Helicopter fault prediction method and device

Abstract

The present invention discloses a helicopter fault prediction method and device, wherein the method comprises: obtaining state information of a helicopter to be tested; inputting the state information of the helicopter to be tested into a helicopter fault prediction model, and outputting a fault prediction result; the helicopter fault prediction model is constructed based on a trained improved grey GM (1,1) prediction model and an Elman neural network model; the trained improved grey GM (1,1) prediction model is trained based on historical state information, historical helicopter fault prediction values, and a correction factor; the correction factor is obtained by simulating pre-acquired performance degradation information of various helicopter components; and the trained Elman neural network model is trained based on weight values ​​of the trained improved grey GM (1,1) prediction model and actual historical helicopter fault values. The present invention can improve the accuracy of helicopter fault prediction results.

Description

Technical Field

[0001] The present invention relates to the technical field of helicopter fault detection, and in particular to a method and device for predicting helicopter faults. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. No statement herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] Helicopters, due to their large payload capacity, vertical takeoff and landing capabilities, and excellent low-speed maneuverability, have been widely used in a variety of missions, including power operations, disaster relief, and material transportation. These aircraft consist of a complex system, including the fuselage, rotor system, tail, landing gear, engine, and transmission. This complex system structure poses numerous challenges to helicopter maintenance and support. The increasing failure rate not only seriously impacts mission execution but also threatens the helicopter's reliable flight.

[0004] The use of Prognostics and Health Management (PHM) technology can transform helicopter maintenance from post-processing and reactive maintenance to regular inspections and proactive protection. This shift not only improves maintenance efficiency but also enhances the overall reliability and safety of helicopters. The core of PHM technology lies in fault diagnosis and prognosis, primarily by analyzing historical system data and current monitoring information to diagnose and predict the current and future health status, performance degradation, and potential failures.

[0005] Currently, PHM technology primarily relies on model-based and data-driven approaches. However, due to the complex operating conditions and strong coupling between components in real-world applications, establishing accurate mathematical models is relatively difficult. Currently, data-driven approaches are the mainstream prediction method, but they suffer from issues such as high computational complexity and inaccurate predictions. Therefore, a helicopter fault prediction method is urgently needed to address these issues. Summary of the Invention

[0006] An embodiment of the present invention provides a helicopter fault prediction method for reducing the amount of calculation and improving the accuracy of helicopter fault prediction. The method includes:

[0007] Obtain the status information of the helicopter to be tested; the status information includes: normal status data and fault status data;

[0008] The state information of the helicopter to be tested is input into the helicopter fault prediction model, and the fault prediction result is output; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values ​​and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults.

[0009] An embodiment of the present invention further provides a helicopter fault prediction device for reducing the amount of calculation and improving the accuracy of helicopter fault prediction. The device includes:

[0010] The status information acquisition module is used to obtain the status information of the helicopter to be tested; the status information includes: normal status data and fault status data;

[0011] The fault prediction result output module is used to input the state information of the helicopter to be tested into the helicopter fault prediction model and output the fault prediction result; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values ​​and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults.

[0012] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned helicopter fault prediction method when executing the computer program.

[0013] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the helicopter fault prediction method is implemented.

[0014] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned helicopter fault prediction method is implemented.

[0015] In an embodiment of the present invention, by obtaining the state information of a helicopter to be tested, the state information includes normal state data and fault state data; the state information of the helicopter to be tested is input into a helicopter fault prediction model, and a fault prediction result is output; the helicopter fault prediction model is constructed based on a trained improved grey GM (1, 1) prediction model and an Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values, and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; and the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and historical helicopter fault actual values. In the above process, the embodiment of the present invention adds a correction factor during the training process of the improved grey GM (1, 1) prediction model to improve the GM (1, 1) model's ability to resist interference from random factors, and combines the trained improved grey GM (1, 1) prediction model with the Elman neural network model, thereby improving the accuracy of the helicopter fault prediction result through the model combination. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0017] Figure 1 Flowchart of a helicopter fault prediction method according to an embodiment of the present invention;

[0018] Figure 2 Schematic diagram of a helicopter fault prediction process according to an embodiment of the present invention;

[0019] Figure 3 This is a flow chart of obtaining a correction factor in an embodiment of the present invention;

[0020] Figure 4 Schematic diagram of a helicopter fault prediction device according to an embodiment of the present invention;

[0021] Figure 5 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0023] Figure 1 Flowchart of a helicopter fault prediction method according to an embodiment of the present invention. The method includes:

[0024] Step 101, obtaining the state information of the helicopter to be tested; the state information includes: normal state data and fault state data;

[0025] Step 102: Input the state information of the helicopter to be tested into the helicopter fault prediction model and output the fault prediction result; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values ​​and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults.

