An airport runway incursion risk assessment method and system based on artificial intelligence

Abstract

The present invention discloses an airport runway intrusion risk assessment method and system based on artificial intelligence, which relates to the field of artificial intelligence technology. The present invention constructs an intelligent assessment system for airport runway intrusion risk through refined management of high-altitude foreign object intrusion data in the dual dimensions of season and weather type, equipment performance time series modeling and multi-risk coupling assessment. The system stores historical high-altitude foreign object intrusion data in a hierarchical directory structure, dynamically predicts the number of high-altitude foreign objects in combination with a time series model and monitors errors in real time, significantly improving the accuracy of high-altitude foreign object intrusion risk prediction under different meteorological conditions. By analyzing the attenuation trend of equipment performance parameters and quantifying the impact of equipment aging on monitoring effects, the system converts equipment performance attenuation into risk increments and incorporates them into the assessment model, avoiding prevention and control misjudgments due to hidden equipment failures, thereby enhancing the risk warning capability of the airport runway intrusion risk assessment system in multiple scenarios.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an airport runway intrusion risk assessment method and system based on artificial intelligence. Background Art

[0002] In the area of airport runway intrusion prevention and control, high-altitude foreign object intrusions primarily include illegal intrusions into runway airspace by flying objects such as birds, drones, and balloons. Bird strikes, due to the biological characteristics of bird activity and their adaptability to the environment, have become a core risk threat to aviation safety. Existing bird-repelling technology systems have significant flaws: On the one hand, the long-term reliance on fixed-pattern sound and light bird-repelling signals, such as fixed-frequency ultrasound or looped playback of the calls of bird predators, has led to the gradual adaptive evolution of bird populations' behavioral patterns. On the other hand, equipment maintenance relies on regular manual inspections and lacks online status monitoring and fault warning mechanisms, making it difficult to detect equipment performance degradation or component failure in real time. This results in a nonlinear decline in prevention and control effectiveness as the equipment ages. Especially under extreme weather conditions, such as strong winds and heavy rain, prevention and control effectiveness declines significantly, leading to increased inaccuracy in bird strike risk assessment models and a significant increase in prevention and control operation and maintenance costs. Summary of the Invention

[0003] The purpose of the present invention is to provide an airport runway incursion risk assessment method and system based on artificial intelligence to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an artificial intelligence-based airport runway incursion risk assessment method, the airport runway incursion risk assessment method comprising the following steps:

[0005] Step S1: Divide the high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area according to season and weather type, obtain the high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area in different years, and construct a high-altitude foreign object intrusion data set after dividing it according to season and weather type;

[0006] Step S1-1: Setting a high-altitude monitoring cycle to monitor the airport runway. The airport runway includes multiple airport high-altitude control and surveillance zones. Any one of the airport high-altitude control and surveillance zones is selected as the research object. The airport high-altitude control and surveillance zone is a specific spatial area within the airport that is specially designated for monitoring and driving away foreign objects in the air to prevent them from threatening the safety of the airport runway.

[0007] Step S1-2: The airport high-altitude air traffic control surveillance area has the same number of seasons in different years, and the weather types in each season in the airport high-altitude air traffic control surveillance area are also the same in different years. The weather types are formed by the seasonal changes in solar radiation angles and the seasonal displacement of pressure and wind belts to form fixed weather types, including sky conditions and precipitation types.

[0008] The seasons are divided into an equal number of seasonal units based on the Earth's revolution period and changes in solar radiation angles according to the law of seasonal changes. Different years are divided into four identical seasons: spring, summer, autumn, and winter, and the weather types in the corresponding seasons of different years remain consistent.

[0009] The sky conditions are consistent with the weather types in the weather forecast, including sunny, cloudy and overcast, and the precipitation types include solid, liquid and mixed states; the solid state includes light snow, moderate snow, heavy snow, hail and sleet; the liquid state includes light rain, moderate rain, heavy rain, torrential rain and thunderstorms; the mixed state includes sleet and freezing rain; the weather types include two major categories, and the sunny, cloudy and overcast, as well as light snow, moderate snow, heavy snow, hail, sleet, light rain, moderate rain, heavy rain, torrential rain, thunderstorms, sleet and freezing rain under the two major categories belong to the same level.

[0010] Step S1-3: Acquire high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area through electromagnetic sensors, and construct a high-altitude foreign object intrusion data set. The high-altitude foreign object intrusion data is represented by the number of high-altitude foreign objects appearing in a high-altitude monitoring cycle in the airport high-altitude control and surveillance area and the high-altitude monitoring cycle to which they belong; the high-altitude foreign object intrusion data set is stored in a hierarchical directory structure: with the season as the root directory, sub-directories are established under the season directory according to weather type, and the high-altitude foreign object intrusion data are stored in the sub-directories.

[0011] By storing data in a hierarchical directory structure based on seasons and weather types, including sky conditions and precipitation types, we achieve refined classification and management of high-altitude foreign object intrusion data. This facilitates rapid retrieval of historical data under specific conditions and provides structured data support for analyzing high-altitude foreign object intrusion patterns, such as bird flocking activity, and evaluating equipment performance.

