QAR data decoding method and device based on large model, equipment and medium

Abstract

The invention discloses a QAR data decoding method and device based on a large model, equipment and a medium, and relates to the field of aviation flight data processing, and the method comprises the steps: carrying out the preprocessing of initial QAR data, generating processed QAR data, and employing a preset QAR analysis large model to analyze structural features, and determining a target protocol version and a parameter number; based on the target protocol version, analyzing the processed QAR data by using a preset QAR analysis large model to generate initial flight data, and determining a target sensor channel type and a target coding mode; converting the initial flight data into converted flight data by using a target mapping function according to the target sensor channel type and the target coding mode; standardized naming mapping is carried out on the fields to generate standardized flight data, parameter units and parameter value domains are unified to obtain decoded flight data, and the decoded flight data are sorted according to a preset time sequence and aligned according to timestamps to generate a visual flight report. And the automatic analysis capability of the QAR data is improved.

Description

Technical Field

[0001] The present invention relates to the field of aviation flight data processing, and in particular to a large model-based QAR data decoding method, device, equipment and medium. Background Art

[0002] The Quick Access Recorder (QAR), a critical flight data recorder on aircraft, records hundreds of critical flight parameters, such as attitude, speed, engine status, and control inputs. Its data format varies depending on the aircraft model, manufacturer, and system architecture. Traditionally, this approach has relied on static, template-based decoding, requiring experts to manually compile metadata such as field meanings, offsets, and scaling factors.

[0003] The current technical bottlenecks in QAR data decoding are as follows: Severe reliance on templates: Unable to parse without the manufacturer's template files; missing or changed templates will lead to decoding failures; Heavy rule configuration: Field annotations require manual maintenance by experts, resulting in heavy configuration workload and high error rate; Poor parsing generalization: Fields vary significantly between different models and manufacturers, lacking adaptability; Lack of intelligent assistance: The current decoding system cannot perform semantic reasoning, field completion, or interactive correction.

[0004] In summary, how to solve the problems of existing decoding methods such as strong dependence on templates, high cost of manual rule maintenance, and poor parsing adaptability is an urgent issue that needs to be solved. Summary of the Invention

[0005] In view of this, the present invention aims to provide a large-scale model-based QAR data decoding method, apparatus, device, and medium, which can solve the problems of existing decoding methods such as strong reliance on templates, high manual rule maintenance costs, and poor parsing adaptability. The specific solution is as follows:

[0006] In a first aspect, the present application provides a QAR data decoding method based on a large model, comprising: Acquire initial QAR data, preprocess the initial QAR data to generate processed QAR data, and analyze structural features of the processed QAR data using a preset QAR parsing model to determine a target protocol version and parameter number corresponding to the processed QAR data; Parsing the data segments corresponding to the parameter numbers in the processed QAR data using the preset QAR parsing model based on the target protocol version to generate initial flight data, and determining the target sensor channel type and target encoding method corresponding to the parameter numbers; constructing a target mapping function based on the initial flight data, and converting the initial flight data into converted flight data representing actual physical quantities using the target mapping function according to the target sensor channel type and target encoding method; The fields in the converted flight data are subjected to standardized naming mapping to generate standardized flight data, the parameter units and parameter value ranges in the standardized flight data are unified to obtain decoded flight data, and the decoded flight data are sorted according to a preset time sequence and aligned by timestamps to generate a visual flight report.

[0007] Optionally, analyzing the structural features of the processed QAR data using a preset QAR parsing model to determine a target protocol version and parameter number corresponding to the processed QAR data includes: Using a preset QAR parsing model to identify data segment features of the processed QAR data based on QAR structural patterns in a preset database; Determining a target protocol version corresponding to the processed QAR data according to the identified data segment features of the processed QAR data; Statistical analysis and semantic matching operations are performed on the processed QAR data to obtain parameter numbers corresponding to the processed QAR data.

