Civil aviation aircraft fuel consumption dynamic optimization system based on real-time route data

Abstract

This application relates to the field of civil aviation optimization technology, and in particular to a dynamic fuel consumption optimization system for civil aircraft based on real-time route data. The system includes modules for data storage, input, multi-source information fusion and feature extraction, intelligent evaluation, adaptive decision support, and output response. The system extracts multi-dimensional features by fusing real-time route and QAR data, constructs a correlation dynamic evaluation model using gradient boosting decision trees, simultaneously calculates flight performance index, fuel efficiency index, and correlation score, and generates a structured report based on a decision rule base, including cost index adjustment suggestions, altitude change schemes, and expected fuel savings predictions. This application enables dynamic and quantitative evaluation of the correlation between route environment and fuel consumption, overcomes the lag of traditional planning, and achieves precise and personalized flight strategy optimization, resulting in significant benefits in reducing costs, improving on-time performance, and reducing emissions.

Description

Technical Field

[0001] This invention belongs to the field of aviation transportation and data processing technology, and more specifically, relates to a dynamic optimization system for fuel consumption of civil aircraft based on real-time route data. Background Technology

[0002] Currently, in the operation and management of the civil aviation transportation industry, optimizing aircraft fuel efficiency is a key aspect of reducing operating costs and responding to energy conservation and emission reduction policies. However, traditional flight planning and real-time flight control models have many limitations in this regard: First, the assessment methods are outdated and isolated: Currently, the assessment and prediction of flight fuel consumption largely rely on static pre-flight planning data or route models based on historical data. Responses to real-time weather conditions (such as high-altitude winds and temperature), dynamic air traffic control instructions, and airspace congestion during flight are primarily based on manual observation by the crew or experience-based judgment by ground dispatchers. These methods not only suffer from significant timeliness and lag, but more importantly, they separate the highly correlated dynamic processes of real-time route environment and flight performance optimization, failing to reflect the immediate impact of external environmental changes on fuel consumption in a real-time and dynamic manner.

[0003] Secondly, the reliance on experience-based decision-making lacks precision: Adjustments to cruise altitude and speed strategies during flight are mostly based on fixed cost indices (CI) or standard operating procedures, or rely on pilots' individual flying experience for rough adjustments. This approach cannot accurately respond to real-time track deviations, real-time weather fluctuations, and uncertain waiting instructions, easily leading to either untimely adjustments to fuel consumption strategies, resulting in unnecessary fuel consumption, or over-adjustments, leading to lost flight time or reduced safety margins.

[0004] Third, the problem of data silos is serious: Civil aviation operations accumulate a large amount of multi-source heterogeneous data, including QAR data collected by airborne sensors, ACARS communication messages, real-time weather radar data, and air traffic control notices. However, this data is usually scattered across airborne systems, airline dispatch centers, or external meteorological services, forming independent "data silos." The lack of effective technical means to deeply integrate, analyze, and extract insights with real-time optimization value from this real-time, multi-modal route data results in the optimization potential of massive amounts of flight data being buried.

[0005] Fourth, causal relationships are difficult to quantify: Although civil aviation practitioners generally recognize that changes in the route environment directly affect fuel consumption, and vice versa, the weight distribution of this mutual influence, and which specific dynamic environmental factors are most closely related to fuel consumption fluctuations, are difficult to quantify precisely. Managers cannot answer precise questions such as, "Under specific crosswind speeds and airspace restrictions, how much fuel benefit will adjusting cruise thrust by 3% bring?" Fuel consumption optimization decisions lack refined data support.

[0006] Therefore, a dynamic optimization system for fuel consumption of civil aircraft based on real-time route data is proposed. Summary of the Invention

