Airport surface operation efficiency optimization system and method based on intelligent technology

Abstract

The present invention discloses an airport surface operation efficiency optimization system and method based on intelligent technology. The present invention realizes intelligent management of airport surface operations by setting up a taxiway data acquisition and prediction module, an aircraft status and scheduling data acquisition module, a taxiway resource real-time monitoring module, a path planning and optimization result generation module, and a global decision and feedback module. The taxiway congestion prediction algorithm based on time series analysis can predict taxiway usage in advance; the path planning and scheduling optimization algorithm based on reinforcement learning can dynamically generate the optimal taxiing path and scheduling plan; and the Bayesian decision network is used to make global operation optimization decisions and generate the global optimal solution. The present invention improves airport operation efficiency, reduces taxiway congestion, and reduces aircraft taxiing time.

Description

Technical Field

[0001] The present invention relates to the technical field of airport surface operation management and optimization, and in particular to an airport surface operation efficiency optimization system and method based on intelligent technology. Background Art

[0002] Existing airport surface operations systems primarily rely on manual monitoring and scheduling, often failing to efficiently and intelligently address complex issues such as taxiway resource allocation, aircraft scheduling, and taxiway path optimization. Especially during peak traffic periods, taxiway congestion, prolonged aircraft taxiing times, and scheduling conflicts severely impact the overall operational efficiency of airports.

[0003] While some airports have introduced automated dispatch systems, existing systems generally lack the ability to predict taxiway congestion and effectively allocate resources based on real-time and historical data. Furthermore, the lack of a system that can comprehensively analyze and optimize decision-making based on the overall state of airport operations makes it difficult to improve the overall efficiency of airport operations.

[0004] Therefore, there is an urgent need for a solution based on intelligent technology that can collect data in real time, predict congestion, optimize scheduling routes, and make global intelligent decisions. Summary of the Invention

[0005] The purpose of the invention is to provide an airport surface operation efficiency optimization system and method based on intelligent technology, which has the advantages of real-time monitoring, congestion prediction, path optimization and global decision-making, and solves the problems of taxiway congestion that cannot be predicted in advance and low efficiency in scheduling and path planning.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an airport surface operation efficiency optimization system based on intelligent technology, the system comprising the following modules:

[0007] Taxiway data collection and prediction module: used to collect real-time taxiway usage and historical data, and use the airport taxiway dynamic congestion prediction algorithm to predict the future taxiway congestion status based on the aircraft taxiing speed and occupancy, and output the taxiway congestion prediction result Y t Aircraft status and scheduling data acquisition module: used to obtain information about the aircraft's physical status, queue status, takeoff and landing schedules, and provide data on the aircraft's current location, taxiing status, and scheduling priority as input to the aircraft's path planning and scheduling optimization algorithms.

[0008] Taxiway resource real-time monitoring module: used to monitor taxiway occupancy and resource allocation status in real time, and compare the current taxiway usage with the taxiway congestion prediction result Y tMatching, providing real-time data support for path planning algorithms;

[0009] Path planning and optimization result generation module: through aircraft path planning and scheduling optimization algorithm, combined with taxiway congestion prediction result Y t and aircraft scheduling data to generate the aircraft's optimal taxiing path and scheduling plan Z t , ensuring efficient aircraft taxiing and avoiding congestion;

[0010] Global Evidence Data Collection and Processing Module: This module is used to collect other evidence data related to global operations, including weather conditions, fuel status, and taxiway resource status, providing support data for the airport's global intelligent decision-making and optimization algorithms.

[0011] Global decision-making and feedback module: Based on the airport's global intelligent decision-making and optimization algorithm, using the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t and other global evidence E, formulate a global optimization decision plan for airport surface operations, including the reallocation of taxiway resources and adjustment of aircraft scheduling priorities.

[0012] The method for optimizing airport surface operation efficiency based on intelligent technology includes the following steps:

[0013] S1. Taxiway data collection and congestion prediction:

[0014] S11: The taxiway data acquisition and prediction module is used to obtain the real-time taxiway usage, historical data, and aircraft taxiing speed information; S12: The collected data is pre-processed, including data cleaning, denoising, and formatting; S13: The airport taxiway dynamic congestion prediction algorithm is used to predict the future taxiway congestion based on real-time and historical data, and generate the taxiway congestion prediction result Y t ;

[0015] S2. Aircraft status and dispatch data collection:

[0016] S21: Obtain aircraft physical status data from the aircraft status and scheduling data acquisition module; S22: Obtain aircraft takeoff and landing plan information, including estimated takeoff time, landing time, and aircraft priority; S23: Integrate aircraft status and scheduling data to generate input data for path planning and scheduling optimization;

[0017] S3: Taxiway resource monitoring and data processing

[0018] S31: Use the taxiway resource real-time monitoring module to monitor the current taxiway occupancy in real time and obtain the taxiway resource usage status; S32: Use the taxiway congestion prediction result Y tCompare and analyze with real-time monitoring data to ensure that the prediction results are consistent with the actual situation or make timely corrections; S33: Process the monitoring data to generate real-time taxiway resource data for path planning, ensuring that the scheduling algorithm obtains the latest taxiway usage;