[0008] Optionally, determining a target encoding method corresponding to the parameter number includes: Determining whether the data segment corresponding to the parameter number in the processed QAR data is encoded in a preset encoding mode according to a preset signal change pattern; The target encoding mode of the parameter number in the processed QAR data is determined according to the obtained judgment result.

[0009] Optionally, performing standardized naming mapping on the fields in the converted flight data to generate standardized flight data, and unifying parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, includes: Identifying the target name meaning of the field in the converted flight data through a pre-trained language model; Mapping the fields in the converted flight data that meet the preset name meaning similarity condition based on the target name meaning to corresponding standard parameter names in the preset knowledge graph to generate standardized flight data; identifying parameter units in the standardized flight data, and converting the parameter units in the standardized flight data into preset units according to preset parameter unit rules to generate unit-converted flight data; The parameter value range in the unit-converted flight data is standardized according to a preset aviation general range to obtain decoded flight data.

[0010] Optionally, sorting the decoded flight data according to a preset time sequence and aligning them by timestamps to generate a visual flight report includes: Using a dynamic time warping algorithm, the time series of each parameter in the decoded flight data are aligned according to a preset time sequence to obtain time-aligned flight data; identifying the time-aligned flight data based on flight parameter variation patterns to determine a target flight phase; Aligning the decoded flight data with target external source information according to timestamps, and then mapping the aligned target external source information to the corresponding decoded flight data to generate flight operation data; wherein the target external source information includes any one or more of aircraft cockpit voice, flight data recorder data, and flight parameter recording system documents; A visual flight report is generated according to the target flight phase and the flight operation data.

[0011] Optionally, the method further includes: Identifying abnormal data segments in the time-aligned flight data using a preset anomaly detection model; Based on a preset completion strategy, abnormal data segments in the time-aligned flight data are filled, and a credibility score is performed on the filled data to generate a visual flight report according to the credibility score.

[0012] Optionally, the visual flight report includes any one or more of a flight situation diagram, a parameter trend diagram, and an alarm event diagram.

[0013] In a second aspect, the present application provides a QAR data decoding device based on a large model, comprising: a version and number determination module, configured to obtain initial QAR data, pre-process the initial QAR data to generate processed QAR data, and analyze structural features of the processed QAR data using a preset QAR parsing model to determine a target protocol version and parameter number corresponding to the processed QAR data; a channel and encoding determination module, configured to parse the data segment corresponding to the parameter number in the processed QAR data using the preset QAR parsing model based on the target protocol version to generate initial flight data, and determine the target sensor channel type and target encoding method corresponding to the parameter number; a data conversion module, configured to construct a target mapping function based on the initial flight data, and convert the initial flight data into converted flight data representing actual physical quantities using the target mapping function according to the target sensor channel type and target encoding method; A report generation module is used to perform standardized naming mapping on the fields in the converted flight data to generate standardized flight data, unify the parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, sort the decoded flight data according to a preset time sequence and align them by timestamp to generate a visual flight report.

[0014] In a third aspect, the present application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned large model-based QAR data decoding method.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned large model-based QAR data decoding method.