[0012] Step S2: Based on the current season, obtain the high-altitude foreign object intrusion data for the corresponding season from the high-altitude foreign object intrusion data set, and predict the number of high-altitude foreign objects invading each airport high-altitude control and surveillance zone on the airport runway in combination with the weather forecast. The predicted number of high-altitude foreign objects is recorded as the predicted value of the number of high-altitude foreign objects; at the same time, obtain the actual number of high-altitude foreign objects invading the airport high-altitude control and surveillance zone, and calculate the error between the predicted value and the actual value;

[0013] Step S2-1: According to the season of the current high-altitude monitoring cycle, match the root directory of the high-altitude foreign object intrusion data set to obtain data information of the subdirectory;

[0014] Step S2-2, further matching the sub-directories through the weather forecast to obtain historical high-altitude foreign object intrusion data corresponding to the weather type;

[0015] Step S2-3: using a time series model based on the acquired historical high-altitude foreign object intrusion data to predict the number of high-altitude foreign objects appearing in the airport high-altitude control surveillance area, which is recorded as the high-altitude foreign object number prediction value;

[0016] Step S2-4: Subtract the predicted value of the number of high-altitude foreign objects from the number of high-altitude foreign objects obtained by the electromagnetic sensor to obtain the predicted error;

[0017] The formula used in the time series model forecast calculation is as follows:

[0018] ;

[0019] Where N dive，t It is expressed as the predicted value of the number of high-altitude foreign objects; A i Expressed as autoregressive coefficient; B j Expressed as moving average coefficient; W k，t A dummy variable representing weather type; J l，t It is represented as a dummy variable for season; z t is the forecast error of the most recent high-altitude monitoring cycle; p is the autoregressive order; q is the moving average order; m is the number of weather types; s is the number of seasons, which is consistent with the weather types divided in step S1-2; A i Expressed as autoregressive coefficient; B j Expressed as moving average coefficient; C k Expressed as the weight coefficient of weather; D l Expressed as the seasonal weight coefficient.

[0020] A dual matching mechanism of season and weather type is used to predict the number of high-altitude foreign objects and calculate errors. First, the root directory of the high-altitude foreign object intrusion data set is located based on the current season. Historical data for the corresponding weather type is then matched with weather forecasts. A time series model is used to predict the number of high-altitude foreign objects. Simultaneously, electromagnetic sensors are used to obtain the actual number of bird flocks in real time, and the prediction error is calculated. This method, by integrating historical data with real-time monitoring data, effectively improves the accuracy of high-altitude foreign object predictions.

[0021] Step S3: Classify and store the errors according to season and weather type to construct an error analysis set, extract all error historical data corresponding to the predicted value of the number of high-altitude foreign objects, analyze and set the error threshold, and calculate the high-altitude foreign object risk value of the airport high-altitude control monitoring area based on the error threshold and the predicted value of the number of high-altitude foreign objects;

[0022] Step S3-1: The storage structure of the error analysis set is synchronized with the storage structure of the high-altitude foreign object intrusion data set, and the predicted errors are stored in the subdirectory;

[0023] Step S3-2: Match the error analysis set based on the season of the current high-altitude monitoring period and the weather forecast information, extract the sub-directory data information corresponding to the weather type, and set the error threshold based on the mean and standard deviation. The fluctuation coefficient is determined based on the airport risk preference, which represents the airport management department's acceptance of the risk of high-altitude foreign object intrusion;

[0024] The error threshold is calculated using the following formula:

[0025] ;

[0026] Where z max It is represented as the error threshold; z is the average error of the sub-directory data information corresponding to the weather type matching the season of the current high-altitude monitoring cycle; e is the fluctuation coefficient; b is the standard deviation of the error of the sub-directory data information corresponding to the weather type matching the season of the current high-altitude monitoring cycle;

[0027] Step S3-3: Calculate the high-altitude foreign object risk value of the airport high-altitude control monitoring area using the high-altitude foreign object quantity prediction value and the error threshold. The high-altitude foreign object risk value calculation uses the following formula:

[0028] ;

[0029] Where R bird It is expressed as the risk value of high-altitude foreign objects in the airport high-altitude control monitoring area; N dive It is expressed as the predicted value of the number of high-altitude foreign objects; w weather Expressed as the risk weight coefficient of weather type; u avg It is expressed as the average error rate of the predicted value of the number of high-altitude foreign objects under the same weather type; u max It is expressed as the error threshold of the predicted value of the number of high-altitude foreign objects under the same weather type;

[0030] Step S3-4: Calculate the weather forecast deviation index based on the actual weather information and the weather forecast, calculate the relative error ratio by the difference between each forecast value and the corresponding actual value, and obtain the weather forecast deviation index of the weather type;

[0031] The weather forecast deviation index is calculated using the following formula:

[0032] ;

[0033] Where Pweather is the weather forecast deviation index of the weather type; n is the sample size of the forecast value of the weather forecast, including temperature, precipitation and wind volume; Fi is the i-th forecast value of the weather forecast; Mi is the actual value in the actual weather;

[0034] Step S3-5: Use the calculated weather forecast deviation index as an adjustment factor to adjust the risk weight coefficient w of the weather type. weather Make corrections and set the weather risk preference coefficient according to the airport management department's tolerance for airport risks. The weather risk preference coefficient represents the degree of importance the airport attaches to weather forecast deviations. The weather risk preference coefficient is multiplied by the value of the weather forecast deviation index plus 1 to obtain the adjustment coefficient, which is then multiplied by the risk weight coefficient w of the original weather type. weather Get the modified weather type risk weight coefficient w weather，new ;

[0035] The revised calculation of the weather type risk weight coefficient uses the following formula:

[0036] ;

[0037] Where w weather，new It is expressed as the modified weather type risk weight coefficient; f is expressed as the weather risk preference coefficient, which is determined by quantitatively analyzing the probability and degree of bird strike risk under different weather types and combining it with the airport's risk tolerance, with a value range of 0-2;

[0038] By constructing an error analysis set based on the dual dimensions of season and weather type, a dynamic assessment of high-altitude foreign object intrusion risk is achieved: First, the prediction errors are categorized and stored according to the same hierarchical structure as the high-altitude foreign object intrusion data set. Then, historical error data for the current season and corresponding weather type are combined to dynamically set error thresholds based on the mean, standard deviation, and airport risk preference. Finally, the high-altitude foreign object risk value is calculated using the predicted number of high-altitude foreign object intrusions, the weather weight coefficient, the average error rate, and the error threshold. This method effectively quantifies the high-altitude foreign object risk under different weather conditions through structured analysis of historical error data and a dynamic threshold adjustment mechanism.