[0019] S4: Aircraft Path Planning and Scheduling Optimization

[0020] S41: Based on taxiway congestion prediction result Y t and aircraft scheduling data, using aircraft path planning and scheduling optimization algorithms to calculate the optimal aircraft taxiing path; S42: Generate the optimal taxiing path and scheduling plan Z through reinforcement learning algorithms t , ensuring that aircraft can take off or land on time and avoid congestion on the taxiway; S43: Generate the optimal taxiing path and scheduling plan Z based on the calculation results t , and output the optimal path and scheduling plan;

[0021] S5: Global evidence collection and analysis

[0022] S51: Obtain global evidence data of current airport surface operations through the global evidence data acquisition and processing module; S52: Analyze the global evidence data to assess the potential impact of external factors on surface operations; S53: Generate auxiliary data for global decision support to ensure that all influencing factors are considered in the global decision-making process;

[0023] S6: Global Operation Optimization Decision Generation

[0024] S61: Based on the airport's global intelligent decision-making and optimization algorithm, combined with the taxiway congestion prediction results Y t , optimal taxiing path and scheduling plan Z t and global evidence E, calculate the global optimization decision of airport surface operations; S62: reallocate taxiway resources and adjust aircraft scheduling priorities based on the global operation optimization decision; S63: generate the final optimized operation plan, output global adjustment suggestions and feed them back to the airport operation control system for execution.

[0025] Preferably, the airport taxiway dynamic congestion prediction algorithm is used to predict the taxiway congestion within a period of time in the future, and its specific formula is as follows:

[0026] where Y t represents the prediction result of taxiway congestion at time t, that is, the taxiway congestion in a certain time period in the future; c: is a constant term, representing the intercept in the model; φ i : is the autoregressive coefficient, which is used to express the impact of taxiway congestion in the past on the current moment; θ i: is the moving average coefficient, which is used to represent the impact of the prediction error at the past moment on the current moment; ε t : is the prediction error at time t.

[0027] Preferably, the aircraft path planning and scheduling optimization algorithm includes:

[0028] Q value update formula: Where Q(s t ,a t ,Y t ):Indicates that in state s t Next, select action a t The expected return value when α: represents the learning rate, which controls the update speed of the Q value; r t : is the reward value, reflecting the quality of the path selection; γ: is the discount factor, indicating the degree of influence of future rewards; :Indicates that in the next state S t+1 The maximum Q value obtained in

[0029] Preferably, the aircraft path planning and scheduling optimization algorithm further includes an optimal path generation formula: Where: Z t : represents the optimal taxiing path and scheduling plan generated at time t; In state s t and congestion prediction Y t Next, select the action a that maximizes the Q value t .

[0030] Preferably, the steps of the aircraft path planning and scheduling optimization algorithm include the following steps:

[0031] S1. Status update: Get the status s of the aircraft at the current time t t , including the physical location of the aircraft and the congestion prediction Y of the taxiway t and queue status; S2, action selection: through reinforcement learning algorithm, according to the state s t and taxiway congestion prediction Y t , select the current optimal action a t , that is, the taxiing path that the aircraft should take; S3, reward calculation: when taking action a t Then, according to the actual congestion of the taxiway and the taxiing time, the corresponding reward r is calculated. t , used to feedback the effect of aircraft taxiing path selection; S4, Q value update: based on reward r t and the state S at the next time t+1 t+1 , update the Q value to continuously optimize the path selection; S5, optimal path generation: through reinforcement learning, the optimal sliding path and scheduling plan Z is finally obtainedt .

[0032] Preferably, the airport global intelligent decision and optimization algorithm is based on the Bayesian decision network, combined with the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t And global evidence E, generate the optimal decision for global operation, the specific formula is as follows:

[0033] Where: P(X|Y t ,Z t ,E): Given the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t , the probability of the airport's global operating state X occurring under the condition of global evidence E, that is, the probability of the global optimization decision; P(Y t ,Z t ,E|X): Taxiway congestion prediction result Y under a specific global decision X t , optimal taxiing path and scheduling plan Z t and the joint probability of other global evidence E; P(X): the prior probability of the global running state X; P(Y t ,Z t ,E): Joint probability of taxiway congestion prediction, path optimization results and global evidence, as a normalization constant. ,

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. The present invention implements a taxiway congestion prediction algorithm based on time series analysis to achieve dynamic prediction of future taxiway usage, effectively avoiding the problems of unreasonable taxiway resource allocation and long aircraft waiting times.

[0036] 2. By adopting a reinforcement learning path optimization and scheduling algorithm, the present invention can dynamically adjust the aircraft's taxiing path and scheduling sequence based on real-time data, minimizing the aircraft's stay time on the taxiway and improving overall operational efficiency.

