Aircraft fault diagnosis method, device and equipment and storage medium
By building and training fault diagnosis models, reading and converting aircraft data, intelligent diagnosis of aircraft faults is realized, and the problems of insufficient intelligence and poor adaptability of fault diagnosis in the existing technology are solved, and the speed and efficiency of fault repair are improved. .
Patent Information
- Application Number
- CN202411445863.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-16
- Publication Date
- 2025-02-18
AI Technical Summary
The existing aircraft fault diagnosis methods are not intelligent enough, have poor adaptability, and cannot accurately judge and repair the aircraft, and have low accuracy and maintenance efficiency.
By building an initial fault diagnosis model, training the model based on the training data and prediction data, obtaining the target fault diagnosis model; reading the aircraft data, converting the data, and obtaining the target fault data; inputting the target fault data into the target fault diagnosis model, Obtain fault diagnosis results and determine the diagnostic strategy implementation based on the results.
It realizes intelligent, fast and automatic evaluation of aircraft failures, accurately determines the location and type of faults, improves the speed and efficiency of aircraft failure repair, and turns to predictive and accurate maintenance.
Smart Images
Figure CN119475150A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of aircraft fault diagnosis, and in particular to an aircraft fault diagnosis method, device, equipment and storage medium. Background Art
[0002] With the rapid development of artificial intelligence technology, aircraft fault diagnosis methods have gradually shifted from relying mainly on manual experience to automation and intelligence.
[0003] At present, common aircraft fault diagnosis methods mainly include: model-based aircraft fault diagnosis method and fault tree analysis-based aircraft fault diagnosis method.
[0004] The model-based aircraft fault diagnosis method mainly builds an aircraft dynamics model, uses sensors to estimate the state, compares it with the actual measurement value, and analyzes the residual to detect the fault;
[0005] The aircraft fault diagnosis method based on fault tree analysis analyzes various possible causes of faults from top to bottom and constructs a fault tree for qualitative and quantitative analysis of faults.
[0006] However, these two methods are not intelligent enough in fault diagnosis and have poor adaptability. They are unable to accurately diagnose and repair aircraft faults, and their accuracy and maintenance efficiency are low. Summary of the invention
[0007] The main purpose of the present invention is to provide an aircraft fault diagnosis method, device, equipment and storage medium, aiming to solve the technical problems of the aircraft fault diagnosis method in the prior art, such as insufficient fault diagnosis intelligence, poor adaptability, inability to accurately judge and repair aircraft faults, and low accuracy and maintenance efficiency.
[0008] In a first aspect, the present invention provides an aircraft fault diagnosis method, the aircraft fault diagnosis method comprising the following steps:
[0009] Building an initial fault diagnosis model, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model;
[0010] Reading aircraft data, acquiring fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion;
[0011] The target fault data is input into a target fault diagnosis model to obtain a fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault diagnosis result.
[0012] Optionally, the building of the initial fault diagnosis model, training the initial fault diagnosis model according to the training data and the prediction data to obtain a trained target fault diagnosis model, includes:
[0013] Obtaining the first connection weight between the neurons of the input layer and the neurons of the hidden layer;
[0014] Obtaining a second connection weight between the neurons in the intermediate layer and the neurons in the output layer;
[0015] Obtaining the number of neurons in the input layer, the hidden layer, and the output layer, and building an initial fault diagnosis model according to the first connection weight, the second connection weight, and the number of neurons;
[0016] The training data and the prediction data are obtained, and the initial fault diagnosis model is trained according to the training data and the prediction data to obtain a trained target fault diagnosis model.
[0017] Optionally, the acquiring of training data and prediction data, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model includes:
[0018] Obtain training data and prediction data corresponding to the current aircraft model from a preset data set;
[0019] Normalizing the training data to obtain normalized target training data, constructing a three-layer BP neural network model, and obtaining corresponding network parameters;
[0020] The initial fault diagnosis model is trained and predicted using the BP neural network model according to the network parameters, the target training data and the predicted data, until the training is completed when the error between the true value and the predicted value of the verification set is within a preset allowable range, thereby obtaining a trained target fault diagnosis model.
[0021] Optionally, the reading of the aircraft data, obtaining fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion includes:
[0022] Acquire aircraft data of the aircraft to be diagnosed;
[0023] Extracting the aircraft data to obtain corresponding fault data;
[0024] The data format of the fault data is converted into a format required by the target fault diagnosis model to obtain the target fault data after the data conversion.
[0025] Optionally, inputting the target fault data into a target fault diagnosis model to obtain a fault diagnosis result, and determining and executing a corresponding diagnosis strategy according to the fault diagnosis result, comprises:
[0026] Inputting the target fault data into a target fault diagnosis model to obtain a fault diagnosis result;
[0027] The fault type and fault location of the target fault are obtained from the fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault type and the fault location.
