Lightning falling area prediction method and system based on improved cellular automaton

Abstract

The invention relates to a lightning falling area prediction method and system based on an improved cellular automaton, and belongs to the field of meteorological data processing, and the method comprises the following steps: (1) obtaining data; (2) carrying out gridding on the acquired data and the geographic area; (3) analyzing real-time data of the weather radar; (4) making a dynamic behavior rule of the cellular automaton; (5) carrying out prediction work based on the formulated behavior rule; and (6) outputting a prediction result. The method has the advantages that the cellular automaton model is improved, which is mainly reflected in improvement of the propagation rule of the cells, and the evolution characteristics of the weather system are given to the cellular automaton, so that the cellular automaton has similar characteristics of the weather system, and the motion law of the cellular automaton is synchronous with that of the weather system at the same frequency; the accuracy of predicting the future falling area of thunder and lightning by using the improved cellular automaton is greatly improved.

Description

Technical Field

[0001] The invention relates to a lightning strike area prediction method based on an improved cellular automaton, and belongs to the field of meteorology. Background Art

[0002] Lightning is a common and extremely harmful natural disaster. Its characteristics are: first, it occurs in a wide range, almost everywhere in my country; second, it is highly harmful, causing casualties, damage to buildings, damage to electrical and electronic equipment, forest and grassland fires, and fires or explosions in flammable and explosive places; third, it occurs over a long period of time, with the length of time between the first and last thunderstorm days in most parts of my country exceeding half a year. To prevent the occurrence of lightning disasters, the best way is to predict the area where lightning will occur in advance so as to take corresponding lightning protection measures in advance.

[0003] There are many methods for predicting lightning, and the common methods are mainly divided into two categories: one is the extrapolation method, which uses regression method, optical flow method, etc. to extrapolate the lightning strike area at the next moment; the other method is the recently emerging machine learning method, which has higher prediction accuracy than ordinary extrapolation methods, but requires computing power to support it. For business applications, it is necessary to have high prediction accuracy and reduce computing power requirements, which requires optimization of machine learning methods. Summary of the invention

[0004] In order to overcome the defects of the prior art, the present invention provides a method for predicting lightning strike areas based on an improved cellular automaton. The technical solution of the present invention is:

[0005] A lightning strike zone prediction method based on an improved cellular automaton comprises the following steps:

[0006] (1) Data acquisition: Acquire real-time lightning location data and combined reflectivity and thunderstorm movement speed data of the forecast area;

[0007] (2) gridding the acquired data and geographic areas;

[0008] (3) Analyze real-time data from weather radar;

[0009] (4) Formulate dynamic behavior rules for cellular automata;

[0010] (5) Carry out forecasting work based on the established behavioral rules;

[0011] (6) Output of prediction results.

[0012] The step (2) is specifically as follows:

[0013] 2.1 Regional grid division: The area to be predicted is evenly divided into grids to ensure that each grid covers the same geographical area;

[0014] 2.2 Lightning zone marking: If lightning occurs and hits a grid at a certain time point, the grid is marked as a lightning zone; this marking indicates that a lightning event has occurred in the grid and is the basis for the initial cell in the cellular automaton model;

[0015] 2.3 Data synchronization: The radar system completes a comprehensive volume scan every six minutes to generate a new set of observation data. In order to make the input data of the model consistent with the actual radar observation data in time, the lightning location data and radar data within the most recent six-minute interval are selected as the initial input data of the model;

[0016] 2.4 Initial data preparation: Using these six minutes of data, determine the initial minefield grid and use it as the starting state of the cellular automaton model.

[0017] The step (3) is specifically as follows:

[0018] 3.1 Real-time data acquisition: By connecting to the weather radar system, the weather data detected by the radar can be obtained in real time;

[0019] 3.2 Radar echo analysis: Based on the real-time data obtained in step 3.1, analyze the intensity evolution trend, moving speed and moving direction of the radar echo. These characteristics will serve as the basis for formulating the behavior rules of the cellular automation.