[0037] 3. The introduction of a global operation optimization decision-making algorithm combines taxiway congestion predictions, path optimization results, and other global evidence for comprehensive analysis to generate the globally optimal decision, ensuring the efficiency and safety of airport surface operations and reducing decision-making delays caused by manual intervention.

[0038] 4. The present invention can monitor the airport's operation status in real time through the setting of real-time monitoring and feedback system, timely discover potential problems and provide feedback on optimization suggestions, thus improving the intelligence and automation level of the airport's overall scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram showing the modular composition of the airport surface operation optimization system based on intelligent technology of the present invention;

[0040] Figure 2 This is a flow chart of the method for optimizing airport surface operation efficiency based on intelligent technology according to the present invention;

[0041] Figure 3 This is a structural diagram of the airport taxiway congestion prediction formula of the present invention;

[0042] Figure 4 A schematic diagram of Q-value update for aircraft path planning and scheduling optimization according to the present invention;

[0043] Figure 5 Generate a flow chart for the optimal path and scheduling of the present invention;

[0044] Figure 6 This is a flowchart of the steps of the aircraft path planning and scheduling optimization algorithm of the present invention. DETAILED DESCRIPTION

[0045] The following will provide a clear and complete description of 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0046] The present invention provides an airport surface operation efficiency optimization system based on intelligent technology. The system consists of multiple functional modules and can collect data in real time, predict taxiway congestion, optimize aircraft taxiing paths and scheduling, and make global intelligent decisions, thereby improving the overall operation efficiency of the airport.

[0047] 1. Taxiway data collection and prediction module: This module is responsible for collecting taxiway usage data and historical data in real time, including taxiway congestion status, aircraft taxiing speed, and taxiway occupancy time. By analyzing the historical taxiway usage and current aircraft taxiing speed, the system uses a time series prediction algorithm to predict future taxiway congestion. The generated congestion prediction result Y tThis module provides a basis for subsequent aircraft routing and scheduling, ensuring the system can proactively identify potential congestion points and optimize resource allocation. This module first uses sensors and monitoring equipment to obtain real-time taxiway occupancy information, and then predicts the future state of the taxiways based on historical data and current taxiing speeds. Through big data analysis and time series models, the system can anticipate potential congestion and take proactive measures to avoid unnecessary delays during taxiing.

[0048] 2. Aircraft status and scheduling data acquisition module: This module is mainly used to collect information on the physical status, queue status, and scheduling plans of aircraft. Through this module, the system can obtain the current position, taxiing status, and scheduled takeoff and landing plans of the aircraft in real time. The data of this module is used as input to the path planning and scheduling optimization algorithm to optimize the taxiing path and scheduling sequence of the aircraft. This module relies on the positioning systems installed throughout the airport and the aircraft's own navigation data to obtain the physical status and scheduling information of the aircraft in real time. This data will be integrated and passed to the system's path optimization module to ensure that the aircraft can taxi along the optimal path and minimize waiting time.

[0049] 3. Taxiway resource real-time monitoring module: The taxiway resource real-time monitoring module is used to monitor the current occupancy of the taxiway in real time to ensure that the system has the latest taxiway usage status. This module also compares the real-time data of the taxiway with the predicted data Y t Comparisons are performed to confirm the accuracy of the predictions, and the path planning is dynamically adjusted based on the latest monitoring data. By monitoring actual taxiway usage, the system can promptly detect discrepancies with the predicted data. If the actual taxiway occupancy changes, the system will dynamically adjust the previous path planning and provide the aircraft with an updated taxi path, ensuring smooth taxiing.

[0050] 4. Path planning and optimization result generation module: This module generates the optimal aircraft taxiing path and scheduling sequence through the "aircraft path planning and scheduling optimization algorithm". The system will calculate the taxiway congestion prediction result Y t Combined with the aircraft scheduling data, the aircraft taxiing path is continuously optimized through the reinforcement learning algorithm to generate the optimal scheduling plan Z t This module uses a reinforcement learning algorithm to dynamically adjust taxi paths based on current taxiway congestion predictions and aircraft status. Each time a path is selected, the system calculates the congestion and time consumption of different routes and selects the path that minimizes taxi time. The system also updates the scheduling plan in real time to ensure that all aircraft's taxi paths do not conflict with each other.

[0051] 5. Global Evidence Data Collection and Processing Module: This module is used to collect other data related to the global operation of the airport, such as weather conditions, aircraft fuel status, and the overall use of taxiway and runway resources. These data will serve as input for global intelligent decision-making and optimization algorithms, helping the system to make more accurate decision-making analysis. Through sensors and external data sources, this module collects external factors related to airport operations (such as weather, fuel conditions, etc.). These data will affect the aircraft's taxi path and scheduling priority. The system will comprehensively consider these global factors to make more accurate scheduling decisions.

