Method, device, equipment and medium for processing in-airport fault data during pilot training based on a data middle platform

Abstract

The present invention discloses a method, device, equipment and medium for processing fault data in an airport during pilot training based on a data middle platform, including: traversing a number of adjustment devices on a target training flight device, and respectively setting a number of outlier values for each adjustment device; binding the adjustment device, an outlier value corresponding to the adjustment device, and a standard processing action of the outlier value under the adjustment device to obtain a fault processing data set corresponding to each adjustment device; adjusting a preset neural network according to a predicted processing action output by the preset neural network and a standard processing action corresponding to the predicted processing action; obtaining a target predicted processing action corresponding to a target outlier value of a target fault device; and determining whether a to-be-corrected fault processing action is qualified. The present invention belongs to the field of aviation data processing. The present invention can improve the efficiency of flight training.

Description

Technical Field

[0001] The present invention relates to the field of aviation data processing, and particularly to a method, device, equipment, and medium for processing in-airport fault data during pilot training based on a data middle platform. Background Art

[0002] Pilot training is a systematic process aimed at training pilots to master flight skills, aviation knowledge, and emergency handling capabilities. This process is usually divided into several stages and may vary according to the pilot's ultimate goal (such as commercial pilot, private pilot, military pilot, etc.).

[0003] Generally, pilots in need of training will be trained in flight training equipment. When a simulation device fails, it is usually handled by an instructor, which is inefficient and costly in terms of labor. Therefore, there is an urgent need for an efficient training method. Summary of the Invention

[0004] By providing a method, device, equipment, and medium for processing in-airport fault data during pilot training based on a data middle platform, the present invention solves the technical problem of low training efficiency in fault handling training in the prior art and achieves the technical effect of improving the training efficiency of fault handling training.

[0005] In a first aspect, the present invention provides a method for processing in-airport fault data during pilot training based on a data middle platform, the method including:

[0006] Traverse a number of adjustment devices on a target training flight device, and respectively set a number of abnormal values for each adjustment device. Wherein, when an adjustment device is in the abnormal value corresponding to the adjustment device, the adjustment device is in a fault state;

[0007] For each adjustment device, bind the adjustment device, an abnormal value corresponding to the adjustment device, and a standard handling action of the abnormal value under the adjustment device to obtain a fault handling data set corresponding to each adjustment device. Each standard handling action includes a handling time limit, an adjustment target, and a body posture amplitude;

[0008] Input the fault handling data sets corresponding to the number of adjustment devices into a preset neural network, and adjust the preset neural network according to the predicted handling action output by the preset neural network and the standard handling action corresponding to the predicted handling action;

[0009] When the target flight device is in a training state, input the target abnormal value of the target fault device into the adjusted preset neural network to obtain a target predicted handling action corresponding to the target abnormal value of the target fault device;

[0010] Obtain the to-be-corrected fault handling action of the first target, and determine whether the to-be-corrected fault handling action is qualified according to the difference between the target predicted handling action and the to-be-corrected fault handling action.

[0011] Further, adjust the preset neural network according to the predicted handling action output by the preset neural network and the standard handling action corresponding to the predicted handling action, including:

[0012] Adjust the neural network parameters of the preset neural network according to the difference between the predicted handling action and the standard handling action corresponding to the predicted handling action;

[0013] When the maximum iteration parameter is reached or the preset training requirement is satisfied, save the latest neural network parameters, and use the preset neural network under the latest neural network parameters as the adjusted preset neural network.

[0014] Further, determining whether the to-be-corrected fault handling action is qualified includes:

[0015] When the difference between the target predicted handling action and the to-be-corrected fault handling action is greater than or equal to the difference threshold, the to-be-corrected fault handling action is unqualified;

[0016] When the difference between the target predicted handling action and the to-be-corrected fault handling action is less than the difference threshold, the to-be-corrected fault handling action is qualified.

[0017] Further, determining the difference between the target predicted handling action and the to-be-corrected fault handling action includes:

[0018] ,

[0019] where, is the difference between the i-th target predicted handling action and the i-th to-be-corrected fault handling action, both a and b are preset weights, is the processing time limit of the i-th target predicted handling action, is the processing time limit of the i-th to-be-corrected fault handling action, is the adjustment target of the i-th target predicted handling action, is the adjustment target of the i-th to-be-corrected fault handling action, is the body posture amplitude vector of the i-th target predicted handling action, is the body posture amplitude vector of the i-th to-be-corrected fault handling action.