[0028] Optionally, obtaining a fault type and a fault location of a target fault from the fault diagnosis result, and determining and executing a corresponding diagnosis strategy according to the fault type and the fault location, includes:
[0029] Obtaining the fault type and fault location of the target fault from the fault diagnosis result;
[0030] estimating long-term performance trend data within a preset time in the future according to the fault type and the fault location;
[0031] Determine the performance deviation time according to the long-term performance trend data, and use the preset parameter threshold to extrapolate the performance deviation time to obtain the remaining estimated life of the single machine;
[0032] A corresponding diagnostic strategy is determined and executed according to the fault type, the fault location and the estimated remaining life of the single machine.
[0033] Optionally, determining and executing a corresponding diagnosis strategy according to the fault type, the fault location and the estimated remaining life of the single machine includes:
[0034] Obtaining a matching diagnostic strategy from a preset diagnostic strategy library according to the fault type, the fault location and the estimated remaining life of the single machine;
[0035] The diagnostic strategy is executed and a prompt is given, and after the faulty aircraft returns to the maintenance area, the faulty aircraft is replaced or maintained.
[0036] In a second aspect, in order to achieve the above-mentioned object, the present invention further provides an aircraft fault diagnosis device, the aircraft fault diagnosis device comprising:
[0037] A training module is used to build an initial fault diagnosis model, train the initial fault diagnosis model according to the training data and the prediction data, and obtain a trained target fault diagnosis model;
[0038] A data conversion module, used for reading aircraft data, obtaining fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion;
[0039] The diagnosis module is used to input the target fault data into a target fault diagnosis model to obtain a fault diagnosis result, determine a corresponding diagnosis strategy according to the fault diagnosis result, and execute it.
[0040] In a third aspect, in order to achieve the above-mentioned purpose, the present invention also proposes an aircraft fault diagnosis device, which includes: a memory, a processor, and an aircraft fault diagnosis program stored in the memory and executable on the processor, wherein the aircraft fault diagnosis program is configured to implement the steps of the aircraft fault diagnosis method as described above.
[0041] In a fourth aspect, in order to achieve the above-mentioned purpose, the present invention further proposes a storage medium, on which an aircraft fault diagnosis program is stored, and when the aircraft fault diagnosis program is executed by a processor, the steps of the aircraft fault diagnosis method as described above are implemented.
[0042] The aircraft fault diagnosis method proposed in the present invention builds an initial fault diagnosis model, trains the initial fault diagnosis model according to training data and prediction data, and obtains a trained target fault diagnosis model; reads aircraft data, obtains fault data from the aircraft data, performs data conversion on the fault data, and obtains target fault data after data conversion; inputs the target fault data into the target fault diagnosis model to obtain a fault diagnosis result, determines and executes a corresponding diagnosis strategy according to the fault diagnosis result, and can form an intelligent fault diagnosis model by training a large amount of historical data to perform fault diagnosis with higher intelligence and adaptability, automatically, quickly, and intelligently evaluate aircraft fault conditions, judge the location and type of aircraft faults, and enable the aircraft to shift from corrective and preventive maintenance to predictive and precise maintenance, thereby improving the speed and efficiency of aircraft fault maintenance. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 A schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention;
[0044] Figure 2 It is a flowchart of a first embodiment of an aircraft fault diagnosis method of the present invention;
[0045] Figure 3 It is a flowchart of a second embodiment of the aircraft fault diagnosis method of the present invention;
[0046] Figure 4Schematic diagram of the structure of the fault diagnosis model in the aircraft fault diagnosis method of the present invention;
[0047] Figure 5 It is a flowchart of a third embodiment of the aircraft fault diagnosis method of the present invention;
[0048] Figure 6 Schematic diagram of the fault diagnosis model training steps in the aircraft fault diagnosis method of the present invention;
[0049] Figure 7 It is a flowchart of a fourth embodiment of the aircraft fault diagnosis method of the present invention;
[0050] Figure 8 It is a schematic diagram of the execution process of the fault diagnosis algorithm in the aircraft fault diagnosis method of the present invention;
[0051] Fig. 9 It is a flowchart of a fifth embodiment of the aircraft fault diagnosis method of the present invention;
[0052] Fig.10 A schematic diagram of a fault diagnosis result in the aircraft fault diagnosis method of the present invention;
[0053] Fig.11 It is a functional module diagram of the first embodiment of the aircraft fault diagnosis device of the present invention.
[0054] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0055] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0056] The solution of the embodiment of the present invention is mainly as follows: by building an initial fault diagnosis model, the initial fault diagnosis model is trained according to the training data and the prediction data to obtain a trained target fault diagnosis model; aircraft data is read, fault data is obtained from the aircraft data, the fault data is converted to obtain target fault data after data conversion; the target fault data is input into the target fault diagnosis model to obtain a fault diagnosis result, and the corresponding diagnosis strategy is determined and executed according to the fault diagnosis result. By training a large amount of historical data, an intelligent fault diagnosis model can be formed to perform fault diagnosis, with a higher degree of intelligence and adaptability, and the aircraft fault condition can be automatically, quickly and intelligently evaluated, and the location and type of the aircraft fault can be determined, so that the aircraft can be transformed from corrective and preventive maintenance to predictive and accurate maintenance, thereby improving the speed and efficiency of aircraft fault maintenance, and solving the technical problems of the aircraft fault diagnosis method in the prior art, that is, the fault diagnosis is not intelligent enough, the adaptability is poor, the aircraft fault cannot be accurately judged and repaired, and the accuracy and maintenance efficiency are low.