[0007] The purpose of this invention is to provide a dynamic optimization system for fuel consumption of civil aircraft based on real-time route data, which can effectively solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A dynamic fuel consumption optimization system for civil aircraft based on real-time route data includes: The data storage module is used to store historical and real-time flight multimodal data; The input module, connected to the data storage module, is used to receive real-time route multimodal data from an external data source; The multi-source information fusion and feature extraction module is connected to the input module and the data storage module respectively. It is used to standardize and normalize the received multimodal data, and extract multidimensional feature vectors for evaluating the correlation from the multimodal data based on a preset feature system for quantifying the dynamic correlation between the route environment and fuel consumption. The route multimodal data includes high-altitude wind field data, ambient temperature deviation data, real-time flight path coordinate data, airspace congestion level data, engine performance monitoring data, and QAR stream data collected by airborne sensors. The feature system includes real-time meteorological impact features, flight path deviation features, air traffic control command constraint features, and flight performance features; the intelligent evaluation core module is connected to the multi-source information fusion and feature extraction module, and contains a correlation dynamic evaluation model. The correlation dynamic evaluation model is an ensemble learning model built on the gradient boosting decision tree algorithm. It is obtained by training historical multidimensional feature vectors and labeled data verified by QAR data through the ensemble learning algorithm. It is used to synchronously calculate and output flight performance index, fuel efficiency index and correlation score representing the correlation strength between the two based on the input real-time multidimensional feature vector. The correlation dynamic evaluation model uses SHAP interpretive technology to quantify the contribution of each input's real-time multidimensional feature vector to the flight performance index and fuel efficiency index, and reveals the correlation strength between different environmental factors and different flight status indicators in the form of a heat map. The adaptive decision support module, connected to the intelligent evaluation core module, is used to generate a structured report containing personalized cost index adjustment suggestions, altitude change schemes, and expected fuel savings predictions based on correlation scores, flight performance indices, fuel efficiency indices, and a preset decision rule base. The preset decision rule base combines production rules with case-based reasoning, and the rules in the decision rule base take the following form: IF correlation score AND fuel efficiency index AND Flight Performance Index THEN Execution: Adjust the cruise cost index to It is recommended to request an ascent to a higher level. It is expected to improve fuel efficiency. Expected flight time changes Among them, the threshold Range A, Range B, Parameter The specific value is dynamically determined based on statistical analysis of historical datasets and the results of case reasoning; The output response module, connected to the adaptive decision support module, is used to visualize the structured report and core evaluation indicators through a graphical user interface. The system also includes a model continuous learning module, connected to the intelligent evaluation core module and the data storage module, respectively, for periodically using newly added flight multimodal data and corresponding validation label data to incrementally train and optimize the correlation dynamic evaluation model in order to update the model parameters.

[0009] Preferably, the input module includes: a standardized data interface unit for receiving the route multimodal data from the airborne communication addressing and reporting system, real-time weather radar, ground dispatch system and air traffic control automation terminal; and an edge computing unit deployed on the airborne processing terminal for performing preliminary cleaning, redundancy removal and formatting processing on the initial data.

[0010] Preferably, the visualization interface generated by the output response module includes a correlation analysis dashboard, which displays the correlation strength between different meteorological factors and different fuel consumption indicators revealed by the correlation dynamic evaluation model in the form of a heat map, and displays the historical changes of core key performance indicators and the predicted trajectory based on current decision recommendations in the form of trend lines.

[0011] Preferably, the system further includes an early warning module, which is connected to the intelligent assessment core module and the adaptive decision support module respectively; the early warning module is configured to automatically trigger multi-level early warning signals when the correlation score is lower than a preset threshold, and push the early warning information and corresponding structured decision suggestions to the user terminal of the designated crew or dispatcher.

[0012] Preferably, when processing multimodal data, the multi-source information fusion and feature extraction module performs Z-score standardization on features of different dimensions and calculates the crosswind drag influence factor based on the real-time crosswind component and Mach number deviation.

[0013] Preferably, the formula for calculating the marginal contribution value of each feature to the output result by the correlation dynamic evaluation model is as follows:

[0014] in, For the first The contribution value of each feature, This is the baseline value output by the model.

[0015] Preferably, the trigger condition for the model continuous learning module to perform incremental training is: the root mean square error between the fuel flow predicted by the model and the actual QAR flow exceeds a preset threshold. .

[0016] Preferably, when performing case reasoning, the adaptive decision support module uses the K-nearest neighbor algorithm to retrieve the historical best fuel-saving schemes under similar routes and weather conditions from the historical flight case database, and adjusts the parameters accordingly. Perform dynamic corrections.

[0017] Preferably, the data storage module adopts a Hadoop-based distributed storage architecture and uses a time-series database to store airborne QAR stream data and high-frequency meteorological monitoring data.