[0020] Further, the method further includes:

[0021] Sort in descending order according to the importance of several adjustment devices to obtain a standard sorting;

[0022] When the number of target faulty devices is greater than 1, obtain the adjustment order of the first target;

[0023] According to the adjustment order of the first target and the standard sorting, determine whether the adjustment order of the first target is qualified.

[0024] Further, according to the adjustment order of the first target and the standard sorting, determining whether the adjustment order of the first target is qualified includes:

[0025] According to the adjustment order of the first target, determine whether the descending order sorting of the adjustment order of the first target can correspond to the standard sorting;

[0026] If it can, the adjustment order of the first target is qualified; if not, the adjustment order of the first target is unqualified.

[0027] Further, performing a missing judgment on the fault processing data set further includes:

[0028] ,

[0029] Among them, is the sequence vector of the fault processing data set, is the mean of the sequence vector, f is the variance of the sequence vector, c and d are both preset weights, is the weighted density of the sequence vector of the fault processing data set.

[0030] In a second aspect, the present invention provides a device for processing in-airport fault data during pilot training based on a data middle platform. The device includes:

[0031] A traversal module for traversing a number of adjustment devices on the target training flight device, and respectively setting a number of outliers for each adjustment device. Among them, when the adjustment device is under the outlier corresponding to the adjustment device, the adjustment device is in a fault state;

[0032] A binding module for, for each adjustment device, binding the adjustment device, an outlier corresponding to the adjustment device, and a standard processing action of the outlier under the adjustment device to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude;

[0033] An adjustment module for inputting the fault processing data sets corresponding to a number of adjustment devices into a preset neural network, and adjusting the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action;

[0034] A prediction module, configured to input the target abnormal value of the target faulty device into the adjusted preset neural network when the target flight device is in a training state, and obtain the target prediction processing action corresponding to the target abnormal value of the target faulty device;

[0035] A judgment module, configured to obtain the to-be-corrected fault processing action of the first target, and judge whether the to-be-corrected fault processing action is qualified according to the difference between the target prediction processing action and the to-be-corrected fault processing action.

[0036] In a third aspect, the present invention provides an electronic device, including:

[0037] A processor;

[0038] A memory for storing executable instructions of the processor;

[0039] Wherein, the processor is configured to execute to implement the method for processing in-airport fault data during pilot training based on the data middle platform provided in the first aspect.

[0040] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor of the electronic device, enabling the electronic device to execute and implement the method for processing in-airport fault data during pilot training based on the data middle platform provided in the first aspect.

[0041] One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

[0042] The present invention provides a method for processing in-airport fault data during pilot training based on a data middle platform. The method includes: traversing a number of adjustment devices on a target training flight device, and respectively setting a number of outlier values for each adjustment device. Wherein, when the adjustment device is under the outlier value corresponding to the adjustment device, the adjustment device is in a fault state; for each adjustment device, binding the adjustment device, an outlier value corresponding to the adjustment device, and a standard processing action of the outlier value under the adjustment device to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude; inputting the fault processing data sets corresponding to a number of adjustment devices into a preset neural network, and adjusting the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action; when the target flight device is in a training state, inputting the target outlier value of the target fault device into the adjusted preset neural network to obtain a target predicted processing action corresponding to the target outlier value of the target fault device; obtaining a to-be-corrected fault processing action of a first target, and judging whether the to-be-corrected fault processing action is qualified according to the difference between the target predicted processing action and the to-be-corrected fault processing action. By first training the neural network to obtain the predicted processing action and comparing the predicted processing action with the to-be-corrected fault processing action, the present invention can better guide the first target, ensure the training effect of the first target, save labor costs, and improve the training efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0044] Figure 1 Flow chart of the method for processing in-airport fault data during pilot training provided by the present invention based on a data middle platform. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] By providing a method for processing in-airport fault data during pilot training in the embodiments of the present invention, the technical problem of low training efficiency in fault handling training in the prior art is solved.