[0016] In summary, the present application first obtains initial QAR data, pre-processes the initial QAR data to generate processed QAR data, and uses a preset QAR parsing model to analyze the structural characteristics of the processed QAR data to determine the target protocol version and parameter number corresponding to the processed QAR data; based on the target protocol version, the preset QAR parsing model is used to parse the data segment corresponding to the parameter number in the processed QAR data to generate initial flight data, and determine the target sensor channel type and target encoding method corresponding to the parameter number; constructs a target mapping function according to the initial flight data, and uses the target mapping function to convert the initial flight data into converted flight data representing actual physical quantities according to the target sensor channel type and target encoding method; performs standardized naming mapping on the fields in the converted flight data to generate standardized flight data, unifies the parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, and sorts the decoded flight data according to a preset time sequence and aligns them by timestamps to generate a visual flight report. As can be seen from the above, this application first obtains the initial QAR data and pre-processes it, then uses the preset QAR parsing large model to analyze the processed data, determine the corresponding target protocol version and parameter number, and then parse the initial flight data based on the protocol version, while clarifying the target sensor channel type and target encoding method corresponding to the parameter number. Afterwards, a target mapping function is constructed based on the initial flight data, and the initial flight data is converted into converted flight data representing the actual physical quantity by combining the target sensor channel and encoding method. Subsequently, the fields of the converted data are standardized in naming and mapping, and the parameter units and value ranges are unified to obtain the decoded flight data. Finally, the data is sorted in a preset time sequence and aligned by timestamp to generate a visual flight report. In this way, this application integrates the semantic understanding, data modeling and knowledge generalization capabilities of the large model, and targets key issues such as the complex source of QAR data, closed protocols, and difficult signal parsing, to build an end-to-end, adaptive decoding process, effectively improving the automatic parsing rate, parsing accuracy and system adaptability of QAR data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0018] Figure 1 This is a flow chart of a QAR data decoding method based on a large model disclosed in this application; Figure 2This is a flowchart of a specific large-model-based QAR data decoding method disclosed in this application; Figure 3 This is a schematic diagram of the structure of a QAR data decoding device based on a large model disclosed in this application; Figure 4 This is a structural diagram of an electronic device disclosed in this application. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] Currently, QAR, a critical flight data recorder on aircraft, records hundreds of key flight parameters, such as attitude, speed, engine status, and control inputs. Its data format varies depending on the aircraft model, manufacturer, and system architecture. For a long time, it has relied on template-based static decoding methods, requiring experts to manually compile metadata such as field meanings, offsets, and scaling factors. Current QAR data decoding faces the following technical bottlenecks: It relies heavily on templates: parsing cannot be performed independently of the manufacturer's template files; missing or altered templates result in decoding failures; cumbersome rule configuration: field annotations require manual maintenance by experts, resulting in a high configuration workload and error rate; poor parsing generalization: fields vary significantly between aircraft models and manufacturers, lacking adaptability; and lacking intelligent assistance: current decoding systems are unable to perform semantic reasoning, field completion, or interactive correction. To address these technical issues, this application discloses a large-scale model-based QAR data decoding method, apparatus, device, and medium. These methods address the existing decoding methods' strong reliance on templates, high manual rule maintenance costs, and poor parsing adaptability.

[0021] See also Figure 1 As shown, the embodiment of the present invention discloses a QAR data decoding method based on a large model, comprising:

[0022] Step S11: Acquire initial QAR data, pre-process the initial QAR data to generate processed QAR data, and use a preset QAR parsing model to analyze the structural characteristics of the processed QAR data to determine the target protocol version and parameter number corresponding to the processed QAR data.

[0023] In this embodiment, initial QAR data is first acquired from the aircraft, and then cleaned and formatted. Initial QAR data is often stored in binary or proprietary formats and contains a large number of undecoded flight parameters and signals. The initial QAR data can be first formatted and decapsulated. By determining whether the initial QAR data is in a standard format such as ARINC 717 or ARINC 429, the packet structure is disassembled and timestamp errors in the initial QAR data are corrected. Signal frames are then reordered in chronological order, and physical layer noise is processed using methods such as sliding window averaging and median filtering. Missing frames are then interpolated and filled to obtain processed QAR data.

[0024] In this embodiment, after obtaining processed QAR data, a pre-set QAR parsing model is used to identify data segment features of the processed QAR data based on QAR structural patterns in a pre-set database. The target protocol version corresponding to the processed QAR data is determined based on the identified data segment features of the processed QAR data. Statistical analysis and semantic matching are then performed on the processed QAR data to obtain the parameter numbers corresponding to the processed QAR data. It should be noted that due to significant differences in QAR protocols used by different aircraft models and manufacturers, traditional methods require manual review of protocol documentation, which is inefficient and has poor error tolerance. To determine the QAR data version on an aircraft, the pre-set QAR parsing model can be used to identify the header, frame length, number of channels, and other features of the processed QAR data segment by comparing them with structural patterns in a known protocol library, thereby determining the target protocol version of the processed QAR data. Simultaneously, statistical analysis and semantic matching are combined to identify characteristic parameters and their byte offsets, thereby obtaining the parameter numbers corresponding to the processed QAR data. Furthermore, based on existing parameter labels, pseudo-labeling and model self-learning can be performed on unlabeled data.