[0018] Preferably, the edge computing unit is configured to perform window smoothing filtering before data upload, and only upload data when the data variation rate exceeds a preset threshold. The upload operation is triggered at that time.

[0019] Compared with the prior art, the present invention has the following beneficial effects: This civil aviation aircraft fuel consumption dynamic optimization system based on real-time route data has achieved dynamic and quantitative assessment and precise management of the correlation between route environment and aircraft fuel consumption by constructing a system that integrates data fusion, intelligent evaluation and decision support. It effectively overcomes the lag, isolation and experience-based defects of traditional flight planning mode, thus generating significant comprehensive benefits in reducing operating costs, improving on-time performance and reducing carbon emissions.

[0020] This civil aircraft fuel consumption dynamic optimization system based on real-time route data systematically accesses, cleans, standardizes, and integrates multimodal data from airborne systems, meteorological services, air traffic control instructions, and historical QAR databases through data storage, input, and multi-source information fusion and feature extraction modules. This breaks down data silos and extracts high-value feature vectors for evaluating correlations based on a specially designed feature system, laying a solid data foundation for precise fuel consumption optimization.

[0021] This civil aviation aircraft fuel consumption dynamic optimization system, based on real-time route data, uses a correlation dynamic evaluation model trained within the intelligent evaluation core module to simultaneously and rapidly calculate flight performance index, fuel efficiency index, and, most importantly, a correlation score representing the strength of the correlation between the two. This changes the previous reliance on pilots' subjective experience and static planned values, achieving a technological leap from "manual estimation" to "data intelligence."

[0022] This civil aviation aircraft fuel consumption dynamic optimization system, based on real-time route data, utilizes an adaptive decision support module and its pre-set decision rule library to automatically generate structured reports containing specific cost index adjustments, altitude change instructions, and expected fuel-saving predictions based on real-time assessed indices and scores. This transforms fuel consumption optimization from "extensive control" to "refined decision-making," significantly improving the economy and responsiveness of civil aviation operations. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the structural principle of the civil aircraft fuel consumption dynamic optimization system of the present invention. Detailed Implementation

[0024] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Example 1: Currently, in the operation and management of the civil aviation transportation industry, the optimization of aircraft fuel efficiency is limited by pain points such as outdated and isolated assessment methods, reliance on experience-based decision-making, severe data silos, and difficulty in quantifying causal relationships. Traditional static flight plans cannot provide real-time insight into the dynamic relationship between changes in the route environment and fuel consumption, resulting in fuel consumption control strategies either being adjusted untimely or being over-adjusted, thus affecting flight efficiency.