[0046] The technical solution of the present invention for solving the above technical problem has the following general idea:

[0047] Method for processing in-airport fault data during pilot training based on a data middle platform. The method includes: traversing a number of adjustment devices on a target training flight device, and respectively setting a number of abnormal values for each adjustment device. Wherein, when the adjustment device is under the abnormal value corresponding to the adjustment device, the adjustment device is in a fault state; for each adjustment device, binding the adjustment device, an abnormal value corresponding to the adjustment device, and a standard processing action of the abnormal value under the adjustment device to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude; inputting the fault processing data sets corresponding to a number of adjustment devices into a preset neural network, and adjusting the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action; when the target flight device is in a training state, inputting the target abnormal value of the target fault device into the adjusted preset neural network to obtain a target predicted processing action corresponding to the target abnormal value of the target fault device; obtaining a to-be-corrected fault processing action of a first target, and judging whether the to-be-corrected fault processing action is qualified according to the difference between the target predicted processing action and the to-be-corrected fault processing action.

[0048] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0049] First, it should be noted that the term "and / or" appearing in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the front and rear associated objects.

[0050] The present invention provides a method for processing in-airport fault data during pilot training based on a data middle platform as shown in Figure 1 and includes steps S11 - S15:

[0051] Regarding step S11, traverse a number of adjustment devices on a target training flight device, and respectively set a number of abnormal values for each adjustment device. Wherein, when the adjustment device is under the abnormal value corresponding to the adjustment device, the adjustment device is in a fault state.

[0052] The target training flight device refers to a flight device (or an airplane) used for training pilots. There are a number of adjustment devices on the airplane. The adjustment device refers to a device on the flight device that needs to be adjusted by the pilot, such as a landing gear system, an instrument panel, an electrical system, and an environmental control system, etc. It can be understood that the adjustment device may malfunction, and when a malfunction occurs, it is necessary to adjust the adjustment device in time to avoid greater losses.

[0053] The outlier of each adjustment device can be obtained based on the data middle platform. When the adjustment device is under the outlier corresponding to the adjustment device, the adjustment device is in a fault state.

[0054] Regarding step S12, for each adjustment device, bind the adjustment device, an outlier corresponding to the adjustment device, and the standard processing action of the outlier under the adjustment device to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude.

[0055] Bind the adjustment device and an outlier corresponding to the adjustment device. Whenever there is an outlier in the adjustment device, bind it once to obtain a fault processing data set corresponding to the adjustment device.

[0056] The standard processing action refers to the reference action for processing the outlier under the adjustment device. Each standard processing action can be described in three aspects: processing time limit, adjustment target, and body posture amplitude. The processing time limit refers to the reference duration required to process the outlier. The adjustment target refers to adjusting the outlier of the adjustment device to the normal value. The body posture amplitude refers to the adjustment actions performed during the processing time limit.

[0057] It can be understood that the more continuous the fault processing data in the fault processing data set, the better the training effect of the preset training model in the following text. Therefore, the present invention also performs a missing value judgment on the fault processing data set, and further includes:

[0058] ,

[0059] Among them, is the sequence vector of the fault processing data set, is the mean value of the sequence vector, f is the variance of the sequence vector, and c and d are both preset weights. is the weighted density of the sequence vector of the fault processing data set.

[0060] When the weighted density of the sequence vector of the fault processing data set is greater than the corresponding set threshold, it indicates that the data in the fault processing data set is less continuous, and the training effect is worse when using this fault processing data set for training. Therefore, when the weighted density of the sequence vector of the fault processing data set is greater than the corresponding set threshold, interpolation can be used to fill the fault processing data set to improve the training effect of the preset training neural network model.

[0061] Regarding step S13, input the fault handling data sets corresponding to several adjustment devices into a preset neural network, and adjust the preset neural network according to the predicted handling actions output by the preset neural network and the standard handling actions corresponding to the predicted handling actions.

[0062] After inputting the fault handling data sets corresponding to several adjustment devices into the preset neural network, it includes: adjusting the neural network parameters of the preset neural network according to the difference between the predicted handling action and the standard handling action corresponding to the predicted handling action; when reaching the maximum iteration parameter or meeting the preset training requirements, save the latest neural network parameters, and use the preset neural network under the latest neural network parameters as the adjusted preset neural network.