[0057] Reference Figure 1 , Figure 1 The figure is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.
[0058] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed RAM memory, or it may be a stable memory (Non-Volatile Memory), such as a disk memory. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.
[0059] Those skilled in the art will understand that Figure 1 The device structure shown in the figure does not constitute a limitation of the device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0060] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating device, a network communication module, a user interface module, and an aircraft fault diagnosis program.
[0061] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and performs the following operations:
[0062] Building an initial fault diagnosis model, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model;
[0063] Reading aircraft data, acquiring fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion;
[0064] The target fault data is input into a target fault diagnosis model to obtain a fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault diagnosis result.
[0065] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and also performs the following operations:
[0066] Obtaining the first connection weight between the neurons of the input layer and the neurons of the hidden layer;
[0067] Obtaining a second connection weight between the neurons in the intermediate layer and the neurons in the output layer;
[0068] Obtaining the number of neurons in the input layer, the hidden layer, and the output layer, and building an initial fault diagnosis model according to the first connection weight, the second connection weight, and the number of neurons;
[0069] The training data and the prediction data are obtained, and the initial fault diagnosis model is trained according to the training data and the prediction data to obtain a trained target fault diagnosis model.
[0070] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and also performs the following operations:
[0071] Obtain training data and prediction data corresponding to the current aircraft model from a preset data set;
[0072] Normalizing the training data to obtain normalized target training data, constructing a three-layer BP neural network model, and obtaining corresponding network parameters;
[0073] The initial fault diagnosis model is trained and predicted using the BP neural network model according to the network parameters, the target training data and the predicted data, until the training is completed when the error between the true value and the predicted value of the verification set is within a preset allowable range, thereby obtaining a trained target fault diagnosis model.
[0074] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and also performs the following operations:
[0075] Acquire aircraft data of the aircraft to be diagnosed;
[0076] Extracting the aircraft data to obtain corresponding fault data;
[0077] The data format of the fault data is converted into a format required by the target fault diagnosis model to obtain the target fault data after the data conversion.
[0078] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and also performs the following operations:
[0079] Inputting the target fault data into a target fault diagnosis model to obtain a fault diagnosis result;
[0080] The fault type and fault location of the target fault are obtained from the fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault type and the fault location.
[0081] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and also performs the following operations:
[0082] Obtaining the fault type and fault location of the target fault from the fault diagnosis result;
[0083] estimating long-term performance trend data within a preset time in the future according to the fault type and the fault location;
[0084] Determine the performance deviation time according to the long-term performance trend data, and use the preset parameter threshold to extrapolate the performance deviation time to obtain the remaining estimated life of the single machine;
[0085] A corresponding diagnostic strategy is determined and executed according to the fault type, the fault location and the estimated remaining life of the single machine.
[0086] The device of the present invention calls the aircraft fault diagnosis program stored in the memory 1005 through the processor 1001, and also performs the following operations:
[0087] Obtaining a matching diagnostic strategy from a preset diagnostic strategy library according to the fault type, the fault location and the estimated remaining life of the single machine;
[0088] The diagnostic strategy is executed and a prompt is given, and after the faulty aircraft returns to the maintenance area, the faulty aircraft is replaced or maintained.
[0089] This embodiment adopts the above scheme, by building an initial fault diagnosis model, training the initial fault diagnosis model according to training data and prediction data, and obtaining a trained target fault diagnosis model; reading aircraft data, obtaining fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion; inputting the target fault data into the target fault diagnosis model to obtain a fault diagnosis result, determining and executing a corresponding diagnosis strategy according to the fault diagnosis result, and being able to train a large amount of historical data to form an intelligent fault diagnosis model to perform fault diagnosis with a higher degree of intelligence and adaptability, automatically, quickly, and intelligently assessing aircraft fault conditions, determining the location and type of aircraft faults, and enabling aircraft to shift from corrective and preventive maintenance to predictive and precise maintenance, thereby improving the speed and efficiency of aircraft fault maintenance.
[0090] Based on the above hardware structure, an embodiment of the aircraft fault diagnosis method of the present invention is proposed.
[0091] Reference Figure 2 , Figure 2 It is a flowchart of the first embodiment of the aircraft fault diagnosis method of the present invention.
[0092] In a first embodiment, the aircraft fault diagnosis method comprises the following steps:
[0093] Step S10: construct an initial fault diagnosis model, train the initial fault diagnosis model according to the training data and the prediction data, and obtain a trained target fault diagnosis model.
[0094] It should be noted that before performing aircraft fault diagnosis, it is necessary to construct a fault diagnosis model and build an initial fault diagnosis model. The initial fault diagnosis model can be trained through pre-set training data and prediction data to obtain a trained target fault diagnosis model.
[0095] Step S20: read the aircraft data, obtain fault data from the aircraft data, perform data conversion on the fault data, and obtain target fault data after data conversion.