[0052] 6. Global decision-making and feedback module: This module is based on global intelligent decision-making and optimization algorithm, combined with taxiway congestion prediction results Y t , path optimization result Z t and global evidence E to globally optimize airport surface operations. The global decision-making plan generated by the system can adjust the allocation of taxiway resources, the scheduling order and priority of aircraft, thereby ensuring the overall efficiency and safety of airport surface operations. The global decision-making and feedback module comprehensively analyzes the predicted congestion of taxiways, aircraft scheduling paths, and other external factors through a Bayesian decision network. The system generates the optimal decision based on the current global operating status and feeds it back to each module for execution to ensure that all parts of the system operate in coordination and maximize the overall operational efficiency of the airport.

[0053] The present invention provides an airport surface operation efficiency optimization method based on intelligent technology. The method mainly realizes intelligent management and optimization of airport surface operation through the following steps:

[0054] S1: Taxiway data collection and congestion prediction

[0055] S11: The taxiway data collection and prediction module collects real-time taxiway usage data, including taxiway occupancy status, aircraft taxiing speed, and historical usage data. This data provides the basis for subsequent congestion prediction.

[0056] S12: Preprocess the collected data, including cleaning, denoising, and formatting, to ensure data accuracy. Data preprocessing can filter out invalid data, ensure the quality of input data, and thus improve the accuracy of prediction results.

[0057] S13: Use the airport taxiway congestion prediction algorithm, combined with real-time data and historical data, to dynamically predict the future use of the taxiway and generate the taxiway congestion prediction result Y t Through time series analysis or machine learning models, possible congestion can be identified in advance, providing early warning and optimization solutions for aircraft scheduling.

[0058] S2: Aircraft status and dispatch data collection

[0059] S21: The Aircraft Status and Scheduling Data Collection Module provides real-time access to aircraft physical status information and scheduling plans, including the aircraft's current location, queue status, and takeoff and landing schedules. This module ensures that the aircraft's real-time status is readily fed back to the system, providing accurate data for route planning.

[0060] S22: Collects aircraft dispatch priority information and adjusts aircraft priorities based on their urgency (e.g., fuel status or flight delays). This information serves as input to path planning and scheduling optimization algorithms, ensuring on-time takeoff and landing, and avoiding unnecessary delays.

[0061] S23: Integrate the physical status of the aircraft and the scheduling priority information to provide input data for path planning, ensuring the rationality of scheduling and optimization of the path.

[0062] S3: Taxiway resource monitoring and data processing

[0063] S31: The Taxiway Resource Monitoring Module monitors taxiway usage in real time, ensuring the system is always aware of taxiway resource allocation. This module allows the system to proactively adjust routes and scheduling when taxiway resources are limited, avoiding conflicts between aircraft.

[0064] S32: Taxiway congestion prediction result Y t Compare and correct the forecast data with real-time monitoring data to ensure that the forecast data is consistent with the actual situation. Through comparative analysis, the system can adjust the forecast results in a timely manner and perform corresponding optimization operations based on the actual situation.

[0065] S33: Process the monitoring data and generate the latest taxiway resource data for path planning. This ensures that the system can obtain the latest taxiway usage status during each path planning, preventing the path planning plan from becoming invalid due to changes in resource status.

[0066] S4: Aircraft Path Planning and Scheduling Optimization

[0067] S41: Based on taxiway congestion prediction result Y t The system uses aircraft routing and scheduling optimization algorithms to calculate the optimal taxi path for each aircraft, taking into account taxiway congestion and aircraft scheduling priorities. This generates an optimal taxi path for each aircraft, ensuring efficient taxiing and minimizing taxiway wait times.

[0068] S42: The system generates the optimal taxiing path and scheduling plan Z based on the feedback of real-time data through reinforcement learning algorithm t Through continuous learning and optimization, the system can continuously update the taxi path according to the dynamically changing airport surface conditions, ensuring that the path is always in the optimal state.

[0069] S43: Generate the optimal taxiing path and scheduling plan Z based on the calculation results t , and outputs optimized routes and scheduling plans. The system ensures that aircraft can take off and land on time, avoiding taxiway congestion.

[0070] S5: Global evidence collection and analysis

[0071] S51: Through the Global Evidence Data Collection and Processing module, the system can collect other evidence data related to the airport's global operations, including weather conditions, fuel status, taxiway resource status, etc. This module ensures that the system can consider the potential impact of the external environment on aircraft scheduling when performing path planning and scheduling optimization.

[0072] S52: Analyzes global evidence data, assesses the impact of external factors on scene operations, and generates global decision-making support data. The system integrates this data into path planning and scheduling optimization algorithms to ensure the accuracy and operability of optimization results.

[0073] S6: Global Operation Optimization Decision Generation

[0074] S61: Based on the airport's global intelligent decision-making and optimization algorithm, combined with the taxiway congestion prediction results Y t , optimal taxiing path and scheduling plan Z t The system performs a global decision analysis based on the data and other global evidence E, generating an optimized decision plan for global operations. The system considers multiple external factors when optimizing taxiway resource allocation and scheduling priorities to ensure comprehensive decision-making.