[0063] The maximum iteration parameter or meeting the preset training requirements can be determined according to the actual situation and is not limited here. The preset neural network is used to predict the handling actions under the abnormal values of the adjustment device to obtain the corresponding predicted handling actions. The neural network parameters can include weights, biases, learning rates, activation functions, and the number and size of hidden layers, etc.

[0064] Regarding step S14, when the target flying device is in the training state, input the target abnormal value of the target fault device into the adjusted preset neural network to obtain the target predicted handling action corresponding to the target abnormal value of the target fault device.

[0065] When the target flying device is in the training state, it means that the flying device is in the simulation state. At this time, the target flying device can simulate the abnormal values of the adjustment device (it should be noted that at this time, the target flying device is in the simulation state and is not in the actual flight process).

[0066] By inputting the target abnormal value of the target fault device into the adjusted preset neural network, the target predicted handling action corresponding to the target abnormal value of the target fault device can be obtained.

[0067] Regarding step S15, obtain the to-be-corrected fault handling action of the first target, and judge whether the to-be-corrected fault handling action is qualified according to the difference between the target predicted handling action and the to-be-corrected fault handling action.

[0068] The first target refers to a pilot who is under training guidance. The pilot is in the target flying device and adjusts the target abnormal value of the target fault device to obtain the to-be-corrected fault handling action of the first target. The to-be-corrected fault handling action also includes the handling time limit, adjustment target, and body posture amplitude.

[0069] Determine whether the fault handling action to be corrected is qualified, including: when the difference between the target prediction handling action and the fault handling action to be corrected is greater than or equal to the difference threshold, the fault handling action to be corrected is unqualified; when the difference between the target prediction handling action and the fault handling action to be corrected is less than the difference threshold, the fault handling action to be corrected is qualified.

[0070] Determine the difference between the target prediction handling action and the fault handling action to be corrected, including:

[0071] ,

[0072] wherein, is the difference between the i-th target prediction handling action and the i-th fault handling action to be corrected, both a and b are preset weights, is the processing time limit of the i-th target prediction handling action, is the processing time limit of the i-th fault handling action to be corrected, is the adjustment target of the i-th target prediction handling action, is the adjustment target of the i-th fault handling action to be corrected, is the body posture amplitude vector of the i-th target prediction handling action, is the body posture amplitude vector of the i-th fault handling action to be corrected.

[0073] The present invention takes the processing time limit as the core, and respectively normalizes the body posture amplitude vector and the adjustment target. After processing, none of the three parameters retains the dimension, only the parameter itself is retained. The present invention takes the processing time limit as the axis to reflect the changes of the body posture amplitude vector and the adjustment target with respect to time, and simultaneously incorporates the processing time limit, the body posture amplitude vector, and the adjustment target as indicators for evaluating the difference between the target prediction handling action and the fault handling action to be corrected, so as to achieve a consistent judgment and can more accurately reflect whether the fault handling action to be corrected is qualified.

[0074] The method further includes: sorting in descending order according to the importance of several adjustment devices to obtain a standard sorting; when the number of target fault devices is greater than 1, obtaining the adjustment order of the first target; and judging whether the adjustment order of the first target is qualified according to the adjustment order of the first target and the standard sorting.

[0075] When the number of target fault devices is greater than 1, it means that multiple fault devices need to be adjusted continuously, but the importance of different adjustment devices is different. Therefore, the fault devices need to be adjusted in sequence.

[0076] Specifically, it includes: judging whether the descending order of the adjustment order of the first target can correspond to the standard order according to the adjustment order of the first target; if it can, the adjustment order of the first target is qualified; if not, the adjustment order of the first target is unqualified.

[0077] For example, the standard order is A, B, C, D, E, F, G (A, B, C, D, E, F, G are all adjustment devices),

[0078] The adjustment order of the first target is A, C, D. A-C-D belongs to the order of A-B-C-D and can correspond to the standard order, so the adjustment order of the first target is qualified.