[0027] In view of this, please refer to Figure 1 As shown, the purpose of this invention is to provide a dynamic fuel consumption optimization system for civil aircraft based on real-time route data. This system includes: The data storage module is used to store historical and real-time flight multimodal data. The data storage module can be a high-performance central server configured with a Hadoop-based distributed storage architecture. The MySQL database is used to store structured flight plans, historical waypoints and cost index settings, while HDFS (Hadoop Distributed File System) and InfluxDB time-series database are used to store massive amounts of airborne QAR streaming data, ACARS communication messages and high-frequency meteorological monitoring data. The input module, connected to the data storage module, is used to receive real-time multimodal route data from external data sources. The input module connects to multiple data sources via a dedicated aviation network and airborne communication links, as detailed below: The RESTful API interface is used to receive JSON-formatted weather reports (such as SIGMET and WINTEM) from ground dispatch systems and external meteorological service agencies. The ARINC429 / 618 protocol parsing interface is used to capture flight status and real-time track coordinates issued by the Airborne Communication Addressing and Reporting System (ACARS) in real time; a dedicated data stream parser is used to process instruction messages (including speed limit, rerouting, and holding instructions) from the Air Traffic Control (ATC) automated terminal. Mobile / airborne EFB terminals serve as human-assisted input terminals, allowing crew members to provide feedback on the real-time turbulence levels or execution status of special flight commands perceived by the cockpit. The multi-source information fusion and feature extraction module is connected to both the input module and the data storage module. It is implemented by a data processing service deployed on a backend server, and its workflow is as follows: The received multimodal data is standardized and normalized (first, the QAR data, weather radar data and track data are denoised to remove outliers, and Z-score standardization is performed on features of different dimensions such as wind speed (kt), altitude (ft), and thrust (%)). Based on a pre-defined feature system for quantifying the dynamic correlation between airway environment and fuel consumption, multi-dimensional feature vectors for evaluating the correlation are extracted from multimodal data. The feature system includes real-time weather impact features, track deviation features, air traffic control command constraint features, and flight performance features. For example, the "crosswind drag impact factor" is calculated based on the "real-time crosswind component" and "Mach number deviation"; the "horizontal track deviation" is extracted based on the "actual flight coordinates" and "planned great circle track"; the "instantaneous performance degradation coefficient" is extracted from the "engine thrust lever angle" and "EGT (exhaust gas temperature)"; and the "airspace congestion response feature" is extracted from the "altitude layer change frequency". The intelligent assessment core module is connected to the multi-source information fusion and feature extraction module, and it contains a correlation dynamic assessment model. The correlation dynamic assessment model is trained by an ensemble learning algorithm (such as the improved XGBoost algorithm) on historical multidimensional feature vectors and labeled data verified by QAR data (through post-accurate calculation of "real fuel efficiency" and "engine performance status"). It is used to synchronously calculate and output the Flight Performance Index (FPI, 0-100), the Fuel Efficiency Index (FEI, 0-100), and the Correlation Score (CS, 0-1) representing the correlation strength between the two based on the input real-time multidimensional feature vectors. This score quantifies the sensitivity and coupling between real-time route environment fluctuations and final fuel consumption fluctuations. The adaptive decision support module, connected to the intelligent assessment core module, generates a structured report containing personalized cost index (CI) adjustment suggestions, altitude change plans, and expected fuel savings predictions based on correlation scores, flight performance indices, fuel efficiency indices, and a pre-set decision rule base (using the Drools rule engine, integrating dozens of flight operation production rules). The specific process for generating the structured report is as follows: The adaptive decision support module receives three key indices output by the intelligent evaluation core module and matches them against operational constraints in the rule base. For example, if a match is found between the "high correlation + low fuel efficiency" rule, strategy optimization is triggered, generating a JSON report containing: “ "、 “ "、 “ "、 “ ”; The output response module, connected to the adaptive decision support module, is a graphical user interface (GUI) developed based on the Vue.js framework. The interface includes a dynamic dashboard, fuel consumption trend analysis chart, and optimization decision cards, which are used to visualize structured reports and core evaluation indicators.

[0028] By constructing the aforementioned system, flight multimodal data, feature engineering, machine learning models, and aviation decision-making rules are integrated into an automated closed loop for the first time. This achieves end-to-end automation from route environment perception to strategy optimization recommendations, solving the problems of "data silos" and "decision lag" in civil aviation operations. The system quantifies the "correlation strength" between complex route environments and fuel consumption using models, and this score serves as the key logic for triggering dynamic optimization. This deep integration of civil aviation operation technology and data science has significant benefits in reducing operating costs, improving on-time performance, and reducing carbon emissions.

[0029] Example 2: Considering the different protocols and data formats of external data sources in civil aviation (such as dispatch systems of different airlines, weather radar, and airborne sensor systems), and the high-frequency noise (such as sensor pulses caused by landing gear vibration) in the raw QAR stream data, direct transmission to the backend would consume a significant amount of computing bandwidth. Furthermore, the complex network environment during flight makes uploading all data prone to degrading real-time performance. Therefore, the input module includes: A standardized data interface unit is used to receive the route multimodal data from the airborne communication addressing and reporting system, real-time weather radar, ground dispatch system, and air traffic control automation terminal; specifically: The standardized data interface unit, serving as an aviation data access gateway, is developed in Java and integrates ARINC618, ARINC429, MQTT, and RESTful protocol stacks. All heterogeneous data incoming from the outside is first converted into a unified internal data structure (such as the Apache Avro format) through schema mapping, ensuring strong consistency in data format. Edge computing units, deployed in airborne processing terminals (such as Electronic Flight Bags (EFBs) or general-purpose airborne processing chassis), are used to perform preliminary cleaning, redundancy removal, and formatting of initial data; specifically: A lightweight computing component (developed in Go) is deployed on the airborne terminal to perform window smoothing filtering before data upload, eliminating sensor sampling glitches. For example, for 1Hz fuel flow data, a moving average is processed using an edge algorithm, and data is only processed when the rate of change exceeds a preset threshold. Uploads are triggered only when needed, which greatly reduces the bandwidth pressure on the air-to-ground link and ensures the accurate and real-time upload of core features. Through the "cloud-edge" collaborative architecture, lightweight preprocessing of route data was achieved, ensuring the real-time performance and high fidelity of model input.