[0020] The step (4) is specifically as follows:

[0021] 4.1 Radar data feature analysis: Analyze the evolution trend of the intensity, moving speed and moving direction of the combined reflectivity of the weather radar.

[0022] 4.2 Establish breeding rules:

[0023] Reproduction capacity: When the radar echo is enhanced, the strength of the combined reflectivity of the radar echo is judged. If the maximum value of the combined reflectivity is increasing, it means that the thunderstorm cloud is developing and strengthening. The development of thunderstorm clouds is often accompanied by an increase in the number of lightning occurrences, indicating that thunderstorm activity is intensifying. At this time, the reproduction capacity of the cellular automaton is also enhanced accordingly; on the contrary, if the maximum value of the combined reflectivity is decreasing, it means that the thunderstorm cloud is weakening, and the number of lightning occurrences will also decrease accordingly;

[0024] Reproduction speed: The reproduction speed of the cells matches the movement speed of the radar echo;

[0025] Reproduction direction: The reproduction direction of the cell is consistent with the movement direction of the radar echo;

[0026] Dynamic adjustment: According to the real-time update of radar data, the behavior rules of the cellular automaton are dynamically adjusted to ensure that the model's predictions are synchronized with the evolution of the actual weather system.

[0027] The step (5) is specifically as follows:

[0028] 5.1 Initialization prediction: The initial state of the cellular automaton model is determined by the actual lightning location data that occurred in the last 6 minutes;

[0029] 5.2 Applying behavioral rules: Based on the behavioral rules formulated in step (4), including reproduction capacity, reproduction speed and reproduction direction, the cellular automaton model begins to simulate the development of the lightning strike zone; 5.3 Performing short-term prediction: The model first performs a short-term prediction for the next 6 minutes to determine the possible location of the lightning strike zone in the short term;

[0030] 5.4 Perform medium-term forecast: The model uses the same behavioral rules to perform a second 6-minute forecast to determine the medium-term lightning strike area.

[0031] 5.5 Long-term forecast: As needed, longer-term forecasts are conducted to provide longer-term predictions of lightning strike areas.

[0032] 5.6 Iteration process: The cellular automaton model is continuously iterated, and each new 6-minute period becomes the input data for a new round of predictions.

[0033] The step (6) is specifically as follows:

[0034] 6.1 Determine the prediction area: Determine the lightning strike area based on the distribution of propagation cells predicted by the cellular automaton model;

[0035] 6.2 Draw ellipses: Draw ellipses along the periphery of the propagated cells. The area enclosed by these ellipses is the predicted lightning strike area;

[0036] 6.3 Graphical representation: Use graphics software or GIS tools to draw these ellipses on a map to visually display the predicted lightning strike area;

[0037] 6.4 Real-time display: The predicted lightning strike area is displayed in real time on the terminal screen, so that users can immediately obtain lightning warning information.

[0038] 6.5 Update prediction: As time goes by and new data is input, the cellular automaton model is continuously iterated, and the prediction results are updated accordingly. The updated prediction results will be continuously displayed on the terminal screen to reflect the latest prediction information;

[0039] The data mentioned in step 3.1 include the strength, position, moving speed and direction information of the radar echo.

[0040] A system for implementing the lightning strike zone prediction method based on improved cellular automata, comprising:

[0041] The data acquisition module is used to obtain the real-time lightning location data of the forecast area and the combined reflectivity and thunderstorm movement speed data of the weather radar;

[0042] A gridding module is used to grid the acquired data and geographic areas;

[0043] Real-time data analysis module, used to analyze real-time data from weather radar;

[0044] Behavior rule formulation module, used to formulate dynamic behavior rules of cellular automata;

[0045] The prediction module is used to carry out prediction work based on the established behavioral rules;

[0046] The result output module is used to output the prediction results.