[0025] Step S12: Based on the target protocol version, the preset QAR parsing model is used to parse the data segment corresponding to the parameter number in the processed QAR data to generate initial flight data, and determine the target sensor channel type and target encoding method corresponding to the parameter number.

[0026] In this embodiment, the processed QAR data is deeply analyzed with the help of a preset QAR analysis model, and each data segment corresponding to the parameter number in the processed QAR data is accurately identified and extracted. Through the calculation and analysis of the model, the relatively scattered and disordered data segments are converted into structured initial flight data. These initial flight data can preliminarily reflect various basic information during the flight process.

[0027] Next, because flight data consists of multiple sensor channels, each with its own encoding scheme and signal amplitude range, it is necessary to identify the target sensor channel types in the initial flight data, such as analog voltage and current signals, digital signals, and event flags. Simultaneously, a determination is made based on a preset signal change pattern to determine whether the data segment corresponding to the parameter number in the processed QAR data uses a preset encoding scheme. Based on the determination result, the target encoding scheme for the parameter number in the processed QAR data is determined. Specifically, the signal change pattern of the processed QAR data can be used to determine whether it uses BCD encoding, two's complement encoding, IEEE 754 floating point, or other encoding schemes, thereby determining the target encoding scheme for the processed QAR data.

[0028] Step S13: constructing a target mapping function based on the initial flight data, and using the target mapping function according to the target sensor channel type and target encoding method to convert the initial flight data into converted flight data representing actual physical quantities.

[0029] In this embodiment, after obtaining initial flight data, a target mapping function is constructed between the channel signal and the real physical quantity, combining parameter characteristics, sensor channel signal characteristics, and encoding rules. This bridges the gap between the channel signal and the real physical quantity, using methods such as linear transformations and logarithmic relationships. These are commonly used. For example, linear transformations are often used for engine speed signals, while logarithmic forms are often used for pressure-related parameters. Subsequently, based on the target sensor channel type and encoding method, the target mapping function is invoked to transform the initial flight data, converting it into intuitive, real-world physical quantity data. This transformed flight data can then be directly used in scenarios such as flight status assessment and fault diagnosis.

[0030] Step S14: perform standardized naming mapping on the fields in the converted flight data to generate standardized flight data, unify the parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, sort the decoded flight data according to a preset time sequence and align them by timestamp to generate a visual flight report.

[0031] In this embodiment, due to the numerous flight data fields, their naming is inconsistent, and they contain numerous abbreviations and aliases. Therefore, semantic normalization of the fields is necessary. A pre-trained language model can be used to identify the target name meanings of the fields in the converted flight data. Based on the target name meanings, the fields in the converted flight data that meet preset name meaning similarity conditions are mapped to corresponding standard parameter names in a preset knowledge graph to generate standardized flight data. Parameter units in the standardized flight data are identified and converted to preset units according to preset parameter unit rules to generate unit-converted flight data. The parameter value ranges in the unit-converted flight data are normalized according to a preset aviation general range to generate decoded flight data. Specifically, the pre-trained language model identifies the target name meanings of the fields, for example, uniformly identifying "ALT," "HGT," and "Altitude" as "height." Semantic vector clustering is performed on similar fields, and these fields are automatically mapped to standard parameters in a unified knowledge graph to generate standardized flight data. Semantic disambiguation is performed using the context of other fields. For example, the context of the upper and lower fields "IAS" and "GS" can be used to help determine whether "SPD" represents airspeed or groundspeed. Understandably, different manufacturers and aircraft models in QAR data may use different units or scaling factors for the same parameter, necessitating unified conversion. This can be done by automatically identifying and converting units such as "ft" and "m," and "knots" and "km / h," generating flight data with unit conversion. Parameters are standardized to a common aviation range, such as a uniform temperature range of -50°C to +150°C. Finally, the initial QAR data is decoded, resulting in decoded flight data.