[0030] Example 3: Considering that traditional fuel-saving methods only focus on "static planned fuel consumption" and "remaining fuel consumption after landing," ignoring "dynamic meteorological environment (such as high-altitude temperature and pressure fields)" and "microscopic deviations in flight status (such as vertical speed fluctuations)," the assessment accuracy is low, making it impossible to determine whether the excessive fuel consumption is due to "crew operation" or "force majeure." Therefore, multimodal data for the flight route includes high-altitude wind field data, ambient temperature deviation data, real-time flight path coordinate data, airspace congestion level data, engine performance monitoring data, and QAR stream data collected by airborne sensors. Among these, the high-altitude wind field data includes three-dimensional wind speed and direction (…). Components); Ambient temperature deviation data ( This reflects the deviation between the actual atmospheric temperature and the standard atmospheric temperature; the QAR flow data collected by airborne sensors includes cruise Mach number (Mach), total weight (GW), total temperature (TAT), thrust lever angle (TLA), and instantaneous fuel flow (FF); engine performance monitoring data includes low-pressure speed (N1), high-pressure speed (N2), and exhaust gas temperature (EGT). By introducing ambient temperature deviation and thrust lever action as correlation indicators, the explanatory dimensions of fuel consumption fluctuations are greatly enriched. Its non-obviousness lies in the fact that it couples the exogenous variable of "dynamic temperature and pressure environment" with the endogenous variable of "engine micro-thrust adjustment" to construct a multi-dimensional fuel efficiency evaluation benchmark.

[0031] Example 4: Due to the extremely high safety requirements for decision-making in civil aviation operations, if machine learning models operate as a "black box," dispatchers will find it difficult to adopt their suggestions for adjusting CI (Critical Index). Furthermore, a single predicted value cannot reveal which route factors are the main causes of fuel consumption. Therefore, the correlation dynamic evaluation model is an ensemble learning model built based on the Gradient Boosting Decision Tree (XGBoost) algorithm. The real-time multi-dimensional feature vector input includes: cruise Mach number, headwind component, total temperature deviation, vertical deviation, thrust lever angle, and instantaneous fuel flow. The model output includes the predicted flight performance index and overall fuel efficiency; specifically: First, the correlation dynamic evaluation model is trained using Python's XGBoost library for regression, and the objective function is optimized to minimize the fuel consumption residual. Secondly, the input feature vector includes real-time weather data (such as headwind component). ) and operating parameters (such as cruise Mach number) It has more than 30 dimensions of features; Then, the model adopts a dual-output structure. One end outputs the Flight Performance Index (FPI) to quantify whether the aircraft is in the optimal aerodynamic state; the other end outputs the Total Fuel Efficiency (TEU) to quantify the fuel consumption efficiency per unit distance. Finally, the model integrates the SHAP interpretation tool to calculate the marginal contribution of each feature to the output. The calculation formula is as follows:

[0032] in, For the first The system displays the contribution values ​​of various meteorological factors and performance indicators in the form of a heat map. The color intensity of the heat map reveals the correlation strength between different wind directions, temperature and pressure conditions, and fuel consumption fluctuations. By utilizing the SHAP interpretability technology, this model enables the abstract "relevance score" to have physical traceability, thereby enhancing the unit's trust in the optimization recommendations.

[0033] Example 5, considering that aircraft performance decreases with increasing flight hours (performance degradation rate) Furthermore, different routes and aircraft types have different operating characteristics in different seasons, causing "accuracy drift" in models with fixed parameters; therefore, the system also includes: The model continuous learning module, connected to both the intelligent evaluation core module and the data storage module, is used to periodically perform incremental training and optimization of the correlation dynamic evaluation model using newly added flight multimodal data and corresponding validation label data, in order to update the model parameters. The specific scheme is as follows: The system backend features an automated retraining pipeline (CI / CD). The model's continuous learning module automatically captures all QAR records and fuel consumption calculation data from the past 30 days each month, using the newly added data as a validation set. If the root mean square error (RMSE) between the model's predicted fuel flow and the actual QAR flow exceeds a preset threshold... If the speed reaches kg / s, it will automatically trigger model fine-tuning based on incremental learning, dynamically updating and improving the weight parameters of the boost tree; This module enables the system to adapt to aircraft performance degradation and seasonal weather changes, ensuring the long-term robustness of the assessment results.