[0047] The advantages of the present invention are:

[0048] 1. The cellular automaton model has been improved, mainly in terms of improving the reproduction rules of cells, giving the characteristics of the evolution of the weather system to the cellular automaton, so that the cellular automaton has similar characteristics to the weather system, and its movement law is synchronized with the weather system. The accuracy of predicting the future lightning strike area with this improved cellular automaton has been greatly improved;

[0049] 2. The improved cellular automaton model provided is simpler to calculate and consumes less computing power than the general machine learning algorithm method for predicting lightning, which is very important for realizing the commercialization of lightning warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is the main structural block diagram of the system of the present invention.

[0051] Figure 2 This is a diagram of a classic cellular automaton model.

[0052] Figure 3 It is a schematic diagram of the reproduction direction of cells in the classic cellular automaton model.

[0053] Figure 4 It is a cellular automaton model diagram of the present invention.

[0054] Figure 5 It is a schematic diagram of the reproduction direction of cells in the cellular automaton model of the present invention.

[0055] Figure 6 The present invention discloses a lightning strike position map from 00:00 to 05:00 on June 10, 2016.

[0056] Figure 7 This is a schematic diagram of the gridded lightning data.

[0057] Figure 8 It is based on the cellular automation model and uses 0-5 minute data to predict the lightning that will occur in 6-11 minutes and compare it with the actual situation. Among them, red is correct, blue is false alarm, and green is missed alarm.

[0058] Fig. 9 This is a comparison chart of the lightning that will occur in 12-17 minutes predicted by using data from 6-10 minutes and the actual situation. Red is correct, blue is false, and green is missed. DETAILED DESCRIPTION

[0059] The present invention will be further described below in conjunction with specific embodiments, and the advantages and features of the present invention will become clearer as the description proceeds. However, these embodiments are exemplary only and do not constitute any limitation to the scope of the present invention. It should be understood by those skilled in the art that the details and forms of the technical solution of the present invention may be modified or replaced without departing from the spirit and scope of the present invention, but these modifications and replacements all fall within the scope of protection of the present invention.

[0060] See also Figures 1 to 9 The present invention relates to a method for predicting lightning strike zones based on an improved cellular automaton, comprising the following steps:

[0061] (1) Data acquisition: Acquire real-time lightning location data and combined reflectivity and thunderstorm movement speed data of the forecast area;

[0062] (2) gridding the acquired data and geographic areas;

[0063] (3) Analyze real-time data from weather radar;

[0064] (4) Formulate dynamic behavior rules for cellular automata;

[0065] (5) Carry out forecasting work based on the established behavioral rules;

[0066] (6) Output of prediction results.

[0067] The lightning strike zone prediction method based on improved cellular automata has the following advantages:

[0068] Improve forecast accuracy: By dynamically adjusting the behavior rules of cellular automata to adapt to real-time weather changes, the development of lightning strike areas can be simulated more accurately.

[0069] Reduced computational complexity: Compared to traditional machine learning algorithms, cellular automata models generally have lower computational complexity, making them more suitable for real-time prediction applications.

[0070] Reduce computing power requirements: This method does not require high-performance computing resources and can be run in a common computing environment, which is conducive to promotion and application.

[0071] Real-time update: The model can process new data in real time and quickly update the prediction results, improving the timeliness of lightning warnings.

[0072] Intuitive forecast results: The forecast results are intuitively displayed on the map in the form of ellipses, making it easier for users to understand and take action.

[0073] Strong adaptability: This method can adapt to different climatic conditions and geographical environments and has strong versatility.

[0074] Scalability: The forecasting framework allows for further integration of other meteorological parameters and forecasting models to improve the forecasting capabilities.

[0075] Decision support: Provide scientific basis for weather forecasters and emergency response teams to help them make more accurate decisions.

[0076] Reduce false alarms and missed alarms: Through accurate predictions, unnecessary false alarms and missed alarms can be reduced, and the reliability of the early warning system can be improved.

[0077] The technical solution of the present invention is to use an improved cellular automaton to predict the future lightning strike area.