[0075] S62: Based on the global decision results, taxiway resources are reallocated and the priority of aircraft scheduling is adjusted to ensure that each aircraft can taxi and take off and land under the best conditions.

[0076] S63: Generate the final optimized operation plan, output global scheduling suggestions, and feed them back to the airport operation control system for execution to ensure the global coordination and optimization of airport surface operations.

[0077] The airport taxiway dynamic congestion prediction algorithm proposed in this paper is primarily used to predict taxiway congestion over a period of time. By analyzing historical and real-time data, the algorithm can proactively identify potential congestion points, providing important decision-making information for subsequent route planning and scheduling optimization. Its prediction model, based on time series analysis, combines the characteristics of the autoregressive (AR) and moving average (MA) models, effectively capturing temporal trends in taxiway usage.

[0078] Congestion prediction formula

[0079] The prediction formula is as follows: In this formula: Y t It represents the prediction result of taxiway congestion at time t, that is, the taxiway congestion in a certain time period in the future;

[0080] c is a constant term, representing the intercept in the model, which usually reflects the baseline level of taxiway resource utilization, for example, the average taxiway occupancy rate or the baseline aircraft taxiing time when other factors are not considered.

[0081] φ i : is the autoregressive coefficient, which is used to express the impact of taxiway congestion in the past on the current moment; coefficient φ i The larger the value, the greater the impact of past congestion on the current situation. The coefficient is obtained through the autoregressive model (AR model) in time series analysis, which fits the time dependence of taxiway congestion based on historical data.

[0082] θ i : This is the moving average coefficient, which represents the impact of past forecast errors on the current timeframe. The moving average model smooths past forecast errors to improve model accuracy. This coefficient is calculated using a moving average model (MA model) in a time series, addressing the issue of random errors in forecasts.

[0083] ε t : is the prediction error at time t, that is, the prediction deviation caused by unobservable factors, such as sudden emergencies, weather changes, and other data that are not in the model.

[0084] Data Collection: The system collects data such as current taxiing speed and occupancy time from real-time taxiway monitoring equipment and combines it with historical data to provide input for the congestion prediction algorithm. The integrity and accuracy of the data directly impact the prediction results, so the system cleans and removes noise during the data collection process.

[0085] Prediction Model Construction: The algorithm uses autoregressive and moving average models, combining historical congestion data and forecast errors to construct a time series model. The autoregressive component (AR) captures the temporal dependence of taxiway congestion, while the moving average component (MA) processes past forecast errors to improve overall forecast accuracy.

[0086] Congestion prediction result generation: After the model is established, the system calculates the taxiway congestion in the future in real time by inputting current and historical data. t It provides the system with early warning of airport taxiway resource allocation and provides key data support for subsequent path planning and scheduling optimization.

[0087] Feedback and correction: The system will give the prediction result Y t Real-time comparisons are performed to verify the consistency of the predictions with actual taxiway occupancy. If discrepancies are found between the predictions and actual conditions, the system automatically adjusts and amends the model based on the new data to ensure the reliability of the predictions.

[0088] The aircraft path planning and scheduling optimization algorithm of the present invention is based on the Q-Learning algorithm of reinforcement learning. The algorithm can generate the optimal taxiing path and scheduling plan for the aircraft at each moment based on the current state of the aircraft and the taxiway congestion prediction results.

[0089] Q value update formula: Where Q(s t ,a t ,Y t ):Indicates that in state s t Next, select action a t The expected reward value when the aircraft is flying is 0; the Q value is updated based on the reinforcement learning algorithm, which optimizes the aircraft's taxiing path by repeatedly adjusting the path selection.

[0090] α: The learning rate controls how quickly the Q value is updated. A larger learning rate allows the system to adapt quickly to new situations. Its value ranges from [0, 1]. A higher learning rate indicates a system that is more sensitive to new information, while a lower learning rate indicates a more conservative system. The learning rate is typically adjusted to an appropriate value through experimentation.

[0091] r t : A reward value that reflects the quality of the route selection. It is usually calculated based on factors such as taxiway congestion and aircraft taxi time. A positive reward value indicates a good route selection, while a negative reward value indicates a poor route selection (e.g., congestion).

[0092] γ: A discount factor that indicates the impact of future rewards; typically between [0, 1]. A larger discount factor indicates that the system prioritizes long-term gains, while a smaller discount factor indicates that the system prioritizes current gains.

[0093] Indicates that in the next state S t+1 Based on the maximum Q value obtained, the system evaluates the possible options at the next moment and selects the path and action that will bring the greatest reward. By continuously updating the Q value, the system gradually learns and optimizes the aircraft's taxi path, allowing the aircraft to taxi efficiently and take off and land on time even in congested conditions.