[0039] Step S4: Acquire the performance parameters of the high-altitude foreign object monitoring equipment, construct a performance degradation trend graph of the equipment, perform analysis and prediction based on the performance degradation trend graph of the high-altitude foreign object monitoring equipment, and calculate the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control surveillance area based on the index standards of the high-altitude foreign object monitoring equipment performance parameters and the analysis and prediction results;

[0040] Step S4-1, obtaining performance parameters of high-altitude foreign object monitoring equipment in multiple historical high-altitude monitoring cycles, wherein the performance parameters include optical high-altitude foreign object monitoring equipment performance parameters and acoustic high-altitude foreign object monitoring equipment performance parameters; the optical high-altitude foreign object monitoring equipment performance parameters include light intensity and spectral wavelength; the acoustic high-altitude foreign object monitoring equipment performance parameters include sound pressure level and frequency range;

[0041] Step S4-2: Use the performance parameter values of the high-altitude foreign object monitoring equipment as the vertical coordinate to establish the y-axis, and use the time series data of the high-altitude monitoring period as the horizontal coordinate to establish the x-axis to construct a device performance decline trend graph.

[0042] Step S4-3: Analyze and predict the performance parameters of each high-altitude foreign object monitoring device using a time series model to obtain predicted values of the performance parameters of each high-altitude foreign object monitoring device, which are recorded as performance predicted values;

[0043] Step S4-4: Calculate the performance risk value of the high-altitude foreign object monitoring equipment for the airport high-altitude control surveillance zone through a high-altitude foreign object monitoring equipment performance risk assessment based on the index standards and performance prediction values of the high-altitude foreign object monitoring equipment performance parameters. The specific calculation process is as follows: subtract the performance prediction value from the index standards of the high-altitude foreign object monitoring equipment performance parameters, divide the difference by the index standards, and then multiply the difference by the risk weight of the high-altitude foreign object monitoring equipment performance parameters to obtain the risk value of the individual high-altitude foreign object monitoring equipment performance parameters. Add the risk values of the optical high-altitude foreign object monitoring equipment performance parameters and the acoustic high-altitude foreign object monitoring equipment performance parameters to obtain the high-altitude foreign object monitoring equipment performance risk value for the airport high-altitude control surveillance zone.

[0044] The performance risk value of high-altitude foreign object monitoring equipment is calculated using the following formula:

[0045] ;

[0046] Where R device It is represented by the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control and monitoring area; n is represented by the number of high-altitude foreign object monitoring equipment in the airport high-altitude control and monitoring area; d i It is represented by the performance parameter weight of the i-th high-altitude foreign object monitoring equipment in the airport high-altitude control monitoring area; S i It is expressed as the lower limit of the performance parameter standard of the ith high-altitude foreign object monitoring equipment in the airport high-altitude control monitoring area; P i It is expressed as the performance prediction value of the i-th high-altitude foreign object monitoring equipment in the airport high-altitude control surveillance area;

[0047] By collecting historical performance parameters, modeling time-series trends, and developing risk assessment models, we achieved a quantitative analysis of the performance risks of high-altitude foreign object monitoring equipment. First, we acquired historical parameters such as light intensity and sound pressure level for optical and acoustic equipment, and constructed a time-series downward trend chart. We then used a time series model to predict future parameter values. Finally, we calculated the equipment performance risk value based on the deviation rate between the indicator standard and the predicted value, combined with parameter weights. This method provides a data-driven decision-making basis for equipment maintenance and effectively identifies potential failure risks.

[0048] Step S5: Calculate the regional comprehensive risk value of the airport high-altitude control surveillance zone intrusion based on the high-altitude foreign object risk value and the high-altitude foreign object monitoring equipment performance risk value through weighted fusion, and calculate the comprehensive risk value of the airport runway intrusion based on the regional comprehensive risk value of each airport high-altitude control surveillance zone;

[0049] The regional comprehensive risk value is calculated based on the high-altitude foreign object risk value in the airport high-altitude control and surveillance area and the performance risk value of the high-altitude foreign object monitoring equipment. The regional comprehensive risk value of the airport runway intrusion is obtained by adding up the regional comprehensive risk values of multiple airport high-altitude control and surveillance areas contained in the airport.

[0050] By weightedly integrating bird strike risk and equipment performance risk, a comprehensive quantitative assessment of the risk of incursions into an airport's high-altitude control surveillance zone is achieved. This approach considers both the dynamic threat of bird flocking and the potential degradation of equipment performance. By overlaying regional risk values, a global risk map for runway incursions is formed. This provides a multi-dimensional risk coupling analysis for airport safety management, effectively improving the comprehensiveness and accuracy of risk assessments.

[0051] Furthermore, an AI-based airport runway incursion risk assessment system includes a data management module, a prediction and analysis module, a high-altitude risk module, an equipment monitoring module, and a comprehensive assessment module;

[0052] The data management module is used to classify and store high-altitude foreign object intrusion data by season and weather type; the prediction analysis module is used to predict the number of high-altitude foreign objects and calculate the prediction error; the high-altitude risk module is used to analyze the error and calculate the high-altitude foreign object risk value; the equipment monitoring module is used to evaluate the performance risk of high-altitude foreign object monitoring equipment; and the comprehensive evaluation module is used to integrate multi-dimensional risks and output the final result.