[0032] Furthermore, after decoding the initial QAR data, the state sequence of the complete flight process needs to be reconstructed in chronological order. This requires using a dynamic time warping algorithm to align the time series of each parameter in the decoded flight data according to a preset chronological order to obtain time-aligned flight data. The time-aligned flight data is then identified based on flight parameter variation patterns to determine the target flight phase. The decoded flight data is then timestamp-aligned with target external information, and the aligned target external information is then mapped to the corresponding decoded flight data to generate flight operation data. The target external information may include any one or more of cockpit voice, flight data recorder data, and flight parameter recording system documents. A visual flight report is generated based on the target flight phase and the flight operation data. Specifically, a dynamic time warping algorithm is used to align the time series of multiple parameters in the decoded flight data to ensure flight state consistency. The target flight phase, such as rollout, takeoff, climb, cruise, descent, and landing, is then identified based on the flight parameter variation patterns in the decoded flight data. A rules engine or weakly supervised model can also be used to automatically label each time window with labels such as "climb segment" or "approach segment." Furthermore, decoded flight data can integrate multiple sources, including cockpit voice, flight data recorder data, and flight parameter recording system documents, aligning data from different systems along a unified timeline. Voice commands in the CVR (Cockpit Voice Recorder) are then mapped to corresponding flight operation data points. A large model is then used to jointly model and reconstruct events across video, audio, text, and flight parameters. Ultimately, the decoded flight data is converted into highly readable and visually friendly structured reports or database formats, generating charts such as flight situation diagrams, parameter trend graphs, and alarm event graphs. Multiple formats, including JSON (JavaScript Object Notation) and CSV (Comma-Separated Values), are supported to facilitate subsequent analysis and archiving. A message queue is also provided to push visualized flight reports to flight quality systems or analytics platforms.

[0033] As can be seen from the above, the embodiment of the present application first obtains the initial QAR data and pre-processes it. Then, it uses the preset QAR parsing large model to analyze the processed data, determine the corresponding target protocol version and parameter number, and then parse the initial flight data based on the protocol version. At the same time, it clarifies the target sensor channel type and target encoding method corresponding to the parameter number. After that, a target mapping function is constructed based on the initial flight data. In combination with the target sensor channel and encoding method, the initial flight data is converted into converted flight data representing the actual physical quantity. Subsequently, the fields of the converted data are standardized and named, and the parameter units and value ranges are unified to obtain the decoded flight data. Finally, the data is sorted in a preset time sequence and aligned by timestamp to generate a visual flight report. In this way, the present application integrates the semantic understanding, data modeling and knowledge generalization capabilities of the large model to address key issues such as the complex source of QAR data, closed protocols, and difficult signal parsing. It constructs an end-to-end, adaptive decoding process, effectively improving the automatic parsing rate, parsing accuracy and system adaptability of QAR data.

[0034] See also Figure 2 As shown, in order to avoid the interference of abnormal data on QAR data analysis, an embodiment of the present invention discloses a QAR data decoding method based on a large model, including:

[0035] Step S21: using a preset anomaly detection model to identify abnormal data segments in the time-aligned flight data.

[0036] In this embodiment, a preset anomaly detection model is used to identify abnormal data segments in the time-aligned flight data. The model integrates rule-based, statistical models, and machine learning models to identify various abnormal data segments such as mutations, static values, drifts, etc. in the time-aligned flight data.

[0037] Step S22: Fill in the abnormal data segments in the time-aligned flight data based on a preset filling strategy, and perform a credibility score on the filled data to generate a visual flight report according to the credibility score.