[0078] (1) Conceptual Model of Cellular Automata

[0079] Cellular automata are nonlinear dynamic systems with discrete characteristics, which are widely used in life sciences, physics, chemistry, materials science, economics, sociology, etc. At present, simulation studies in the fields of fluid mechanics, atmospheric circulation, ecology, material growth, urban expansion, and information dissemination using cellular automata have achieved very good results; predictions using the game of life model, forest and grassland fire spread model, and image recognition model established by cellular automata have a high degree of fit with the real scene.

[0080] The occurrence and location of lightning are discrete in time and space, and cellular automata are themselves discrete and abstract computing systems, so the two are very compatible in terms of characteristics. In addition, the thunderstorm weather system that carries lightning is essentially a dynamic system, and cellular automata are also a dynamic system, so using cellular automata to simulate and predict the future location of lightning is highly feasible and applicable.

[0081] The reproduction rules of cells in cellular automata are generally fixed. This method is not completely the same as the various typical cellular automata models currently available. This method has improved the behavior rules of cells. The improvements are: the behavior rules of cells are dynamic and change in real time according to the strength and movement direction of weather radar echoes. The reproduction ability, reproduction direction, reproduction speed and other factors of cells in the improved cellular automata are no longer fixed, but change in real time according to the development of the weather system. In other words, the evolution and development of the improved cellular automata model also have similar characteristics to the weather system, and are basically synchronized with the development of the weather system. Therefore, the accuracy of predicting lightning strike areas using such a model will be higher.

[0082] In the classic cellular automaton model, the cells propagate in all directions with equal probability or randomly. Figure 2 The square 0 in the middle represents the original cell, and the square 1 represents the cell that is reproduced for the first time. Figure 3 Square No. 2 in the middle is the cell for the second reproduction.

[0083] (2) Improved cellular automaton model

[0084] The number and direction of reproduction of the improved cellular automaton are not random or equiprobable, but change according to the development of the weather system. If the weather system is developing more, the number of reproduction will increase; otherwise, the number of reproduction will decrease ( Figure 4 The direction of cell reproduction is determined by the movement direction of the weather system, and its direction is consistent with the movement direction of the weather system ( Figure 5 ).

[0085] The step (2) is specifically as follows:

[0086] 2.1 Regional grid division: The area to be predicted is evenly divided into grids, with each grid size of 10km×10km, ensuring that each grid covers the same geographical area;

[0087] 2.2 Lightning zone marking: If lightning occurs and hits a grid at a certain time point, the grid is marked as a lightning zone; this marking indicates that a lightning event has occurred in the grid and is the basis for the initial cell in the cellular automaton model;

[0088] 2.3 Data synchronization: The weather radar commonly used in meteorological services is Doppler radar. This radar system completes a comprehensive volume scan every six minutes to generate a new set of observation data. In order to make the input data of the model consistent with the actual radar observation data in time, the lightning location data and radar data within the most recent six-minute interval are selected as the initial input data of the model;

[0089] 2.4 Initial data preparation: Using these six minutes of data, determine the initial minefield grid and use it as the starting state of the cellular automaton model.

[0090] The step (3) is specifically as follows:

[0091] 3.1 Real-time data acquisition: By connecting to the weather radar system, the weather data detected by the radar can be obtained in real time;

[0092] 3.2 Radar echo analysis: Based on the real-time data obtained in step 3.1, analyze the intensity evolution trend, moving speed and moving direction of the radar echo. These characteristics will serve as the basis for formulating the behavior rules of the cellular automaton. These characteristics directly affect the parameters such as the reproduction ability, reproduction direction and reproduction speed of the cells in the cellular automaton.

[0093] The most commonly used product for judging the intensity of weather radar echoes is the radar's combined reflectivity. An increase in the maximum value of the combined reflectivity indicates that the thunderstorm cloud is developing and strengthening. To judge the movement of thunderstorm clouds, the movement of strong echo areas (areas with combined reflectivity > 30dBZ) is often used to indicate this. The average moving speed and average moving direction of the strong echo area are the moving speed and direction of the thunderstorm clouds.