[0053] The output end of the data management module is electrically connected to the input end of the prediction and analysis module; the output end of the prediction and analysis module is electrically connected to the input end of the high-altitude risk module; the output end of the high-altitude risk module is electrically connected to the input end of the equipment monitoring module; the output end of the equipment monitoring module is electrically connected to the input end of the comprehensive assessment module;

[0054] The data management module includes a seasonal weather classification unit and a hierarchical data storage unit; the seasonal weather classification unit is used to divide the historical data of the airport high-altitude control monitoring area according to season and weather type; the hierarchical data storage unit is used to construct a high-altitude foreign object intrusion data set with season as the root directory and weather type as the subdirectory;

[0055] The prediction and analysis module includes a time series model prediction unit and an error calculation and matching unit; the time series model prediction unit is used to predict the number of high-altitude foreign objects in the current season and weather type based on historical high-altitude foreign object intrusion data; the error calculation and matching unit is used to generate error data based on the difference between the actual number of high-altitude foreign objects and the predicted value;

[0056] The high-altitude risk module includes a threshold dynamic setting unit and a risk coefficient calculation unit; the threshold dynamic setting unit is used to set the error threshold according to historical error data and airport risk preference; the risk coefficient calculation unit is used to calculate the high-altitude foreign object risk value based on the high-altitude foreign object number prediction value, the error threshold and the weather weight;

[0057] The equipment monitoring module includes a performance trend modeling unit and a parameter risk assessment unit; the performance trend modeling unit is used to construct an equipment performance degradation trend graph and predict parameters through time series data; the parameter risk assessment unit is used to calculate the equipment performance risk value based on the indicator standard and the predicted value;

[0058] The comprehensive assessment module includes a risk weight fusion unit and a global risk output unit; the risk weight fusion unit is used to weight the high-altitude foreign object risk and equipment risk to calculate the regional comprehensive risk value; the global risk output unit is used to summarize the regional comprehensive risk values of each airport's high-altitude control monitoring area to obtain the airport runway intrusion comprehensive risk value.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. This invention achieves a deep coupling of analysis between the patterns of high-altitude foreign object intrusions, such as bird flock activity, and equipment performance through refined management of high-altitude foreign object intrusion data across both seasonal and weather types. By storing historical data in a hierarchical directory structure, combining it with a time series model to predict the number of high-altitude foreign objects and calculate errors in real time, this effectively improves the accuracy of high-altitude foreign object predictions across different seasons and meteorological conditions. This provides data support for the dynamic adjustment of high-altitude monitoring strategies and significantly reduces the risk of control failures caused by the adaptive evolution of high-altitude foreign object intrusions, such as bird flocks.

[0061] 2. This invention reveals the inherent correlation between equipment performance degradation and the increased risk of high-altitude foreign object intrusion through modeling the time series trends of equipment performance parameters and quantitative risk assessment. By analyzing the historical data and downward trends of parameters such as light intensity and sound pressure level of optical and acoustic equipment, the impact of equipment performance degradation on high-altitude monitoring effectiveness is quantified in real time. Through risk weight calculation, equipment performance degradation is converted into an incremental risk of high-altitude foreign object intrusion, thereby dynamically incorporating equipment aging factors into the risk assessment model, avoiding misjudgments of prevention and control capabilities due to the hidden decline in equipment performance, and significantly improving the comprehensiveness of airport runway intrusion risk assessment and the timeliness of early warning.

[0062] 3. This invention constructs a global dynamic map of airport runway incursion risk through a weighted fusion assessment of high-altitude foreign object risk and equipment risk. By comprehensively considering the dual impact of high-altitude foreign object count prediction errors and equipment performance degradation, and superimposing risk values from multiple monitoring areas, it generates an airport-level comprehensive risk index. This provides a multi-dimensional decision-making basis for airport safety management, optimizes resource allocation, enhances prevention and control capabilities in extreme weather conditions, and comprehensively improves the systematicity and accuracy of runway incursion risk assessments. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of an airport runway incursion risk assessment method based on artificial intelligence according to the present invention;

[0064] Figure 2 This is a structural schematic diagram of an airport runway intrusion risk assessment system based on artificial intelligence in the present invention. DETAILED DESCRIPTION

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0066] Example 1: Figure 1 As shown, the present invention provides a technical solution, an airport runway incursion risk assessment method based on artificial intelligence, and the airport runway incursion risk assessment method includes the following steps:

[0067] Step S1: Divide the high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area according to season and weather type, obtain the high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area in different years, and construct a high-altitude foreign object intrusion data set after dividing it according to season and weather type;

[0068] Step S1-1: Setting a high-altitude monitoring cycle to monitor the airport runway. The airport runway includes multiple airport high-altitude control and surveillance zones. Any one of the airport high-altitude control and surveillance zones is selected as the research object. The airport high-altitude control and surveillance zone is a specific spatial area within the airport that is specially designated for monitoring and driving away foreign objects in the air to prevent them from threatening the safety of the airport runway.

[0069] Step S1-2: The airport high-altitude air traffic control surveillance area has the same number of seasons in different years, and the weather types in each season in the airport high-altitude air traffic control surveillance area are also the same in different years. The weather types are formed by the seasonal changes in solar radiation angles and the seasonal displacement of pressure and wind belts to form fixed weather types, including sky conditions and precipitation types.

[0070] Step S1-3: Acquire high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area through electromagnetic sensors, and construct a high-altitude foreign object intrusion data set. The high-altitude foreign object intrusion data is represented by the number of high-altitude foreign objects appearing in a high-altitude monitoring cycle in the airport high-altitude control and surveillance area and the high-altitude monitoring cycle to which they belong; the high-altitude foreign object intrusion data set is stored in a hierarchical directory structure: with the season as the root directory, sub-directories are established under the season directory according to weather type, and the high-altitude foreign object intrusion data are stored in the sub-directories.