[0038] In this embodiment, after identifying abnormal data segments, the abnormal data segments in the time-aligned flight data are filled based on a preset filling strategy. The filling strategy covers adjacent data interpolation, reconstruction based on physical modeling, and predictive filling based on historical models. At the same time, the credibility of the filled data is scored, and finally a visual flight report is generated based on these credibility scores, providing valuable reference for subsequent related decisions.

[0039] As can be seen from the above, the present embodiment prevents abnormal data from interfering with subsequent analysis at the source, ensuring the fundamental reliability of the data used for analysis. Furthermore, the mechanism for assigning credibility scores to supplemented data adds a clear reliability indicator to the supplemented data. The visual flight report generated based on this score also provides a more reliable and intuitive reference for subsequent decision-making. Together, these two factors enhance the accuracy and practicality of flight data processing.

[0040] See also Figure 3 As shown, the embodiment of the present invention discloses a QAR data decoding device based on a large model, comprising: A version and number determination module 11 is configured to obtain initial QAR data, pre-process the initial QAR data to generate processed QAR data, and analyze structural features of the processed QAR data using a preset QAR parsing model to determine a target protocol version and parameter number corresponding to the processed QAR data; a channel and encoding determination module 12 for parsing the data segment corresponding to the parameter number in the processed QAR data using the preset QAR parsing model based on the target protocol version to generate initial flight data, and determining the target sensor channel type and target encoding method corresponding to the parameter number; a data conversion module 13, configured to construct a target mapping function based on the initial flight data, and convert the initial flight data into converted flight data representing actual physical quantities using the target mapping function according to the target sensor channel type and target encoding method; The report generation module 14 is used to perform standardized naming mapping on the fields in the converted flight data to generate standardized flight data, unify the parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, sort the decoded flight data according to a preset time sequence and align them by timestamp to generate a visual flight report.

[0041] As can be seen from the above, this application first obtains the initial QAR data and pre-processes it, then uses the preset QAR parsing large model to analyze the processed data, determine the corresponding target protocol version and parameter number, and then parse the initial flight data based on the protocol version, while clarifying the target sensor channel type and target encoding method corresponding to the parameter number. Afterwards, a target mapping function is constructed based on the initial flight data, and the initial flight data is converted into converted flight data representing the actual physical quantity by combining the target sensor channel and encoding method. Subsequently, the fields of the converted data are standardized in naming and mapping, and the parameter units and value ranges are unified to obtain the decoded flight data. Finally, the data is sorted in a preset time sequence and aligned by timestamp to generate a visual flight report. In this way, this application integrates the semantic understanding, data modeling and knowledge generalization capabilities of the large model, and targets key issues such as the complex source of QAR data, closed protocols, and difficult signal parsing, to build an end-to-end, adaptive decoding process, effectively improving the automatic parsing rate, parsing accuracy and system adaptability of QAR data.

[0042] In some specific implementations, the version and number determination module 11 includes: A feature recognition unit, configured to recognize data segment features of the processed QAR data based on QAR structural patterns in a preset database using a preset QAR parsing model; a version determining unit, configured to determine a target protocol version corresponding to the processed QAR data according to the identified data segment features of the processed QAR data; The number acquisition unit is used to perform statistical analysis and semantic matching operations on the processed QAR data to obtain the parameter number corresponding to the processed QAR data.

[0043] In some specific implementations, the version and number determination module 11 includes: a data segment determination unit, configured to determine whether a data segment corresponding to a parameter number in the processed QAR data is encoded in a preset encoding mode according to a preset signal change pattern; The encoding mode determination unit is used to determine the target encoding mode of the parameter number in the processed QAR data according to the obtained judgment result.

[0044] In some specific embodiments, the report generating module 14 includes: a name meaning recognition unit, configured to recognize the target name meaning of a field in the converted flight data by using a pre-trained language model; A first data generating unit is configured to map, based on the target name meaning, fields in the converted flight data that meet a preset name meaning similarity condition to corresponding standard parameter names in a preset knowledge graph to generate standardized flight data; a second data generating unit, configured to identify parameter units in the standardized flight data and convert the parameter units in the standardized flight data into preset units according to preset parameter unit rules to generate unit-converted flight data; The third data generating unit is configured to perform standardization processing on the parameter value range in the flight data after the unit conversion according to a preset aviation general range to obtain decoded flight data.