[0096] It should be understood that after reading the aircraft data, corresponding fault data can be obtained from the aircraft data, and after performing data format conversion on the fault data, target fault data after data conversion can be obtained.
[0097] Step S30: input the target fault data into a target fault diagnosis model to obtain a fault diagnosis result, and determine and execute a corresponding diagnosis strategy according to the fault diagnosis result.
[0098] It can be understood that by inputting the target fault data into the target fault diagnosis model, a fault diagnosis result can be obtained, and then a corresponding diagnosis strategy can be determined and executed according to the fault diagnosis result.
[0099] This embodiment adopts the above scheme, by building an initial fault diagnosis model, training the initial fault diagnosis model according to training data and prediction data, and obtaining a trained target fault diagnosis model; reading aircraft data, obtaining fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion; inputting the target fault data into the target fault diagnosis model to obtain a fault diagnosis result, determining and executing a corresponding diagnosis strategy according to the fault diagnosis result, and being able to train a large amount of historical data to form an intelligent fault diagnosis model to perform fault diagnosis with a higher degree of intelligence and adaptability, automatically, quickly, and intelligently assessing aircraft fault conditions, determining the location and type of aircraft faults, and enabling aircraft to shift from corrective and preventive maintenance to predictive and precise maintenance, thereby improving the speed and efficiency of aircraft fault maintenance.
[0100] Further, Figure 3 FIG. 1 is a flow chart of a second embodiment of the aircraft fault diagnosis method of the present invention. Figure 3 As shown, a second embodiment of the aircraft fault diagnosis method of the present invention is proposed based on the first embodiment. In this embodiment, the step S10 specifically includes the following steps:
[0101] Step S11, obtaining the first connection weight between the neurons of the input layer and the neurons of the hidden layer.
[0102] It should be noted that the fault diagnosis model of this embodiment is a three-layer deep learning neural network model. First, the connection weights between the neurons in the input layer and the neurons in the hidden layer, that is, the first connection weights, can be obtained.
[0103] Step S12: Obtain a second connection weight between the neurons in the intermediate layer and the neurons in the output layer.
[0104] It can be understood that the connection weight between the neurons in the intermediate layer and the neurons in the output layer is obtained, that is, the second connection weight.
[0105] Step S13, obtaining the number of neurons in the input layer, the hidden layer and the output layer, and building an initial fault diagnosis model according to the first connection weight, the second connection weight and the number of neurons.
[0106] It should be understood that after obtaining the number of neurons corresponding to the neurons of the input layer, the neurons of the hidden layer and the neurons of the output layer, an initial fault diagnosis model can be built according to the first connection weight, the second connection weight and the number of neurons.
[0107] Step S14: acquiring training data and prediction data, and training the initial fault diagnosis model according to the training data and the prediction data to obtain a trained target fault diagnosis model.
[0108] It is understandable that after obtaining the training data and the prediction data, the initial fault diagnosis model can be trained according to the training data and the prediction data to obtain a trained target fault diagnosis model.
[0109] In the specific implementation, refer to Figure 4 , Figure 4 FIG. 1 is a schematic diagram of the structure of a fault diagnosis model in the aircraft fault diagnosis method of the present invention. Figure 4 As shown, the fault diagnosis model used in the present invention is a 3-layer neural network model. ij is the connection weight between the i-th neuron in the input layer and the j-th neuron in the hidden layer, w jk is the connection weight between the jth neuron in the middle layer and the kth neuron in the output layer, x pi is the input of the ith neuron in the input layer, c pk is the output of the kth neuron in the output layer, i = 1, 2, ..., n; j = 1, 2, ..., l; k = 1, 2, ..., m; n, l, m are the number of neurons in the input layer, hidden layer and output layer respectively; p = 1, 2 ..., P; P is the total number of samples.
[0110] This embodiment adopts the above scheme, by obtaining the first connection weight between the neurons of the input layer and the neurons of the hidden layer; obtaining the second connection weight between the neurons of the middle layer and the neurons of the output layer; obtaining the number of neurons in the input layer, the hidden layer and the output layer, and building an initial fault diagnosis model according to the first connection weight, the second connection weight and the number of neurons; obtaining training data and prediction data, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model, which can quickly build an intelligent fault diagnosis model to perform fault diagnosis, with a higher degree of intelligence and adaptability, automatically, quickly and intelligently evaluate aircraft fault conditions, and improve the speed of aircraft fault diagnosis.
[0111] Further, Figure 5 FIG. 1 is a flow chart of a third embodiment of the aircraft fault diagnosis method of the present invention. Figure 5As shown, a third embodiment of the aircraft fault diagnosis method of the present invention is proposed based on the second embodiment. In this embodiment, the step S14 specifically includes the following steps:
[0112] Step S141: Obtain training data and prediction data corresponding to the current aircraft model from a preset data set.
[0113] It should be noted that the training data and prediction data corresponding to the aircraft model of the current aircraft can be obtained from the preset data set, and different types of aircraft correspond to different training data and prediction data.