[0071] In specific implementation, the high-altitude foreign object intrusion data will be divided into two dimensions according to seasons and weather types (including sky conditions and precipitation patterns) based on the Earth's revolution period and the changing laws of solar radiation angles. A hierarchical storage structure will be constructed with seasons as the root directory and weather types as subdirectories to achieve temporal and spatial association and rapid retrieval of data. The consistency of seasonal divisions in different years must be ensured, such as the fixed use of the Gregorian calendar's four seasons of spring, summer, autumn and winter. The classification of weather types must strictly follow meteorological standards so that the divided weather types can match the weather types given in the weather forecast. At the same time, the deployment of electromagnetic sensors must cover the entire monitoring area to avoid blind spots in data collection.

[0072] Step S2: Based on the current season, obtain the high-altitude foreign object intrusion data for the corresponding season from the high-altitude foreign object intrusion data set, and predict the number of high-altitude foreign objects invading each airport high-altitude control and surveillance zone on the airport runway in combination with the weather forecast. The predicted number of high-altitude foreign objects is recorded as the predicted value of the number of high-altitude foreign objects; at the same time, obtain the actual number of high-altitude foreign objects invading the airport high-altitude control and surveillance zone, and calculate the error between the predicted value and the actual value;

[0073] Step S2-1: According to the season of the current high-altitude monitoring cycle, match the root directory of the high-altitude foreign object intrusion data set to obtain data information of the subdirectory;

[0074] Step S2-2, further matching the sub-directories through the weather forecast to obtain historical high-altitude foreign object intrusion data corresponding to the weather type;

[0075] Step S2-3: using a time series model based on the acquired historical high-altitude foreign object intrusion data to predict the number of high-altitude foreign objects appearing in the airport high-altitude control surveillance area, which is recorded as the high-altitude foreign object number prediction value;

[0076] Step S2-4: Subtract the predicted value of the number of high-altitude foreign objects from the number of high-altitude foreign objects obtained by the electromagnetic sensor to obtain the predicted error.

[0077] In specific implementation, taking bird strikes as an example, based on the season-weather dual matching mechanism, time series models (such as LSTM) are used to learn the bird flock activity patterns in historical data and predict the number of bird strikes in the current period, while calculating the prediction error through real-time monitoring data; the model training needs to differentiate the weights of different weather types (such as heavy rain has a higher weight than sunny days), and ensure that the real-time data is synchronized with the timestamp of the prediction period.

[0078] Step S3: Classify and store the errors according to season and weather type to construct an error analysis set, extract all error historical data corresponding to the predicted value of the number of high-altitude foreign objects, analyze and set the error threshold, and calculate the high-altitude foreign object risk value of the airport high-altitude control monitoring area based on the error threshold and the predicted value of the number of high-altitude foreign objects;

[0079] Step S3-1: The storage structure of the error analysis set is synchronized with the storage structure of the high-altitude foreign object intrusion data set, and the predicted errors are stored in the subdirectory;

[0080] Step S3-2: Match the error analysis set based on the season of the current high-altitude monitoring period and the weather forecast information, extract the sub-directory data information corresponding to the weather type, and set the error threshold based on the mean and standard deviation. The fluctuation coefficient is determined based on the airport risk preference, which represents the airport management department's acceptance of the risk of high-altitude foreign object intrusion;

[0081] Step S3-3: Calculate the high-altitude foreign object risk value of the airport high-altitude control monitoring area using the high-altitude foreign object quantity prediction value and the error threshold. The high-altitude foreign object risk value calculation uses the following formula:

[0082] ;

[0083] Where R bird It is expressed as the risk value of high-altitude foreign objects in the airport high-altitude control monitoring area; N dive It is expressed as the predicted value of the number of high-altitude foreign objects; w weather Expressed as the risk weight coefficient of weather type; u avgIt is expressed as the average error rate of the predicted value of the number of high-altitude foreign objects under the same weather type; u max It is expressed as the error threshold of the predicted value of the number of high-altitude foreign objects under the same weather type;

[0084] Step S3-4: Calculate the weather forecast deviation index based on the actual weather information and the weather forecast, calculate the relative error ratio by the difference between each forecast value and the corresponding actual value, and obtain the weather forecast deviation index of the weather type;

[0085] Step S3-5: Use the calculated weather forecast deviation index as an adjustment factor to adjust the risk weight coefficient w of the weather type. weather Make corrections and set the weather risk preference coefficient according to the airport management department's tolerance for airport risks. The weather risk preference coefficient represents the degree of importance the airport attaches to weather forecast deviations. The weather risk preference coefficient is multiplied by the value of the weather forecast deviation index plus 1 to obtain the adjustment coefficient, which is then multiplied by the risk weight coefficient w of the original weather type. weather Get the modified weather type risk weight coefficient w weather，new ;

[0086] In specific implementation, taking bird strikes as an example, the prediction errors are classified and stored according to the same structure as the high-altitude foreign object intrusion data set, and the error threshold is dynamically set through statistical methods (mean, standard deviation). The volatility coefficient is adjusted in combination with the airport risk preference to quantify the bird strike risk under weather conditions. The rationality of the threshold needs to be verified (such as based on a 95% confidence interval), and the weather weight coefficient needs to be calibrated according to the correlation between historical bird strike events and weather (such as thunderstorm weight is higher than light snow).

[0087] Step S4: Acquire the performance parameters of the high-altitude foreign object monitoring equipment, construct a performance degradation trend graph of the equipment, perform analysis and prediction based on the performance degradation trend graph of the high-altitude foreign object monitoring equipment, and calculate the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control surveillance area based on the index standards of the high-altitude foreign object monitoring equipment performance parameters and the analysis and prediction results;

[0088] Step S4-1, obtaining performance parameters of high-altitude foreign object monitoring equipment in multiple historical high-altitude monitoring cycles, wherein the performance parameters include optical high-altitude foreign object monitoring equipment performance parameters and acoustic high-altitude foreign object monitoring equipment performance parameters; the optical high-altitude foreign object monitoring equipment performance parameters include light intensity and spectral wavelength; the acoustic high-altitude foreign object monitoring equipment performance parameters include sound pressure level and frequency range;

[0089] Step S4-2: Use the performance parameter values of the high-altitude foreign object monitoring equipment as the vertical coordinate to establish the y-axis, and use the time series data of the high-altitude monitoring period as the horizontal coordinate to establish the x-axis to construct a device performance decline trend graph.