[0094] The step (4) is specifically as follows:

[0095] 4.1 Radar data feature analysis: Analyze the evolution trend of the intensity, moving speed and moving direction of the combined reflectivity of the weather radar.

[0096] 4.2 Establish breeding rules:

[0097] Reproduction capacity: When the radar echo is enhanced, the strength of the combined reflectivity of the radar echo is judged. If the maximum value of the combined reflectivity is increasing, it means that the thunderstorm cloud is developing and strengthening. The development of thunderstorm clouds is often accompanied by an increase in the number of lightning occurrences, indicating that thunderstorm activity is intensifying. At this time, the reproduction capacity of the cellular automaton is also enhanced accordingly; on the contrary, if the maximum value of the combined reflectivity is decreasing, it means that the thunderstorm cloud is weakening, and the number of lightning occurrences will also decrease accordingly;

[0098] Reproduction speed: The reproduction speed of the cells matches the movement speed of the radar echo. The faster the radar echo moves, the faster the cells reproduce.

[0099] Reproduction direction: The cell's reproduction direction is consistent with the direction of the radar echo. For example, if the radar echo moves to the southeast, the cell will also reproduce in the southeast direction, but not in other directions.

[0100] Dynamic adjustment: According to the real-time update of radar data, the behavior rules of the cellular automaton are dynamically adjusted to ensure that the model's predictions are synchronized with the evolution of the actual weather system.

[0101] According to the evolution trend of the combined reflectivity intensity, moving speed and moving direction of weather radar, the behavior rules of cellular automata are formulated. When the radar echo is enhanced, the reproduction capacity of cellular automata is also enhanced, such as the original reproduction of 3 at a time, now it reproduces 4 or more at a time; the reproduction speed and proliferation direction of cells are also consistent with the moving direction and speed of radar echoes. For example, when the radar echo moves to the southeast, the cells also reproduce to the southeast, and do not reproduce in other directions.

[0102] The step (5) is specifically as follows:

[0103] 5.1 Initialization prediction: The initial state of the cellular automaton model is determined by the actual lightning location data that occurred in the last 6 minutes; these data provide the starting point for the model prediction.

[0104] 5.2 Applying behavioral rules: Based on the behavioral rules formulated in step (4), including reproduction capacity, reproduction speed and reproduction direction, the cellular automaton model begins to simulate the development of the lightning strike zone; 5.3 Performing short-term prediction: The model first performs a short-term prediction for the next 6 minutes to determine the possible location of the lightning strike zone in the short term;

[0105] 5.4 Perform medium-term forecast: The model uses the same behavioral rules to perform a second 6-minute forecast to determine the medium-term lightning strike area.

[0106] 5.5 Long-term forecast: As needed, longer forecasts such as 12 minutes, 18 minutes or longer are conducted to provide longer-term predictions of lightning strike areas.

[0107] 5.6 Iteration process: The cellular automaton model is continuously iterated, and each new 6-minute period becomes the input data for a new round of predictions. The model dynamically adjusts the behavior rules based on the latest radar data and previous prediction results to achieve continuous predictions.

[0108] The step (6) is specifically as follows:

[0109] 6.1 Determine the prediction area: Determine the lightning strike area based on the distribution of breeding cells predicted by the cellular automaton model; breeding cells represent the area where lightning may occur during the prediction period.

[0110] 6.2 Drawing ellipses: Draw ellipses along the periphery of the reproduced cells. The areas enclosed by these ellipses are the predicted lightning strike areas. The major and minor axes of the ellipses are adjusted according to the density and range of the cell distribution to cover all possible lightning strike points.

[0111] 6.3 Graphical representation: Use graphics software or GIS tools to draw these ellipses on a map to visually display the predicted lightning strike area;

[0112] 6.4 Real-time display: The predicted lightning strike area is displayed in real time on the terminal screen, such as a computer monitor, console or other display device. This allows users to obtain lightning warning information immediately.