[0114] Step S142: normalize the training data to obtain normalized target training data, construct a three-layer BP neural network model, and obtain corresponding network parameters.
[0115] It can be understood that after the training data is normalized, normalized target training data can be obtained, and a three-layer structured (Back Propagation, BP) neural network model can be constructed, and corresponding network parameters can be obtained.
[0116] Step S143, training and predicting the initial fault diagnosis model using the BP neural network model according to the network parameters, the target training data and the predicted data, until the training is completed when the error between the true value and the predicted value of the verification set is within a preset allowable range, thereby obtaining a trained target fault diagnosis model.
[0117] It should be understood that the initial fault diagnosis model is trained and predicted using the BP neural network model based on the network parameters, the target training data and the predicted data, and the training is completed when it can be verified that the error between the true value and the predicted value of the verification set corresponding to the initial fault diagnosis model is within a preset allowable range, thereby obtaining a trained target fault diagnosis model.
[0118] In the specific implementation, see Figure 6 , Figure 6 FIG. 1 is a schematic diagram of the steps of training the fault diagnosis model in the aircraft fault diagnosis method of the present invention. Figure 6 As shown in the figure, after building the fault diagnosis model, it needs to be trained. First, set the training data and prediction data, then normalize the training data, then build a 3-layer BP neural network model and set the network parameters, and then use the BP neural network model for training and prediction. Repeat the above steps until the error between the true value and the predicted value of the verification set is within the operating range, and the training is completed.
[0119] This embodiment adopts the above scheme, obtains training data and prediction data corresponding to the current aircraft model from a preset data set; normalizes the training data to obtain normalized target training data, constructs a three-layer BP neural network model, and obtains corresponding network parameters; uses the BP neural network model to train and predict the initial fault diagnosis model according to the network parameters, the target training data and the prediction data, until the training is completed when the error between the true value and the prediction value of the verification set is within a preset allowable range, and obtains a trained target fault diagnosis model. The trained target fault diagnosis model can be obtained to perform fault diagnosis, which has a higher degree of intelligence and adaptability, and can automatically, quickly and intelligently evaluate aircraft fault conditions, thereby improving the speed of aircraft fault diagnosis.
[0120] Further, Figure 7 FIG. 4 is a flow chart of a fourth embodiment of the aircraft fault diagnosis method of the present invention. Figure 7 As shown, a fourth embodiment of the aircraft fault diagnosis method of the present invention is proposed based on the first embodiment. In this embodiment, step S20 specifically includes the following steps:
[0121] Step S21: Acquire aircraft data of the aircraft to be diagnosed.
[0122] It should be noted that firstly, the aircraft data of the aircraft to be diagnosed can be obtained.
[0123] Step S22: extract the aircraft data to obtain corresponding fault data.
[0124] It can be understood that after data extraction is performed on the aircraft data, corresponding fault data can be obtained.
[0125] Step S23: convert the data format of the fault data into the format required by the target fault diagnosis model to obtain the target fault data after data conversion.
[0126] It should be understood that the data format of the fault data is converted into the format required by the target fault diagnosis model, thereby obtaining the target fault data after data conversion.
[0127] In the specific implementation, see Figure 8 , Figure 8 FIG. 1 is a schematic diagram of the execution process of the fault diagnosis algorithm in the aircraft fault diagnosis method of the present invention. Figure 8As shown, after the fault diagnosis model is trained, it can be used to complete aircraft fault diagnosis; first, load the fault diagnosis model, as well as input data such as historical test data, fault information, and monitoring results; then perform data extraction, extract the fault diagnosis data from the input data, and perform data conversion on it, converting its format into the format required by the fault diagnosis model; then load the fault diagnosis data in the standard format into the fault diagnosis model, and perform fault diagnosis through the fault diagnosis model execution engine.
[0128] This embodiment adopts the above scheme, by acquiring the aircraft data of the current aircraft to be diagnosed; performing data extraction on the aircraft data to obtain corresponding fault data; converting the data format of the fault data into the format required by the target fault diagnosis model to obtain the target fault data after data conversion; and being able to quickly obtain the fault data after data conversion.
[0129] Further, Fig. 9 FIG. 5 is a flow chart of a fifth embodiment of the aircraft fault diagnosis method of the present invention. Fig. 9 As shown, a fifth embodiment of the aircraft fault diagnosis method of the present invention is proposed based on the first embodiment. In this embodiment, step S30 specifically includes the following steps:
[0130] Step S31: input the target fault data into a target fault diagnosis model to obtain a fault diagnosis result.
[0131] It should be noted that by inputting the target fault data into the target fault diagnosis model, a fault diagnosis result can be obtained.
[0132] In specific implementations, fault diagnosis requires data from many sources, and the data in each system is complex. To extract the data required for fault diagnosis from numerous business information systems, it is necessary to perform data extraction, conversion, loading and other steps, and then form a standardized data set that meets the diagnostic standards.
[0133] Step S32: Obtain the fault type and fault location of the target fault from the fault diagnosis result, determine and execute a corresponding diagnosis strategy according to the fault type and the fault location.