[0090] Step S4-3: Analyze and predict the performance parameters of each high-altitude foreign object monitoring device using a time series model to obtain predicted values of the performance parameters of each high-altitude foreign object monitoring device, which are recorded as performance predicted values;

[0091] Step S4-4: Calculate the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control monitoring area through the high-altitude foreign object monitoring equipment performance risk assessment based on the index standards and performance prediction values of the performance parameters of the high-altitude foreign object monitoring equipment. The specific calculation process is: subtract the performance prediction value from the index standards of the performance parameters of the high-altitude foreign object monitoring equipment from the index standards, divide it by the index standards, and then multiply it by the risk weight of the performance parameters of the high-altitude foreign object monitoring equipment to obtain the risk value of the performance parameters of a single high-altitude foreign object monitoring equipment. Add the risk values of the performance parameters of the optical high-altitude foreign object monitoring equipment and the performance parameters of the acoustic high-altitude foreign object monitoring equipment to obtain the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control monitoring area.

[0092] In specific implementation, taking bird strikes as an example, by analyzing the time-series decline trend of optical and acoustic equipment performance parameters (light intensity, sound pressure level, etc.), building a prediction model and calculating the performance deviation rate, the impact of equipment aging on the bird repellent effect is quantified; it is necessary to establish a dynamic baseline of equipment performance parameters (such as the sound pressure level attenuation curve), and assign risk weights based on the importance of equipment functions (such as the light intensity weight is higher than the frequency range), and distinguish between normal attenuation and sudden failures (such as abnormal sensor data mutations).

[0093] Step S5: Calculate the regional comprehensive risk value of the airport high-altitude control surveillance zone intrusion based on the high-altitude foreign object risk value and the high-altitude foreign object monitoring equipment performance risk value through weighted fusion, and calculate the comprehensive risk value of the airport runway intrusion based on the regional comprehensive risk value of each airport high-altitude control surveillance zone;

[0094] The regional comprehensive risk value is calculated based on the high-altitude foreign object risk value in the airport high-altitude control and surveillance area and the performance risk value of the high-altitude foreign object monitoring equipment. The regional comprehensive risk value of the airport runway intrusion is obtained by adding up the regional comprehensive risk values of multiple airport high-altitude control and surveillance areas contained in the airport.

[0095] In specific implementation, taking bird strikes as an example, the airport-level comprehensive risk index is generated by weighted fusion of high-altitude foreign object risks and equipment performance risks, and superimposing multi-region risk values; the weight distribution needs to be reversed through historical accident data or expert review, and a separate weight distribution data table is established when the system is built.

[0096] Example 2, as Figure 2 As shown, the present invention provides an airport runway incursion risk assessment system based on artificial intelligence. The intelligent management system includes a data management module, a prediction and analysis module, a high-altitude risk module, an equipment monitoring module and a comprehensive assessment module;

[0097] The data management module is used to classify and store high-altitude foreign object intrusion data by season and weather type; the prediction analysis module is used to predict the number of high-altitude foreign objects and calculate the prediction error; the high-altitude risk module is used to analyze the error and calculate the high-altitude foreign object risk value; the equipment monitoring module is used to evaluate the performance risk of high-altitude foreign object monitoring equipment; and the comprehensive evaluation module is used to integrate multi-dimensional risks and output the final result.

[0098] The output end of the data management module is electrically connected to the input end of the prediction and analysis module; the output end of the prediction and analysis module is electrically connected to the input end of the high-altitude risk module; the output end of the high-altitude risk module is electrically connected to the input end of the equipment monitoring module; the output end of the equipment monitoring module is electrically connected to the input end of the comprehensive assessment module;

[0099] The data management module includes a seasonal weather classification unit and a hierarchical data storage unit; the seasonal weather classification unit is used to divide the historical data of the airport high-altitude control monitoring area according to season and weather type; the hierarchical data storage unit is used to construct a high-altitude foreign object intrusion data set with season as the root directory and weather type as the subdirectory;

[0100] The prediction and analysis module includes a time series model prediction unit and an error calculation and matching unit; the time series model prediction unit is used to predict the number of high-altitude foreign objects in the current season and weather type based on historical high-altitude foreign object intrusion data; the error calculation and matching unit is used to generate error data based on the difference between the actual number of high-altitude foreign objects and the predicted value;

[0101] The high-altitude risk module includes a threshold dynamic setting unit and a risk coefficient calculation unit; the threshold dynamic setting unit is used to set the error threshold according to historical error data and airport risk preference; the risk coefficient calculation unit is used to calculate the high-altitude foreign object risk value based on the high-altitude foreign object number prediction value, the error threshold and the weather weight;

[0102] The equipment monitoring module includes a performance trend modeling unit and a parameter risk assessment unit; the performance trend modeling unit is used to construct an equipment performance degradation trend graph and predict parameters through time series data; the parameter risk assessment unit is used to calculate the equipment performance risk value based on the indicator standard and the predicted value;

[0103] The comprehensive assessment module includes a risk weight fusion unit and a global risk output unit; the risk weight fusion unit is used to weight the high-altitude foreign object risk and equipment risk to calculate the regional comprehensive risk value; the global risk output unit is used to summarize the regional comprehensive risk values of each airport's high-altitude control monitoring area to obtain the airport runway intrusion comprehensive risk value.