[0034] Example 6: Considering that traditional fuel consumption management relies heavily on pilots consulting Quick Checklists (QRH) or referring to static cost index tables, it cannot cope with the complex coupled scenario of "high-altitude wind shear + air traffic control speed limits + heavy weight"; therefore, the decision rule base preset in the adaptive decision support module is a combination of production rules and case reasoning, and the rules in the decision rule base are in the following form: IF correlation score AND fuel efficiency index AND Flight Performance Index For example: IF correlation score AND fuel efficiency index AND Flight Performance Index ; THEN Execution: Adjust the cruise cost index ( )to It was suggested that a climb to a higher altitude of $Y$ft be requested, which would be expected to improve fuel efficiency. Expected flight time changes min; for example: THEN implementation: It is recommended to reduce the cost index to min. And request to climb to a higher level. (37,000 ft), expected fuel efficiency improvement The expected flight time will increase. min.

[0035] The system backend maintains a global flight path case database. When a decision is triggered, the system uses the K-Nearest Neighbor (KNN) algorithm to retrieve successful fuel-saving solutions under similar historical flight paths and weather conditions. The rules mentioned above... It will be dynamically adjusted based on the best historical case retrieved (for example, if the history is similar to crosswinds). If the effect is better, then from Revised to This enables personalized and precise navigation optimization.

[0036] Example 7, considering complex flight evaluation indicators (such as...) If the response to changes is presented in a traditional tabular format, it will be difficult for pilots and dispatchers to respond quickly under the pressure of their work schedules. Therefore, the visualization interface generated by the output response module includes a correlation analysis dashboard, which displays the correlation strength between different meteorological factors and different fuel consumption indicators as revealed by the correlation dynamic assessment model in the form of a heat map, and displays the historical changes of core key performance indicators and the predicted trajectory based on current decision recommendations in the form of trend lines.

[0037] The system also includes an early warning module, which is connected to the intelligent assessment core module and the adaptive decision support module, respectively. The early warning module is configured to automatically trigger multi-level early warning signals when the correlation score is lower than a preset threshold, and push the early warning information and corresponding structured decision suggestions to the user terminals of designated crew members or dispatch personnel. Specifically: The system uses the ECharts visualization component to develop a correlation analysis dashboard. The heat map dynamically displays the contribution intensity of the headwind component to thrust demand; the trend line plots the fuel flow curve over the past 2 hours in real time, and uses dashed lines to plot the future fuel consumption prediction trajectory after implementing optimization decisions (such as climb height).

[0038] The early warning module monitors real-time correlation scores (CS). A two-tiered threshold is set: when... When this is triggered, a Level II yellow alert is displayed on the airborne EFB interface with a prominent color indicating "The current route environment has an abnormal impact on fuel consumption"; when Furthermore, when the fuel efficiency index continues to decline, a Level 1 red alarm is triggered. The system automatically calls the satellite link or the ground ACARS gateway to directly push structured decision recommendations to the dispatch monitoring screen and send collaborative instructions to the dispatcher.

[0039] By employing the aforementioned technical means, the complex nonlinear fuel consumption correlation is transformed into intuitive and visual decision support, significantly enhancing the collaborative fuel-saving decision-making capabilities of the crew and dispatch in complex and dynamic route environments.

