edge coordinate set; Connecting the extended edge coordinate sets in sequence to construct a scene acquisition range; According to the scene collection range, the terrain feature information, the building distribution information and the traffic road data are collected to construct the nuclear power plant calibration range scene model.

4. The method according to claim 1, characterized in that Traversing the building distribution information, the item location list, and the item type list, performing a spread analysis based on the type of the nuclear power plant accident, and obtaining a set of possible spread locations, including: The building distribution information also includes a building type list and a building location list; Traversing the building type list and the item type list, performing a spread probability analysis based on the nuclear power plant accident type, and obtaining a building type spread probability list and an item type spread probability list; According to the building type spread probability list and the item type spread probability, sorting the building type list and the item type list into spreadable building types and spreadable item types that are greater than or equal to a spread probability threshold; According to the building location list and the item location list, the spreadable building type and the spreadable item type are matched to obtain the spreadable location set.

5. The method according to claim 4, characterized in that Traversing the building type list and the item type list, performing a spread probability analysis based on the nuclear power plant accident type, and obtaining a building type spread probability list and an item type spread probability list, including: Extracting a target to be analyzed from the building type list and the object type list; Collecting a first historical nuclear power plant accident monitoring log of the nuclear power plant accident type, wherein the accident location of any historical nuclear power plant accident in the first historical nuclear power plant accident monitoring log is unobstructed from the target to be analyzed and is less than or equal to a preset distance; The frequency ratio of the target to be analyzed in the first historical nuclear power plant accident monitoring log is obtained, set as the target to be analyzed spread probability, and added to the building type spread probability list or the object type spread probability list.

6. The method according to claim 1, characterized in that According to the location of the nuclear power plant accident source, a spread path is predicted based on the set of spreadable locations to obtain a predicted spread path of the nuclear power plant accident, including: Traversing the set of spreading target types of the set of spreading locations to calibrate the spreading distance, and obtaining a set of spreading distance thresholds; Based on the nuclear power plant accident source location and the spread distance threshold set, selecting a primary spread location set from the spreadable location set; Traversing the primary spreading position set and the spreading distance threshold set, and selecting a secondary spreading position set from the spreadable position set; Until all the spreadable location sets are traversed, an N-level spread location set is obtained; The nuclear power plant accident source location, the first-level spread location set, the second-level spread location set, and finally the N-level spread location set are connected to obtain the predicted spread path of the nuclear power plant accident.

7. The method according to claim 6, characterized in that Traversing the set of spreading target types of the set of spreading locations to calibrate the spreading distance, and obtaining a set of spreading distance thresholds, including: Obtaining a first spreading target type, collecting a second historical nuclear power plant accident monitoring log of the nuclear power plant accident type, wherein the second historical nuclear power plant accident monitoring log includes a distribution distance record value between a fault area and the first spreading target type; Performing de-extreme value processing on the distribution distance record value to obtain a concentrated distribution distance record value; The maximum value of the concentrated distribution distance record value is extracted, set as the first spreading target type spreading distance threshold, and added into the spreading distance threshold set.

Images