[0094] The system calculates the optimal taxiing path and scheduling plan Z through reinforcement learning t , the final path selection is based on the action selection that maximizes the Q value, that is:

[0095] Optimal path generation formula: in:

[0096] Z t : represents the optimal taxiing path and scheduling plan generated at time t;

[0097] In state s t and congestion prediction Y t Next, select the action a that maximizes the Q value t ,The system determines the optimal gliding path and scheduling strategy through repeated learning.

[0098] The execution steps of the aircraft path planning and scheduling optimization algorithm are:

[0099] S1. Status update: The system first obtains the status s of the aircraft at the current time t t , including the physical location of the aircraft and the congestion prediction Y of the taxiway t and queuing status. By collecting these data, the system provides basic information for subsequent path planning.

[0100] S2, action selection: through reinforcement learning algorithm, according to the state s t and taxiway congestion prediction Y t , select the current optimal action a t , which is the taxi path the aircraft should take; the system evaluates the Q value of each feasible path and selects the path that can bring the greatest reward.

[0101] S3, reward calculation: after taking action a t Then, according to the actual congestion of the taxiway and the taxiing time, the corresponding reward r is calculated. t, used to feedback the effect of aircraft taxiing path selection; the higher the reward value, the more optimized the current path selection.

[0102] S4, Q value update: based on reward r t and the state S at the next time t+1 t+1 , update the Q value to continuously optimize the path selection;

[0103] S5. Optimal path generation: Through reinforcement learning, the optimal taxiing path and scheduling plan Z are finally obtained. t , ensuring that aircraft can taxi efficiently and take off and land on time even in congested conditions.

[0104] The airport global intelligent decision-making and optimization algorithm is based on the Bayesian decision network, combined with the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t And global evidence E, generate the optimal decision for global operation, the specific formula is as follows:

[0105] Where: P(X|Y t ,Z t ,E): Given the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t , the probability of the airport's global operating state X occurring under the condition of global evidence E, that is, the probability of the global optimization decision; P(Y t ,Z t ,E|X): Taxiway congestion prediction result Y under a specific global decision X t , optimal taxiing path and scheduling plan Z t The joint probability of the global evidence E and other global evidence E; this value indicates the degree of match between the predicted taxiway condition and the actual situation when the system makes a decision. P(X): The prior probability of the global operating state X; this value is usually set based on historical data and experience. t ,Z t ,E): The joint probability of taxiway congestion prediction, path optimization results and global evidence, is used as a normalization constant to ensure the validity of the probability distribution of the Bayesian formula.

[0106] Data input and processing: The system first collects the taxiway congestion prediction results Y t , optimal taxiing path and scheduling plan Z t and global evidence E, these data will serve as input for global decision analysis, helping the system to comprehensively evaluate the operation status of the airport surface.

[0107] Global state assessment: Using a Bayesian network, the system can calculate the probability of the airport's global operating state X based on the current input data. This state X includes a comprehensive assessment of factors such as taxiway resource allocation, aircraft scheduling priority, and taxi time.

[0108] Optimal decision generation: P(X|Y t ,Z t ,E), the system can generate the optimal global operational decision X. This means that the system not only considers the current taxiway conditions and aircraft scheduling, but also comprehensively adjusts the allocation and scheduling priorities of airport resources based on external factors (such as weather) to ensure the efficiency and safety of overall operations. Feedback and Optimization: The system continuously monitors taxiway usage and aircraft scheduling results during actual operations and feeds this data back into the Bayesian decision network to update the model parameters. If the predicted results do not match the actual results, the system automatically corrects the decision strategy to ensure more accurate future optimization results.

[0109] Example 1: Taxiway resource optimization and aircraft scheduling during airport peak hours

[0110] Peak hours at airports are critical for airport surface operations, often leading to taxiway resource shortages and extended aircraft wait times. Traditional airport taxiway scheduling methods are unable to dynamically adjust taxi paths based on real-time data, resulting in irrational taxiway resource allocation, frequent congestion, and inefficient aircraft scheduling. To address this, the present invention provides an intelligent airport surface operation efficiency optimization system that utilizes taxiway congestion prediction and a global intelligent scheduling algorithm to alleviate these issues.

[0111] The technical purpose of this embodiment is to improve the efficiency of taxiway resource utilization, optimize aircraft scheduling strategies, reduce congestion incidence, and shorten aircraft taxiing time during airport peak hours by applying the system of the present invention, thereby improving the overall operating efficiency of the airport.

[0112] This invention utilizes a time-series-based taxiway congestion prediction algorithm and a reinforcement learning-based path planning algorithm. By combining real-time and historical data, it dynamically adjusts aircraft taxi paths and scheduling priorities. Compared to existing static scheduling methods, this invention can automatically adjust taxiing plans based on congestion conditions, significantly improving the system's intelligence.

[0113] This embodiment is particularly suitable for optimizing taxiway resources during peak hours at airports, when aircraft take off and land frequently. During peak hours, airports experience a surge in the number of aircraft. This invention can rapidly adjust scheduling plans in this complex environment, minimizing aircraft waiting time on taxiways.