[0045] In some specific embodiments, the report generating module 14 includes: a data acquisition unit, configured to align the time series of the parameters in the decoded flight data according to a preset time sequence using a dynamic time warping algorithm, and acquire time-aligned flight data; a flight phase determination unit, configured to identify the time-aligned flight data based on a flight parameter variation pattern to determine a target flight phase; a flight operation data generating unit, configured to align the decoded flight data with target exogenous information according to timestamps, and then map the aligned target exogenous information to the corresponding decoded flight data to generate flight operation data; wherein the target exogenous information includes any one or more of aircraft cockpit voice, flight data recorder data, and flight parameter recording system documents; The flight report generating unit is used to generate a visual flight report according to the target flight phase and the flight operation data.

[0046] In some specific implementations, the large model-based QAR data decoding device may further include: an abnormal data segment identification module, configured to identify abnormal data segments in the time-aligned flight data using a preset anomaly detection model; The visual flight report generation module is used to fill in the abnormal data segments in the time-aligned flight data based on a preset filling strategy, and to perform credibility scoring on the filled data to generate a visual flight report according to the credibility score.

[0047] In some specific implementations, the visual flight report includes any one or more of a flight situation diagram, a parameter trend diagram, and an alarm event diagram.

[0048] Furthermore, the embodiment of the present application also discloses an electronic device, Figure 4 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content in the diagram should not be considered as any limitation to the scope of application of the present application.

[0049] Figure 4This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of the present application. The electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps of the large model-based QAR data decoding method disclosed in any of the aforementioned embodiments. Furthermore, the electronic device 20 in this embodiment may be a computer.

[0050] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device. The communication protocol it follows is any communication protocol that can be applied to the technical solution of this application and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world. Its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0051] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or CD, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0052] The operating system 221 is used to manage and control the hardware devices and computer program 222 on the electronic device 20, and can be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of implementing the large model-based QAR data decoding method disclosed in any of the aforementioned embodiments and executed by the electronic device 20, the computer program 222 may further include a computer program capable of performing other specific tasks.

[0053] Furthermore, this application discloses a computer-readable storage medium for storing a computer program. When executed by a processor, the computer program implements the aforementioned large-model-based QAR data decoding method. The specific steps of this method can be found in the corresponding content disclosed in the aforementioned embodiments and will not be further elaborated here.

[0054] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. Reference can be made to the descriptions of the identical or similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and the relevant parts can be referred to the descriptions of the methods.

[0055] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0056] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0057] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0058] The above is a detailed introduction to the technical solution provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for those skilled in the art, according to the ideas of the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A QAR data decoding method based on a large model, characterized in that: include: Acquire initial QAR data, preprocess the initial QAR data to generate processed QAR data, and analyze structural features of the processed QAR data using a preset QAR parsing model to determine a target protocol version and parameter number corresponding to the processed QAR data; Parsing the data segments corresponding to the parameter numbers in the processed QAR data using the preset QAR parsing model based on the target protocol version to generate initial flight data, and determining the target sensor channel type and target encoding method corresponding to the parameter numbers; constructing a target mapping function based on the initial flight data, and converting the initial flight data into converted flight data representing actual physical quantities using the target mapping function according to the target sensor channel type and target encoding method; The fields in the converted flight data are subjected to standardized naming mapping to generate standardized flight data, the parameter units and parameter value ranges in the standardized flight data are unified to obtain decoded flight data, and the decoded flight data are sorted according to a preset time sequence and aligned by timestamps to generate a visual flight report.