[0113] 6.5 Update prediction: As time goes by and new data is input, the cellular automaton model is continuously iterated, and the prediction results are updated accordingly. The updated prediction results will be continuously displayed on the terminal screen to reflect the latest prediction information;

[0114] Interpretation of results: The predicted lightning strike area can provide decision support for weather forecasters, help them issue lightning warnings in a timely manner, and guide the public and relevant departments to take preventive measures.

[0115] The data mentioned in step 3.1 include the strength, position, moving speed and direction information of the radar echo.

[0116] The present invention also relates to a system for realizing a lightning strike zone prediction method based on an improved cellular automation, comprising:

[0117] Data acquisition module 1, used to obtain real-time lightning location data and combined reflectivity and thunderstorm movement speed data of the forecast area by weather radar;

[0118] A gridding module 2, used for gridding the acquired data and the geographical area;

[0119] Real-time data analysis module 3, used to analyze real-time data of weather radar;

[0120] Behavior rule formulation module 4, used to formulate dynamic behavior rules of cellular automata;

[0121] Prediction module 5, used to carry out prediction work based on the established behavioral rules;

[0122] The result output module 6 is used to output the prediction result.

[0123] In the embodiment of the present invention, a severe thunderstorm weather process occurred in Shandong on June 10, 2016, and the selected data is the lightning location data of the lightning location monitoring network of the Shandong Meteorological Department. The improved cellular automation is used to predict the lightning strike area. After inspection and evaluation, the predicted lightning strike area is basically consistent with the actual lightning strike area in terms of position, trend and orientation.

[0124] Figure 6 These are the lightning locations from 00:00 to 5 minutes on June 10, 2016. The data comes from the lightning location network built by the Shandong Provincial Meteorological Bureau.

[0125] Figure 7 It is the state after the lightning data is gridded, which is also the original data for predicting lightning.

[0126] Figure 8 This is the result of the first cell reproduction, i.e. the prediction of the next 6 minutes. The color of the grid points is red for correct, blue for false, and green for missed. The area enclosed by the ellipse is the prediction area.

[0127] Fig. 9 This is the result of the second cell reproduction, i.e. the prediction of the next 12 minutes. The color of the grid points is red for correct, blue for false, and green for missed. The area enclosed by the ellipse is the prediction area.

[0128] After comparison and verification with the actual situation, the predicted hit rate score reached more than 80%, which can basically meet the business needs of lightning prediction.

[0129] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A lightning strike zone prediction method based on improved cellular automata, characterized in that: The following steps are involved: (1) Data acquisition: Acquire real-time lightning location data and combined reflectivity and thunderstorm movement speed data of the forecast area; (2) gridding the acquired data and geographic areas; (3) Analyze real-time data from weather radar; (4) Formulate dynamic behavior rules for cellular automata; (5) Carry out forecasting work based on the established behavioral rules; (6) Output of prediction results.

2. The lightning strike zone prediction method based on improved cellular automation according to claim 1, characterized in that: The step (2) is specifically as follows: 2.1 Regional grid division: The area to be predicted is evenly divided into grids to ensure that each grid covers the same geographical area; 2.2 Lightning zone marking: If lightning occurs and hits a grid at a certain time point, the grid is marked as a lightning zone; this marking indicates that a lightning event has occurred in the grid and is the basis for the initial cell in the cellular automaton model; 2.3 Data synchronization: The radar system completes a comprehensive volume scan every six minutes to generate a new set of observation data. In order to make the input data of the model consistent with the actual radar observation data in time, the lightning location data and radar data within the most recent six-minute interval are selected as the initial input data of the model; 2.4 Initial data preparation: Using these six minutes of data, determine the initial minefield grid and use it as the starting state of the cellular automaton model.

3. The lightning strike zone prediction method based on improved cellular automation according to claim 1 or 2, characterized in that: The step (3) is specifically as follows: 3.1 Real-time data acquisition: By connecting to the weather radar system, the weather data detected by the radar can be obtained in real time; 3.2 Radar echo analysis: Based on the real-time data obtained in step 3.1, analyze the intensity evolution trend, moving speed and moving direction of the radar echo. These characteristics will serve as the basis for formulating the behavior rules of the cellular automation.