[0134] It can be understood that the fault type and fault location of the target fault are obtained from the fault diagnosis result, and then the corresponding diagnosis strategy can be determined and executed according to the fault type and the fault location.
[0135] In the specific implementation, the fault diagnosis algorithm is executed, and the generated diagnostic conclusions are handed over to the diagnostic result output part to be stored in the database. The fault diagnosis query and analysis is mainly based on the generated diagnostic conclusions to realize the query and analysis of the detailed parameters of the fault diagnosis of key components (such as servo structures and steering gears), and to have a detailed understanding of the health performance parameters of key components, from which the information required by the user is obtained. According to the user configuration information and the results of the failure diagnosis of key components, the corresponding fault diagnosis analysis report is generated.
[0136] Furthermore, the step S32 specifically includes the following steps:
[0137] Obtaining the fault type and fault location of the target fault from the fault diagnosis result;
[0138] estimating long-term performance trend data within a preset time in the future according to the fault type and the fault location;
[0139] Determine the performance deviation time according to the long-term performance trend data, and use the preset parameter threshold to extrapolate the performance deviation time to obtain the remaining estimated life of the single machine;
[0140] A corresponding diagnostic strategy is determined and executed according to the fault type, the fault location and the estimated remaining life of the single machine.
[0141] It should be understood that long-term performance trend data within a preset future time period can be estimated based on the fault type and fault location; the performance excess time can be determined based on the long-term performance trend data, and then the performance excess time can be extrapolated using the preset parameter threshold to obtain the remaining estimated life of the single machine; thus, the corresponding diagnostic strategy can be determined and executed based on the fault type, the fault location and the remaining estimated life of the single machine.
[0142] Furthermore, the step determines and executes a corresponding diagnostic strategy according to the fault type, the fault location and the estimated remaining life of the single machine, and specifically includes the following steps:
[0143] Obtaining a matching diagnostic strategy from a preset diagnostic strategy library according to the fault type, the fault location and the estimated remaining life of the single machine;
[0144] The diagnostic strategy is executed and a prompt is given, and after the faulty aircraft returns to the maintenance area, the faulty aircraft is replaced or maintained.
[0145] It can be understood that a matching diagnostic strategy is obtained from a preset diagnostic strategy library based on the fault type, the fault location and the estimated remaining life of the single aircraft. After executing the diagnostic strategy, corresponding prompts can be given for feedback. After detecting that the faulty aircraft has returned to the maintenance area, the faulty aircraft can be replaced or maintained.
[0146] In the specific implementation, see Fig.10 , Fig.10 FIG. 1 is a schematic diagram of the fault diagnosis result of the aircraft fault diagnosis method of the present invention, as shown in FIG. Fig.10 As shown, the fault diagnosis method can calculate the fault type and location, and output long-term performance trend data and curves for a period of time in the future. It can also use parameter thresholds to extrapolate the performance deviation time, estimate the remaining life of a single machine, and prompt the site and the warehouse to replace and maintain the equipment after returning.
[0147] Through the above scheme, this embodiment obtains a fault diagnosis result by inputting the target fault data into a target fault diagnosis model; obtains the fault type and fault location of the target fault from the fault diagnosis result, determines and executes a corresponding diagnosis strategy based on the fault type and the fault location; can train a large amount of historical data to form an intelligent fault diagnosis model to perform fault diagnosis with higher intelligence and adaptability, automatically, quickly and intelligently evaluate aircraft fault conditions, determine the location and type of aircraft faults, and enable the aircraft to shift from corrective and preventive maintenance to predictive and precise maintenance, thereby improving the speed and efficiency of aircraft fault maintenance.
[0148] Accordingly, the present invention further provides an aircraft fault diagnosis device.
[0149] Reference Fig.11 , Fig.11 It is a functional module diagram of the first embodiment of the aircraft fault diagnosis device of the present invention.
[0150] In a first embodiment of the aircraft fault diagnosis device of the present invention, the aircraft fault diagnosis device comprises:
[0151] The training module 10 is used to build an initial fault diagnosis model, train the initial fault diagnosis model according to the training data and the prediction data, and obtain a trained target fault diagnosis model.
[0152] The data conversion module 20 is used to read the aircraft data, obtain fault data from the aircraft data, perform data conversion on the fault data, and obtain target fault data after data conversion.
[0153] The diagnosis module 30 is used to input the target fault data into a target fault diagnosis model to obtain a fault diagnosis result, determine a corresponding diagnosis strategy according to the fault diagnosis result, and execute the strategy.
[0154] The training module 10 is also used to obtain a first connection weight between neurons in the input layer and neurons in the hidden layer; obtain a second connection weight between neurons in the middle layer and neurons in the output layer; obtain the number of neurons in the input layer, the hidden layer and the output layer, and build an initial fault diagnosis model based on the first connection weight, the second connection weight and the number of neurons; obtain training data and prediction data, train the initial fault diagnosis model based on the training data and the prediction data, and obtain a trained target fault diagnosis model.