2. The QAR data decoding method based on a large model according to claim 1, characterized in that: The using a preset QAR parsing model to analyze the structural characteristics of the processed QAR data to determine the target protocol version and parameter number corresponding to the processed QAR data includes: Using a preset QAR parsing model to identify data segment features of the processed QAR data based on QAR structural patterns in a preset database; Determining a target protocol version corresponding to the processed QAR data according to the identified data segment features of the processed QAR data; Statistical analysis and semantic matching operations are performed on the processed QAR data to obtain parameter numbers corresponding to the processed QAR data.

3. The large model-based QAR data decoding method according to claim 1, characterized in that: Determining the target encoding method corresponding to the parameter number includes: Determining whether the data segment corresponding to the parameter number in the processed QAR data is encoded in a preset encoding mode according to a preset signal change pattern; The target encoding mode of the parameter number in the processed QAR data is determined according to the obtained judgment result.

4. The large model-based QAR data decoding method according to claim 1, characterized in that: The step of performing standardized naming mapping on the fields in the converted flight data to generate standardized flight data, and unifying the parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, includes: Identifying the target name meaning of the field in the converted flight data through a pre-trained language model; Mapping the fields in the converted flight data that meet the preset name meaning similarity condition based on the target name meaning to corresponding standard parameter names in the preset knowledge graph to generate standardized flight data; identifying parameter units in the standardized flight data, and converting the parameter units in the standardized flight data into preset units according to preset parameter unit rules to generate unit-converted flight data; The parameter value range in the unit-converted flight data is standardized according to a preset aviation general range to obtain decoded flight data.

5. The large model-based QAR data decoding method according to any one of claims 1 to 4, characterized in that: The step of sorting the decoded flight data according to a preset time sequence and aligning the data by timestamps to generate a visual flight report includes: Using a dynamic time warping algorithm, the time series of each parameter in the decoded flight data are aligned according to a preset time sequence to obtain time-aligned flight data; identifying the time-aligned flight data based on flight parameter variation patterns to determine a target flight phase; Aligning the decoded flight data with target external source information according to timestamps, and then mapping the aligned target external source information to the corresponding decoded flight data to generate flight operation data; wherein the target external source information includes any one or more of aircraft cockpit voice, flight data recorder data, and flight parameter recording system documents; A visual flight report is generated according to the target flight phase and the flight operation data.

6. The large model-based QAR data decoding method according to claim 5, characterized in that: Also includes: Identifying abnormal data segments in the time-aligned flight data using a preset anomaly detection model; Based on a preset completion strategy, abnormal data segments in the time-aligned flight data are filled, and a credibility score is performed on the filled data to generate a visual flight report according to the credibility score.

7. The QAR data decoding method based on a large model according to claim 1, characterized in that: The visual flight report includes any one or more of a flight situation diagram, a parameter trend diagram, and an alarm event diagram.

8. A QAR data decoding device based on a large model, characterized in that: include: a version and number determination module, configured to obtain initial QAR data, pre-process the initial QAR data to generate processed QAR data, and analyze structural features of the processed QAR data using a preset QAR parsing model to determine a target protocol version and parameter number corresponding to the processed QAR data; a channel and encoding determination module, configured to parse the data segment corresponding to the parameter number in the processed QAR data using the preset QAR parsing model based on the target protocol version to generate initial flight data, and determine the target sensor channel type and target encoding method corresponding to the parameter number; a data conversion module, configured to construct a target mapping function based on the initial flight data, and convert the initial flight data into converted flight data representing actual physical quantities using the target mapping function according to the target sensor channel type and target encoding method; A report generation module is used to perform standardized naming mapping on the fields in the converted flight data to generate standardized flight data, unify the parameter units and parameter value ranges in the standardized flight data to obtain decoded flight data, sort the decoded flight data according to a preset time sequence and align them by timestamp to generate a visual flight report.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the large model-based QAR data decoding method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that Used to store a computer program; wherein, when the computer program is executed by a processor, the large model-based QAR data decoding method according to any one of claims 1 to 7 is implemented.

Images