[0104] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An artificial intelligence-based airport runway incursion risk assessment method, characterized by: The airport runway incursion risk assessment method comprises the following steps: Step S1: Divide the high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area according to season and weather type, obtain the high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area in different years, and construct a high-altitude foreign object intrusion data set after dividing it according to season and weather type; Step S2: Based on the current season, obtain the high-altitude foreign object intrusion data for the corresponding season from the high-altitude foreign object intrusion data set, and predict the number of high-altitude foreign objects invading each airport high-altitude control and surveillance zone on the airport runway in combination with the weather forecast. The predicted number of high-altitude foreign objects is recorded as the predicted value of the number of high-altitude foreign objects; at the same time, obtain the actual number of high-altitude foreign objects invading the airport high-altitude control and surveillance zone, and calculate the error between the predicted value and the actual value; Step S3: Classify and store the errors according to season and weather type to construct an error analysis set, extract all error historical data corresponding to the predicted value of the number of high-altitude foreign objects, analyze and set the error threshold, and calculate the high-altitude foreign object risk value of the airport high-altitude control monitoring area based on the error threshold and the predicted value of the number of high-altitude foreign objects; Step S4: Acquire the performance parameters of the high-altitude foreign object monitoring equipment, construct a performance degradation trend graph of the equipment, perform analysis and prediction based on the performance degradation trend graph of the high-altitude foreign object monitoring equipment, and calculate the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control surveillance area based on the index standards of the high-altitude foreign object monitoring equipment performance parameters and the analysis and prediction results; Step S5: Based on the high-altitude foreign object risk value and the high-altitude foreign object monitoring equipment performance risk value, the regional comprehensive risk value of the airport high-altitude control monitoring area intrusion is calculated through weighted fusion, and the comprehensive risk value of the airport runway intrusion is calculated based on the regional comprehensive risk value of each airport high-altitude control monitoring area.

2. The method for assessing airport runway incursion risk based on artificial intelligence according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1: Setting a high-altitude monitoring cycle to monitor the airport runway. The airport runway includes multiple airport high-altitude control and surveillance zones. Any one of the airport high-altitude control and surveillance zones is selected as the research object. The airport high-altitude control and surveillance zone is a specific spatial area within the airport that is specially designated for monitoring and driving away foreign objects in the air to prevent them from threatening the safety of the airport runway. Step S1-2: The airport high-altitude air traffic control monitoring area has the same number of seasons in different years, and the weather types in each season of the airport high-altitude air traffic control monitoring area are also the same in different years. The weather types are fixed due to the change of solar radiation angle caused by the change of seasons and the seasonal displacement of pressure belts and wind belts. The weather types are divided according to the weather forecast, including sky conditions and precipitation types. Step S1-3: Acquire high-altitude foreign object intrusion data in the airport high-altitude control and surveillance area through electromagnetic sensors, and construct a high-altitude foreign object intrusion data set. The high-altitude foreign object intrusion data is represented by the number of high-altitude foreign objects appearing in a high-altitude monitoring cycle in the airport high-altitude control and surveillance area and the high-altitude monitoring cycle to which they belong; the high-altitude foreign object intrusion data set is stored in a hierarchical directory structure: with the season as the root directory, sub-directories are established under the season directory according to weather type, and the high-altitude foreign object intrusion data are stored in the sub-directories.

3. The method for assessing airport runway incursion risk based on artificial intelligence according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: According to the season of the current high-altitude monitoring cycle, match the root directory of the high-altitude foreign object intrusion data set to obtain data information of the subdirectory; Step S2-2, further matching the sub-directories through the weather forecast to obtain historical high-altitude foreign object intrusion data corresponding to the weather type; Step S2-3: using a time series model based on the acquired historical high-altitude foreign object intrusion data to predict the number of high-altitude foreign objects appearing in the airport high-altitude control surveillance area, which is recorded as the high-altitude foreign object number prediction value; Step S2-4: Subtract the predicted value of the number of high-altitude foreign objects from the number of high-altitude foreign objects obtained by the electromagnetic sensor to obtain the predicted error.

4. The method for assessing airport runway incursion risk based on artificial intelligence according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1: The storage structure of the error analysis set is synchronized with the storage structure of the high-altitude foreign object intrusion data set, and the predicted errors are stored in the subdirectory; Step S3-2: Matching the error analysis set based on the season and weather forecast information of the current high-altitude monitoring period, specifically matching the same subdirectories under the same root directory; extracting the subdirectory data information corresponding to the weather type, and setting the error threshold based on the mean and standard deviation calculation, and the fluctuation coefficient is determined according to the airport risk preference, which represents the airport management department's acceptance of the risk of high-altitude foreign object intrusion; Step S3-3: Calculate the high-altitude foreign object risk value of the airport high-altitude control monitoring area using the high-altitude foreign object quantity prediction value and the error threshold. The high-altitude foreign object risk value calculation uses the following formula: ； Where R bird It is expressed as the risk value of high-altitude foreign objects in the airport high-altitude control monitoring area; N dive It is expressed as the predicted value of the number of high-altitude foreign objects; w weather The risk weight coefficient expressed as weather type; u avg It is expressed as the average error rate of the predicted value of the number of high-altitude foreign objects under the same weather type; u max It is expressed as the error threshold of the predicted value of the number of high-altitude foreign objects under the same weather type; Step S3-4: Calculate the weather forecast deviation index based on the actual weather information and the weather forecast, calculate the relative error ratio by the difference between each forecast value and the corresponding actual value, and obtain the weather forecast deviation index of the weather type; Step S3-5: Use the calculated weather forecast deviation index as an adjustment factor to adjust the risk weight coefficient w of the weather type. weather Make corrections and set a weather risk preference coefficient based on the airport management department's tolerance for airport risks. The weather risk preference coefficient represents the degree of importance the airport places on weather forecast deviations. The adjustment coefficient is obtained by multiplying the weather risk preference coefficient by the weather forecast deviation index plus 1, and then multiplying the adjustment coefficient by the risk weight coefficient w of the original weather type. weather Get the modified weather type risk weight coefficient w weather，new .