[0155] The training module 10 is also used to obtain training data and prediction data corresponding to the current aircraft model from a preset data set; normalize the training data to obtain normalized target training data, construct a three-layer BP neural network model, and obtain corresponding network parameters; use the BP neural network model to train and predict the initial fault diagnosis model according to the network parameters, the target training data and the prediction data, until the training is completed when the error between the true value and the predicted value of the verification set is within a preset allowable range, and a trained target fault diagnosis model is obtained.
[0156] The data conversion module 20 is also used to obtain aircraft data of the aircraft to be diagnosed; extract the aircraft data to obtain corresponding fault data; convert the data format of the fault data into the format required by the target fault diagnosis model to obtain the target fault data after data conversion.
[0157] The diagnostic module 30 is also used to input the target fault data into a target fault diagnosis model to obtain a fault diagnosis result; obtain the fault type and fault location of the target fault from the fault diagnosis result, and determine and execute a corresponding diagnostic strategy based on the fault type and the fault location.
[0158] The diagnostic module 30 is also used to obtain the fault type and fault location of the target fault from the fault diagnosis result; estimate the long-term performance trend data within a preset future time based on the fault type and the fault location; determine the performance excess time based on the long-term performance trend data, and use the preset parameter threshold to extrapolate the performance excess time to obtain the remaining estimated life of the single machine; determine and execute the corresponding diagnostic strategy based on the fault type, the fault location and the remaining estimated life of the single machine.
[0159] The diagnostic module 30 is also used to obtain a matching diagnostic strategy from a preset diagnostic strategy library according to the fault type, the fault location and the estimated remaining life of the single aircraft; execute the diagnostic strategy and give a prompt, and replace or maintain the faulty aircraft after the faulty aircraft returns to the maintenance area.
[0160] Among them, the steps implemented by each functional module of the aircraft fault diagnosis device can refer to the various embodiments of the aircraft fault diagnosis method of the present invention, and will not be repeated here.
[0161] In addition, an embodiment of the present invention further provides a storage medium, wherein an aircraft fault diagnosis program is stored on the storage medium, and when the aircraft fault diagnosis program is executed by a processor, the following operations are implemented:
[0162] Building an initial fault diagnosis model, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model;
[0163] Reading aircraft data, acquiring fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion;
[0164] The target fault data is input into a target fault diagnosis model to obtain a fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault diagnosis result.
[0165] Furthermore, when the aircraft fault diagnosis program is executed by the processor, the following operations are also implemented:
[0166] Obtaining the first connection weight between the neurons of the input layer and the neurons of the hidden layer;
[0167] Obtaining a second connection weight between the neurons in the intermediate layer and the neurons in the output layer;
[0168] Obtaining the number of neurons in the input layer, the hidden layer, and the output layer, and building an initial fault diagnosis model according to the first connection weight, the second connection weight, and the number of neurons;
[0169] The training data and the prediction data are obtained, and the initial fault diagnosis model is trained according to the training data and the prediction data to obtain a trained target fault diagnosis model.
[0170] Furthermore, when the aircraft fault diagnosis program is executed by the processor, the following operations are also implemented:
[0171] Obtain training data and prediction data corresponding to the current aircraft model from a preset data set;
[0172] Normalizing the training data to obtain normalized target training data, constructing a three-layer BP neural network model, and obtaining corresponding network parameters;
[0173] The initial fault diagnosis model is trained and predicted using the BP neural network model according to the network parameters, the target training data and the predicted data, until the training is completed when the error between the true value and the predicted value of the verification set is within a preset allowable range, thereby obtaining a trained target fault diagnosis model.
[0174] Furthermore, when the aircraft fault diagnosis program is executed by the processor, the following operations are also implemented:
[0175] Acquire aircraft data of the aircraft to be diagnosed;
[0176] Extracting the aircraft data to obtain corresponding fault data;
[0177] The data format of the fault data is converted into a format required by the target fault diagnosis model to obtain the target fault data after the data conversion.
[0178] Furthermore, when the aircraft fault diagnosis program is executed by the processor, the following operations are also implemented:
[0179] Inputting the target fault data into a target fault diagnosis model to obtain a fault diagnosis result;
[0180] The fault type and fault location of the target fault are obtained from the fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault type and the fault location.
[0181] Furthermore, when the aircraft fault diagnosis program is executed by the processor, the following operations are also implemented:
[0182] Obtaining the fault type and fault location of the target fault from the fault diagnosis result;
[0183] estimating long-term performance trend data within a preset time in the future according to the fault type and the fault location;
[0184] Determine the performance deviation time according to the long-term performance trend data, and use the preset parameter threshold to extrapolate the performance deviation time to obtain the remaining estimated life of the single machine;
[0185] A corresponding diagnostic strategy is determined and executed according to the fault type, the fault location and the estimated remaining life of the single machine.
[0186] Furthermore, when the aircraft fault diagnosis program is executed by the processor, the following operations are also implemented:
[0187] Obtaining a matching diagnostic strategy from a preset diagnostic strategy library according to the fault type, the fault location and the estimated remaining life of the single machine;
[0188] The diagnostic strategy is executed and a prompt is given, and after the faulty aircraft returns to the maintenance area, the faulty aircraft is replaced or maintained.