[0114] To verify the effectiveness of this invention during peak hours at airports, we designed several comparative experiments. These experiments covered both peak and normal periods, and compared the proposed method with a traditional static scheduling algorithm (the prior art). Comparison criteria included taxi time, aircraft waiting time, taxiway congestion, number of scheduling adjustments, scheduling response time, aircraft takeoff and landing on-time performance, taxiway resource utilization, and system resource utilization.

[0115]

[0116] Experimental results show that during peak hours at airports, the system of the present invention significantly reduces aircraft taxiing and waiting times, improving dispatch response speed and taxiway resource utilization. Compared with traditional dispatch systems, the present invention demonstrates significant advantages in multiple indicators, including congestion and punctuality.

[0117] Example 2: Airport Global Scheduling Optimization under Severe Weather Conditions

[0118] Inclement weather conditions severely impact airport taxiway and runway availability, further complicating aircraft scheduling. Traditional scheduling systems often fail to quickly respond to changes in the external environment, resulting in prolonged taxiing times and frequent takeoff and landing delays. To address this issue, the present invention provides a global intelligent scheduling optimization system based on a Bayesian decision network to address scheduling challenges in inclement weather.

[0119] The technical purpose of this embodiment is to introduce a global intelligent scheduling system, combined with real-time taxiway monitoring data, weather conditions and other external factors, to dynamically adjust taxiway resource allocation and aircraft scheduling strategies to reduce delays and improve airport operation efficiency.

[0120] Based on a Bayesian decision network, this system comprehensively evaluates taxiway congestion, aircraft status, and global evidence data (such as weather and runway conditions) in adverse weather conditions to generate globally optimized scheduling decisions. Compared to existing technologies, this system can respond to changes in the external environment in real time and automatically adjust taxiing plans, thereby maintaining efficient airport operations under adverse conditions.

[0121] This embodiment is suitable for scenarios where airport operations are significantly impacted by adverse weather conditions such as heavy rain or strong winds. Through the global intelligent scheduling system of the present invention, airports can dynamically adjust scheduling strategies based on changes in the external environment, minimizing delays and congestion caused by weather.

[0122] The experiment compared the performance of the system of the present invention with that of the existing dispatch system under adverse weather conditions. The evaluation indicators included taxiing time, number of delays, aircraft take-off and landing time deviation, dispatch response speed, number of dispatch adjustments, aircraft cancellation rate, passenger satisfaction, system scalability, etc.

[0123]

[0124]

[0125] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The airport surface operation efficiency optimization system based on intelligent technology is characterized by: The system includes the following modules: Taxiway data collection and prediction module: used to collect real-time taxiway usage and historical data, and use the airport taxiway dynamic congestion prediction algorithm to predict the future taxiway congestion status based on the aircraft taxiing speed and occupancy, and output the taxiway congestion prediction result Y t ; Aircraft status and scheduling data acquisition module: used to obtain information about the aircraft's physical status, queue status, takeoff and landing schedules, and provide data on the aircraft's current location, taxiing status, and scheduling priority as input to aircraft path planning and scheduling optimization algorithms; Taxiway resource real-time monitoring module: used to monitor taxiway occupancy and resource allocation status in real time, and compare the current taxiway usage with the taxiway congestion prediction result Y t Matching, providing real-time data support for path planning algorithms; Path planning and optimization result generation module: through aircraft path planning and scheduling optimization algorithm, combined with taxiway congestion prediction result Y t and aircraft scheduling data to generate the aircraft's optimal taxiing path and scheduling plan Z t , ensuring efficient aircraft taxiing and avoiding congestion; Global Evidence Data Collection and Processing Module: This module is used to collect other evidence data related to global operations, including weather conditions, fuel status, and taxiway resource status, providing support data for the airport's global intelligent decision-making and optimization algorithms. Global decision-making and feedback module: Based on the airport's global intelligent decision-making and optimization algorithm, using the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t and other global evidence E, formulate a global optimization decision plan for airport surface operations, including the reallocation of taxiway resources and adjustment of aircraft scheduling priorities.