[0079] In summary, the present invention provides a method for processing in-airport fault data during pilot training based on a data middle platform. The method includes: traversing a number of adjustment devices on a target training flight device, and respectively setting a number of abnormal values for each adjustment device. Wherein, when the adjustment device is under the abnormal value corresponding to the adjustment device, the adjustment device is in a fault state; for each adjustment device, binding the adjustment device, an abnormal value corresponding to the adjustment device, and the standard processing action of the abnormal value under the adjustment device to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude; inputting the fault processing data sets corresponding to a number of adjustment devices into a preset neural network, and adjusting the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action; when the target flight device is in a training state, inputting the target abnormal value of the target fault device into the adjusted preset neural network to obtain a target predicted processing action corresponding to the target abnormal value of the target fault device; obtaining the to-be-corrected fault processing action of the first target, and judging whether the to-be-corrected fault processing action is qualified according to the difference between the target predicted processing action and the to-be-corrected fault processing action. By first training the neural network to obtain the predicted processing action and comparing the predicted processing action with the to-be-corrected fault processing action, the present invention can better guide the first target, ensure the training effect of the first target, save labor costs, and improve the training efficiency.

[0080] Based on the same inventive concept, the present invention provides a device for processing in-airport fault data during pilot training based on a data middle platform. The device includes:

[0081] A traversing module, configured to traverse a number of adjustment devices on a target training flight device, and respectively set a number of abnormal values for each adjustment device. Wherein, when the adjustment device is under the abnormal value corresponding to the adjustment device, the adjustment device is in a fault state;

[0082] A binding module, for each adjustment device, binds the adjustment device, an outlier value corresponding to the adjustment device, and a standard processing action for the outlier value under the adjustment device, to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude;

[0083] An adjustment module, for inputting the fault processing data sets corresponding to several adjustment devices into a preset neural network, and adjusting the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action;

[0084] A prediction module, when the target flight device is in a training state, inputs the target outlier value of the target fault device into the adjusted preset neural network, and obtains the target predicted processing action corresponding to the target outlier value of the target fault device;

[0085] A judgment module, for obtaining the to-be-corrected fault processing action of the first target, and judging whether the to-be-corrected fault processing action is qualified according to the difference between the target predicted processing action and the to-be-corrected fault processing action.

[0086] Based on the same inventive concept, the present invention provides an electronic device, including:

[0087] A processor;

[0088] A memory for storing instructions executable by the processor;

[0089] Wherein, the processor is configured to execute to implement the method for processing in-airport fault data during pilot training based on the data middle platform as provided above.

[0090] Based on the same inventive concept, the present invention further provides a non-transitory computer-readable storage medium. When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device can execute to implement the method for processing in-airport fault data during pilot training based on the data middle platform as provided above.

[0091] Since the electronic device introduced in this embodiment is the electronic device used for implementing the method for information processing in the embodiment of the present invention, based on the method for information processing introduced in the embodiment of the present invention, those skilled in the art can understand the specific implementation manners and various variation forms of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiment of the present invention will not be described in detail here. As long as the electronic device used by those skilled in the art to implement the method for information processing in the embodiment of the present invention belongs to the scope protected by the present invention.

[0092] Those skilled in the art will appreciate that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0093] The present invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart illustrations and / or block diagrams, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device create means for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0094] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0095] These computer program instructions may also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are performed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0096] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0097] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. A method for processing in-airport fault data during pilot training based on a data middle platform, characterized in that, The method includes: Traverse a number of adjustment devices on the target training flight device, and set a number of outlier values for each adjustment device. When the adjustment device is under the outlier value corresponding to the adjustment device, the adjustment device is in a faulty state; For each adjustment device, bind the adjustment device, an outlier value corresponding to the adjustment device, and a standard processing action of the outlier value under the adjustment device to obtain a fault processing data set corresponding to each adjustment device. Each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude; Input the fault processing data sets corresponding to a number of adjustment devices into a preset neural network, and adjust the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action; When the target flight device is in a training state, input the target outlier value of the target faulty device into the adjusted preset neural network to obtain a target predicted processing action corresponding to the target outlier value of the target faulty device; Obtain the to-be-corrected fault processing action of the first target, and judge whether the to-be-corrected fault processing action is qualified according to the difference between the target predicted processing action and the to-be-corrected fault processing action. Determining the difference between the target predicted processing action and the to-be-corrected fault processing action includes: ， in, For the The difference between the target prediction processing action and the i-th fault processing action to be corrected, a and b are both preset weights, Predict the processing time limit of the processing action for the i-th target, is the processing time limit of the i-th fault processing action to be corrected, The adjustment target for the i-th target prediction processing action, is the adjustment target of the i-th fault handling action to be corrected, The body posture magnitude vector for the i-th target prediction processing action, is the body posture amplitude vector of the i-th fault handling action to be corrected.