4. The method for predicting lightning strike areas based on improved cellular automata according to claim 1 or 2, characterized in that: The step (4) is specifically as follows: 4.1 Radar data feature analysis: Analyze the evolution trend of the intensity, moving speed and moving direction of the combined reflectivity of the weather radar. 4.2 Establish breeding rules: Reproduction capacity: When the radar echo is enhanced, the strength of the combined reflectivity of the radar echo is judged. If the maximum value of the combined reflectivity is increasing, it means that the thunderstorm cloud is developing and strengthening. The development of thunderstorm clouds is often accompanied by an increase in the number of lightning occurrences, indicating that thunderstorm activity is intensifying. At this time, the reproduction capacity of the cellular automaton is also enhanced accordingly; on the contrary, if the maximum value of the combined reflectivity is decreasing, it means that the thunderstorm cloud is weakening, and the number of lightning occurrences will also decrease accordingly; Reproduction speed: The reproduction speed of the cells matches the movement speed of the radar echo; Reproduction direction: The reproduction direction of the cell is consistent with the movement direction of the radar echo; Dynamic adjustment: According to the real-time update of radar data, the behavior rules of the cellular automaton are dynamically adjusted to ensure that the model's predictions are synchronized with the evolution of the actual weather system.

5. The method for predicting lightning strike areas based on improved cellular automation according to claim 1 or 2, characterized in that: The step (5) is specifically as follows: 5.1 Initialization prediction: The initial state of the cellular automaton model is determined by the actual lightning location data that occurred in the last 6 minutes; 5.2 Applying behavioral rules: According to the behavioral rules formulated in step (4), including reproduction capacity, reproduction speed and reproduction direction, the cellular automaton model begins to simulate the development of the lightning strike area; 5.3 Make short-term predictions: The model first makes a short-term prediction for the next 6 minutes to determine the possible location of the lightning strike area in the short term; 5.4 Perform medium-term forecast: The model uses the same behavioral rules to perform a second 6-minute forecast to determine the medium-term lightning strike area. 5.5 Long-term forecast: As needed, longer-term forecasts are conducted to provide longer-term predictions of lightning strike areas. 5.6 Iteration process: The cellular automaton model is continuously iterated, and each new 6-minute period becomes the input data for a new round of predictions.

6. The method for predicting lightning strike zones based on improved cellular automation according to claim 1 or 2, characterized in that the step (6) is specifically: 6.1 Determine the prediction area: Determine the lightning strike area based on the distribution of propagation cells predicted by the cellular automaton model; 6.2 Draw ellipses: Draw ellipses along the periphery of the propagated cells. The area enclosed by these ellipses is the predicted lightning strike area; 6.3 Graphical representation: Use graphics software or GIS tools to draw these ellipses on a map to visually display the predicted lightning strike area; 6.4 Real-time display: The predicted lightning strike area is displayed in real time on the terminal screen, so that users can immediately obtain lightning warning information. 6.5 Update prediction: As time goes by and new data is input, the cellular automaton model is continuously iterated, and the prediction results are updated accordingly. The updated prediction results will be continuously displayed on the terminal screen to reflect the latest prediction information.

7. The method for predicting lightning strike areas based on improved cellular automation according to claim 1 or 2, characterized in that: The data mentioned in step 3.1 include the strength, position, moving speed and direction information of the radar echo.

8. A system for implementing the lightning strike zone prediction method based on improved cellular automation as claimed in any one of claims 1 to 7, characterized in that: include: The data acquisition module is used to obtain the real-time lightning location data of the forecast area and the combined reflectivity and thunderstorm movement speed data of the weather radar; A gridding module is used to grid the acquired data and geographic areas; Real-time data analysis module, used to analyze real-time data from weather radar; Behavior rule formulation module, used to formulate dynamic behavior rules of cellular automata; The prediction module is used to carry out prediction work based on the established behavioral rules; The result output module is used to output the prediction results.

Images