2. An intelligent technology-based method for optimizing airport surface operation efficiency, characterized by: The following steps are involved: S1. Taxiway data collection and congestion prediction: S11: The taxiway data acquisition and prediction module obtains real-time taxiway usage, historical data, and aircraft taxiing speed information; S12: Preprocess the collected data, including data cleaning, denoising and formatting; S13: Use the airport taxiway dynamic congestion prediction algorithm to predict future taxiway congestion based on real-time and historical data, and generate the taxiway congestion prediction result Y t ; S2. Aircraft status and dispatch data collection: S21: Acquire physical status data of the aircraft from the aircraft status and scheduling data acquisition module; S22: Obtaining aircraft takeoff and landing plan information, including estimated takeoff time, landing time, and aircraft priority; S23: Integrate aircraft status and scheduling data to generate input data for path planning and scheduling optimization; S3: Taxiway resource monitoring and data processing S31: Using the taxiway resource real-time monitoring module, the current taxiway occupancy status is monitored in real time to obtain the taxiway resource usage status; S32: Taxiway congestion prediction result Y t Compare and analyze with real-time monitoring data to ensure that the forecast results are consistent with the actual situation or make timely corrections; S33: Processes monitoring data to generate real-time taxiway resource data for path planning, ensuring that the scheduling algorithm receives the latest taxiway usage information; S4: Aircraft Path Planning and Scheduling Optimization S41: Based on taxiway congestion prediction result Y t and aircraft scheduling data, using aircraft path planning and scheduling optimization algorithms to calculate the optimal aircraft taxi path; S42: Generate the optimal taxiing path and scheduling plan Z through reinforcement learning algorithm t , ensuring that aircraft can take off or land on time and avoid congestion on the taxiways; S43: Generate the optimal taxiing path and scheduling plan Z based on the calculation results t , and output the optimal path and scheduling plan; S5: Global evidence collection and analysis S51: Obtaining global evidence data of current airport surface operations through the global evidence data collection and processing module; S52: Analyze global evidence data and assess the potential impact of external factors on scene operations; S53: Generate auxiliary data for global decision support to ensure that all influencing factors are taken into account in the global decision process; S6: Global Operation Optimization Decision Generation S61: Based on the airport's global intelligent decision-making and optimization algorithm, combined with the taxiway congestion prediction results Y t , optimal taxiing path and scheduling plan Z t and global evidence E, the global optimization decision of the computer field operation; S62: Reallocate taxiway resources and adjust aircraft scheduling priorities based on global operational optimization decisions; S63: Generate the final optimized operation plan, output global adjustment suggestions and feed them back to the airport operation control system for execution.

3. The method for optimizing airport surface operation efficiency based on intelligent technology according to claim 2 is characterized by: The airport taxiway dynamic congestion prediction algorithm is used to predict the taxiway congestion in the future. The specific formula is as follows: where Y t It represents the prediction result of taxiway congestion at time t, that is, the taxiway congestion in a certain time period in the future; c: is a constant term, representing the intercept in the model; φ i : is the autoregressive coefficient, which is used to represent the impact of taxiway congestion in the past on the current moment; θ i : is the moving average coefficient, which is used to represent the impact of the prediction error at the past moment on the current moment; ε t : is the prediction error at time t.

4. The method for optimizing airport surface operation efficiency based on intelligent technology according to claim 2 is characterized by: The aircraft path planning and scheduling optimization algorithm includes: Q value update formula: Where Q(s t ,a t ,Y t ):Indicates that in state s t Next, select action a t The expected return value when α: represents the learning rate, which controls the update speed of the Q value; r t : is the reward value, reflecting the quality of path selection; γ: is the discount factor, which indicates the impact of future rewards; Indicates that in the next state S t+1 The maximum Q value obtained in 5. The method for improving airport surface operation efficiency based on intelligent technology according to claim 4, characterized in that: The aircraft path planning and scheduling optimization algorithm also includes the optimal path generation formula: in: Z t : represents the optimal taxiing path and scheduling plan generated at time t; In state s t and congestion prediction Y t Next, select the action a that maximizes the Q value t .

6. The intelligent technology-based airport surface operation efficiency optimization system and method according to claim 5 is characterized by: The steps of the aircraft path planning and scheduling optimization algorithm include the following steps: S1. Status update: Get the status s of the aircraft at the current time t t , including the physical location of the aircraft and the congestion prediction Y of the taxiway t and queue status S2, action selection: through reinforcement learning algorithm, according to the state s t and taxiway congestion prediction Y t , select the current optimal action a t , which is the glide path the aircraft should take; S3, reward calculation: after taking action a t Then, according to the actual congestion of the taxiway and the taxiing time, the corresponding reward r is calculated. t , used to provide feedback on the effectiveness of aircraft taxi path selection; S4, Q value update: based on reward r t and the state S at the next moment t+1 t+1 , update the Q value to continuously optimize the path selection; S5. Optimal path generation: Through reinforcement learning, the optimal taxiing path and scheduling plan Z are finally obtained. t .

7. The method for optimizing airport surface operation efficiency based on intelligent technology according to claim 2 is characterized by: The airport global intelligent decision-making and optimization algorithm is based on the Bayesian decision network, combined with the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t And global evidence E, generate the optimal decision for global operation, the specific formula is as follows: Where: P(X|Y t ,Z t ,E): Given the taxiway congestion prediction result Y t , optimal taxiing path and scheduling plan Z t , the probability of the airport's global operating state X occurring under the condition of global evidence E, that is, the probability of the global optimization decision; P(Y t ,Z t ,E|X): Taxiway congestion prediction result Y under a specific global decision X t , optimal taxiing path and scheduling plan Z t and the joint probability of other global evidence E; P(X): prior probability of the global operating state X; P(Y t ,Z t ,E): Joint probability of taxiway congestion prediction, path optimization results and global evidence, as a normalization constant.

Images