2. The method for processing in-airport fault data during pilot training based on a data middle platform according to claim 1, wherein Adjusting the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action includes: Adjust the neural network parameters of the preset neural network according to the difference between the predicted processing action and the standard processing action corresponding to the predicted processing action; When the maximum iteration parameter is reached or the preset training requirement is met, save the latest neural network parameters, and use the preset neural network under the latest neural network parameters as the adjusted preset neural network.

3. The method for processing in-airport fault data during pilot training based on a data middle platform according to claim 1, wherein, Judging whether the to-be-corrected fault processing action is qualified includes: When the difference between the target predicted processing action and the to-be-corrected fault processing action is greater than or equal to the difference threshold, the to-be-corrected fault processing action is unqualified; When the difference between the target predicted processing action and the to-be-corrected fault processing action is less than the difference threshold, the to-be-corrected fault processing action is qualified.

4. The method for processing in-airport fault data during pilot training based on a data middle platform according to claim 1, wherein, The method further includes: Perform a descending order sorting according to the importance levels of a number of adjustment devices to obtain a standard sorting; When the number of target faulty devices is greater than 1, obtain the adjustment order of the first target; Judge whether the adjustment order of the first target is qualified according to the adjustment order of the first target and the standard sorting.

5. The method for processing in-airport fault data during pilot training based on a data middle platform according to claim 4, wherein Judging whether the adjustment order of the first target is qualified according to the adjustment order of the first target and the standard sorting includes: Judge whether the descending order sorting of the adjustment order of the first target can correspond to the standard sorting according to the adjustment order of the first target; If it can, the adjustment order of the first target is qualified; if not, the adjustment order of the first target is unqualified.

6. The method for processing in-airport fault data during pilot training based on a data middle platform according to claim 1, wherein Performing a missing judgment on the fault processing data set further includes: ， Among them, is the sequence vector of the fault handling data set, is the mean of the sequence vector, f is the variance of the sequence vector, and both c and d are preset weights, is the weighted density of the sequence vector of the fault handling data set.

7. An in-airport fault data processing device for pilot training based on a data middle platform, characterized in that Applied to the method for processing in-airport fault data during pilot training based on a data middle platform according to any one of claims 1-6, the device includes: A traversal module, configured to traverse a plurality of adjustment devices on a target training flight device, and respectively set a plurality of outlier values for each adjustment device, wherein when the adjustment device is under the outlier value corresponding to the adjustment device, the adjustment device is in a fault state; A binding module, configured to, for each adjustment device, bind the adjustment device, an outlier value corresponding to the adjustment device, and a standard processing action of the outlier value under the adjustment device, to obtain a fault processing data set corresponding to each adjustment device, and each standard processing action includes a processing time limit, an adjustment target, and a body posture amplitude; An adjustment module, configured to input the fault processing data sets corresponding to a plurality of adjustment devices into a preset neural network, and adjust the preset neural network according to the predicted processing action output by the preset neural network and the standard processing action corresponding to the predicted processing action; A prediction module, configured to, when the target flight device is in a training state, input the target outlier value of the target fault device into the adjusted preset neural network, and obtain a target predicted processing action corresponding to the target outlier value of the target fault device; A judgment module, configured to obtain a to-be-corrected fault processing action of a first target, and judge whether the to-be-corrected fault processing action is qualified according to the difference between the target predicted processing action and the to-be-corrected fault processing action.

8. An electronic device, characterized in that, Includes: A processor; A memory for storing executable instructions of the processor; Wherein, the processor is configured to execute to implement the method for processing in-airport fault data during pilot training based on a data middle platform according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device can execute to implement the method for processing in-airport fault data during pilot training based on a data middle platform according to any one of claims 1 to 6.

Images