5. The method for assessing airport runway incursion risk based on artificial intelligence according to claim 4, characterized in that: The specific steps of step S4 are as follows: Step S4-1, obtaining performance parameters of high-altitude foreign object monitoring equipment in multiple historical high-altitude monitoring cycles, wherein the performance parameters include optical high-altitude foreign object monitoring equipment performance parameters and acoustic high-altitude foreign object monitoring equipment performance parameters; the optical high-altitude foreign object monitoring equipment performance parameters include light intensity and spectral wavelength; the acoustic high-altitude foreign object monitoring equipment performance parameters include sound pressure level and frequency range; Step S4-2: Use the performance parameter values of the high-altitude foreign object monitoring equipment as the vertical coordinate to establish the y-axis, and use the time series data of the high-altitude monitoring period as the horizontal coordinate to establish the x-axis to construct a device performance decline trend graph.

6. The method for airport runway incursion risk assessment based on artificial intelligence according to claim 5, characterized in that: In step S4, it also includes: Step S4-3: Analyze and predict the performance parameters of each high-altitude foreign object monitoring device using a time series model to obtain predicted values of the performance parameters of each high-altitude foreign object monitoring device, which are recorded as performance predicted values; Step S4-4: Calculate the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control monitoring area through the high-altitude foreign object monitoring equipment performance risk assessment based on the index standards and performance prediction values of the performance parameters of the high-altitude foreign object monitoring equipment. The specific calculation process is: subtract the performance prediction value from the index standards of the performance parameters of the high-altitude foreign object monitoring equipment from the index standards, divide it by the index standards, and then multiply it by the risk weight of the performance parameters of the high-altitude foreign object monitoring equipment to obtain the risk value of the performance parameters of a single high-altitude foreign object monitoring equipment. Add the risk values of the performance parameters of the optical high-altitude foreign object monitoring equipment and the performance parameters of the acoustic high-altitude foreign object monitoring equipment to obtain the performance risk value of the high-altitude foreign object monitoring equipment in the airport high-altitude control monitoring area.

7. The method for assessing airport runway incursion risk based on artificial intelligence according to claim 6, characterized in that: In step S5, a regional comprehensive risk value is calculated based on the high-altitude foreign object risk value in the airport high-altitude control monitoring area and the high-altitude foreign object monitoring equipment performance risk value, and the regional comprehensive risk values of multiple airport high-altitude control monitoring areas contained in the airport are added together to obtain the comprehensive risk value of the airport runway intrusion.

8. An artificial intelligence-based airport runway incursion risk assessment system, applied to the artificial intelligence-based airport runway incursion risk assessment method according to any one of claims 1 to 7, characterized in that: The airport runway incursion risk assessment system includes a data management module, a prediction analysis module, a high-altitude risk module, an equipment monitoring module and a comprehensive assessment module; The data management module is used to classify and store high-altitude foreign object intrusion data by season and weather type; the prediction analysis module is used to predict the number of high-altitude foreign objects and calculate the prediction error; the high-altitude risk module is used to analyze the error and calculate the high-altitude foreign object risk value; the equipment monitoring module is used to evaluate the performance risk of high-altitude foreign object monitoring equipment; and the comprehensive evaluation module is used to integrate multi-dimensional risks and output the final result. The output end of the data management module is electrically connected to the input end of the prediction and analysis module; the output end of the prediction and analysis module is electrically connected to the input end of the high-altitude risk module; the output end of the high-altitude risk module is electrically connected to the input end of the equipment monitoring module; the output end of the equipment monitoring module is electrically connected to the input end of the comprehensive evaluation module.

9. The artificial intelligence-based airport runway incursion risk assessment system according to claim 8, characterized in that: The data management module includes a seasonal weather classification unit and a hierarchical data storage unit; the seasonal weather classification unit is used to divide the historical data of the airport high-altitude control monitoring area according to season and weather type; the hierarchical data storage unit is used to construct a high-altitude foreign object intrusion data set with season as the root directory and weather type as the subdirectory; The prediction and analysis module includes a time series model prediction unit and an error calculation and matching unit; the time series model prediction unit is used to predict the number of high-altitude foreign objects in the current season and weather type based on historical high-altitude foreign object intrusion data; the error calculation and matching unit is used to generate error data based on the difference between the actual number of high-altitude foreign objects and the predicted value; The high-altitude risk module includes a threshold dynamic setting unit and a risk coefficient calculation unit; the threshold dynamic setting unit is used to set the error threshold based on historical error data and airport risk preference; the risk coefficient calculation unit is used to calculate the high-altitude foreign object risk value through the predicted value of the number of high-altitude foreign objects, the error threshold and the weather weight.

10. The artificial intelligence-based airport runway incursion risk assessment system according to claim 8, characterized in that: The equipment monitoring module includes a performance trend modeling unit and a parameter risk assessment unit; the performance trend modeling unit is used to construct an equipment performance degradation trend graph and predict parameters through time series data; The parameter risk assessment unit is used to calculate the equipment performance risk value based on the indicator standard and the predicted value; The comprehensive assessment module includes a risk weight fusion unit and a global risk output unit; The risk weight fusion unit is used to weight the high-altitude foreign object risk and the equipment risk to calculate the regional comprehensive risk value; The global risk output unit is used to summarize the regional comprehensive risk values of each airport high-altitude control surveillance zone to obtain the airport runway intrusion comprehensive risk value.

Images