[0189] Those skilled in the art will appreciate that all or part of the steps in the above-mentioned implementation method can be accomplished by instructing related hardware through a program, and the program is stored in a storage medium, including a number of instructions for enabling a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each implementation method of the present application; and the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.
[0190] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0191] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0192] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. An aircraft fault diagnosis method, characterized in that: The aircraft fault diagnosis method comprises: Building an initial fault diagnosis model, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model; Reading aircraft data, acquiring fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion; The target fault data is input into a target fault diagnosis model to obtain a fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault diagnosis result.
2. The aircraft fault diagnosis method according to claim 1, characterized in that: The initial fault diagnosis model is constructed, and the initial fault diagnosis model is trained according to the training data and the prediction data to obtain a trained target fault diagnosis model, including: Obtaining the first connection weight between the neurons of the input layer and the neurons of the hidden layer; Obtaining a second connection weight between the neurons in the intermediate layer and the neurons in the output layer; Obtaining the number of neurons in the input layer, the hidden layer, and the output layer, and building an initial fault diagnosis model according to the first connection weight, the second connection weight, and the number of neurons; The training data and the prediction data are obtained, and the initial fault diagnosis model is trained according to the training data and the prediction data to obtain a trained target fault diagnosis model.
3. The aircraft fault diagnosis method according to claim 2, characterized in that: The acquiring of training data and prediction data, training the initial fault diagnosis model according to the training data and the prediction data, and obtaining a trained target fault diagnosis model comprises: Obtain training data and prediction data corresponding to the current aircraft model from a preset data set; Normalizing the training data to obtain normalized target training data, constructing a three-layer BP neural network model, and obtaining corresponding network parameters; The initial fault diagnosis model is trained and predicted using the BP neural network model according to the network parameters, the target training data and the predicted data, until the training is completed when the error between the true value and the predicted value of the verification set is within a preset allowable range, thereby obtaining a trained target fault diagnosis model.
4. The aircraft fault diagnosis method according to claim 1, characterized in that: The reading of the aircraft data, obtaining fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion includes: Acquire aircraft data of the aircraft to be diagnosed; Extracting the aircraft data to obtain corresponding fault data; The data format of the fault data is converted into a format required by the target fault diagnosis model to obtain the target fault data after the data conversion.
5. The aircraft fault diagnosis method according to claim 1, characterized in that: The step of inputting the target fault data into a target fault diagnosis model to obtain a fault diagnosis result, and determining and executing a corresponding diagnosis strategy according to the fault diagnosis result includes: Inputting the target fault data into a target fault diagnosis model to obtain a fault diagnosis result; The fault type and fault location of the target fault are obtained from the fault diagnosis result, and a corresponding diagnosis strategy is determined and executed according to the fault type and the fault location.
6. The aircraft fault diagnosis method according to claim 5, characterized in that: The obtaining of the fault type and the fault location of the target fault from the fault diagnosis result, and determining and executing a corresponding diagnosis strategy according to the fault type and the fault location, comprises: Obtaining the fault type and fault location of the target fault from the fault diagnosis result; estimating long-term performance trend data within a preset time in the future according to the fault type and the fault location; Determine the performance deviation time according to the long-term performance trend data, and use the preset parameter threshold to extrapolate the performance deviation time to obtain the remaining estimated life of the single machine; A corresponding diagnostic strategy is determined and executed according to the fault type, the fault location and the estimated remaining life of the single machine.
7. The aircraft fault diagnosis method according to claim 6, characterized in that: The determining and executing a corresponding diagnostic strategy according to the fault type, the fault location and the estimated remaining life of the single machine includes: Obtaining a matching diagnostic strategy from a preset diagnostic strategy library according to the fault type, the fault location and the estimated remaining life of the single machine; The diagnostic strategy is executed and a prompt is given, and after the faulty aircraft returns to the maintenance area, the faulty aircraft is replaced or maintained.
8. An aircraft fault diagnosis device, characterized in that: The aircraft fault diagnosis device comprises: A training module is used to build an initial fault diagnosis model, train the initial fault diagnosis model according to the training data and the prediction data, and obtain a trained target fault diagnosis model; A data conversion module, used for reading aircraft data, obtaining fault data from the aircraft data, performing data conversion on the fault data, and obtaining target fault data after data conversion; The diagnosis module is used to input the target fault data into a target fault diagnosis model to obtain a fault diagnosis result, determine a corresponding diagnosis strategy according to the fault diagnosis result, and execute it.
9. An aircraft fault diagnosis device, characterized in that: The aircraft fault diagnosis device comprises: a memory, a processor, and an aircraft fault diagnosis program stored in the memory and executable on the processor, wherein the aircraft fault diagnosis program is configured to implement the steps of the aircraft fault diagnosis method according to any one of claims 1 to 7.
10. A storage medium, characterized in that: The storage medium stores an aircraft fault diagnosis program, which, when executed by the processor, implements the steps of the aircraft fault diagnosis method according to any one of claims 1 to 7.