[0040] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic fuel consumption optimization system for civil aircraft based on real-time route data, characterized in that, include: The data storage module is used to store historical and real-time flight multimodal data; The input module, connected to the data storage module, is used to receive real-time route multimodal data from an external data source; The multi-source information fusion and feature extraction module is connected to the input module and the data storage module respectively. It is used to standardize and normalize the received multimodal data, and extract multidimensional feature vectors for evaluating the correlation from the multimodal data based on a preset feature system for quantifying the dynamic correlation between the route environment and fuel consumption. The route multimodal data includes high-altitude wind field data, ambient temperature deviation data, real-time flight path coordinate data, airspace congestion level data, engine performance monitoring data, and QAR stream data collected by airborne sensors. The feature system includes real-time meteorological impact features, flight path deviation features, air traffic control command constraint features, and flight performance features; the intelligent evaluation core module is connected to the multi-source information fusion and feature extraction module, and contains a correlation dynamic evaluation model. The correlation dynamic evaluation model is an ensemble learning model built on the gradient boosting decision tree algorithm. It is obtained by training historical multidimensional feature vectors and labeled data verified by QAR data through the ensemble learning algorithm. It is used to synchronously calculate and output flight performance index, fuel efficiency index and correlation score representing the correlation strength between the two based on the input real-time multidimensional feature vector. The correlation dynamic evaluation model uses SHAP interpretive technology to quantify the contribution of each input's real-time multidimensional feature vector to the flight performance index and fuel efficiency index, and reveals the correlation strength between different environmental factors and different flight status indicators in the form of a heat map. The adaptive decision support module, connected to the intelligent evaluation core module, is used to generate a structured report containing personalized cost index adjustment suggestions, altitude change schemes, and expected fuel savings predictions based on correlation scores, flight performance indices, fuel efficiency indices, and a preset decision rule base. The preset decision rule base combines production rules with case-based reasoning, and the rules in the decision rule base take the following form: IF correlation score AND fuel efficiency index AND Flight Performance Index THEN Execution: Adjust the cruise cost index to It is recommended to request a climb to a higher level. It is expected to improve fuel efficiency. Expected flight time changes ; Among them, threshold Range A, Range B, Parameter The specific value is dynamically determined based on statistical analysis of historical datasets and the results of case reasoning. The output response module, connected to the adaptive decision support module, is used to visualize the structured report and core evaluation indicators through a graphical user interface. The system also includes a model continuous learning module, connected to the intelligent evaluation core module and the data storage module, respectively, for periodically using newly added flight multimodal data and corresponding validation label data to incrementally train and optimize the correlation dynamic evaluation model in order to update the model parameters.

2. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, The input module includes: a standardized data interface unit for receiving the route multimodal data from the airborne communication addressing and reporting system, real-time weather radar, ground dispatch system and air traffic control automation terminal; and an edge computing unit deployed on the airborne processing terminal for performing preliminary cleaning, redundancy removal and formatting processing on the initial data.

3. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, The visualization interface generated by the output response module includes a correlation analysis dashboard, which displays the correlation strength between different meteorological factors and different fuel consumption indicators revealed by the correlation dynamic evaluation model in the form of a heat map, and displays the historical changes of core key performance indicators and the predicted trajectory based on current decision recommendations in the form of trend lines.

4. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that: The system also includes an early warning module, which is connected to the intelligent assessment core module and the adaptive decision support module respectively. The early warning module is configured to automatically trigger multi-level early warning signals when the correlation score is lower than a preset threshold, and push the early warning information and corresponding structured decision suggestions to the user terminal of the designated crew or dispatcher.

5. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, When processing multimodal data, the multi-source information fusion and feature extraction module performs Z-score standardization on features of different dimensions and calculates the crosswind drag influence factor based on the real-time crosswind component and Mach number deviation.

6. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, The formula for calculating the marginal contribution of each feature to the output result by the correlation dynamic evaluation model is as follows: in, For the first The contribution value of each feature, This is the baseline value output by the model.

7. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, The trigger condition for the model's continuous learning module to perform incremental training is: the root mean square error between the fuel flow predicted by the model and the actual QAR flow exceeds a preset threshold. .

8. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, When performing case reasoning, the adaptive decision support module uses the K-nearest neighbor algorithm to retrieve the historical best fuel-saving solutions under similar routes and weather conditions from the historical flight case database, and adjusts the parameters accordingly. Perform dynamic corrections.

9. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 1, characterized in that, The data storage module adopts a Hadoop-based distributed storage architecture and uses a time-series database to store airborne QAR stream data and high-frequency meteorological monitoring data.

10. The civil aircraft fuel consumption dynamic optimization system based on real-time route data according to claim 2, characterized in that, The edge computing unit is configured to perform window smoothing filtering before data upload and only upload data if the data variation rate exceeds a preset threshold. The upload operation is triggered at that time.

Images