[0026] Figure 2The figure is a schematic diagram of the helicopter fault prediction process in an embodiment of the present invention. In this embodiment, historical information is first queried from system operating data. The historical information includes the helicopter's status information, the failure rate of key components of the helicopter in each cycle, the failure category, the cause of the failure, and historical treatment measures. A fault database is established based on the historical information. Feature parameters related to the helicopter's fault status information are extracted from the fault database, and corresponding feature thresholds are set. Before training the improved gray GM (1, 1) prediction model, a level ratio test is first performed on the feature parameters related to the status information to determine whether the feature parameters meet the level ratio requirements. The level ratio is calculated for the training sample to determine whether the level ratio falls within the level ratio coverage range. If the level ratio does not fall within the coverage range, the feature parameter data is translated and transformed until the level ratio requirements are met. After confirming that the level ratio requirements are met, the improved gray GM (1, 1) prediction model is trained to establish an Elman neural network model. The established Elman neural network model is trained based on the weights of the trained improved gray GM (1, 1) prediction model. Before training the Elman, the output data of the Elman neural network model is normalized. During the Elman training process, the parameters of the Elman neural network model were adjusted according to the actual accuracy requirements. The resulting helicopter fault prediction model uses the improved grey GM(1,1) prediction model as the first-layer training network and the Elman neural network model as the second-layer training network. When the prediction result is greater than a set threshold, an alarm is triggered and repair recommendations are given based on historical maintenance records, expert experience database, and corresponding maintenance manuals. When the prediction result is less than the set threshold, no alarm is generated.

[0027] Each step is described in detail below.

[0028] In step 101, the state information of the helicopter to be tested is obtained; the state information includes: normal state data and fault state data.

[0029] In a specific embodiment, historical information is retrieved from system operating data. The historical information includes helicopter status information, failure rates of key helicopter components within each cycle, failure types, failure causes, and historical treatment measures. Based on the historical information, characteristic parameters affecting the key helicopter components are determined.

[0030] In step 102, the state information of the helicopter to be tested is input into the helicopter fault prediction model, and the fault prediction result is output; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values ​​and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults.

[0031] In one embodiment, a helicopter failure prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model, including:

[0032] A helicopter fault prediction model is constructed using the improved grey GM (1, 1) prediction model as the first layer training network and the Elman neural network model as the second layer training network.

[0033] In a specific embodiment, before training the improved grey GM (1, 1) prediction model, the level ratio is calculated for the training sample to determine whether the level ratio falls within the level ratio coverage range, that is, , is the magnitude ratio of the data sequence, t is the number of data sequence elements. After the training samples meet the level ratio requirement, the improved grey GM (1, 1) prediction model is trained.

[0034] In a specific embodiment, the first-order differential equation of the improved grey GM (1,1) prediction model is:

[0035] ,

[0036] in, is a monotonically increasing sequence, As a correction factor, a correction factor is added to the first-order differential equation of the improved grey GM (1,1) prediction model , thereby improving the GM(1,1) model's ability to resist interference from random factors in fault prediction, and improving the GM(1,1) model's ability to track the evolution process of mechanical system faults.

[0037] Based on the first-order differential equation of the improved grey GM (1,1) prediction model, the final improved grey GM (1,1) prediction model formula is determined as follows:

[0038] ;

[0039] in, Represents the original data output by the improved grey GM(1,1) prediction model, Represents the data output by the improved grey GM(1,1) prediction model after first-order processing, Represents the original state information, is the correction factor, To develop gray numbers, b To control the ash number.

[0040] Figure 3 This is a flow chart of obtaining correction factors in an embodiment of the present invention. In one embodiment, simulating pre-acquired performance degradation information of various components of a helicopter to obtain correction factors includes:

[0041] Step 301: simulating the previously acquired performance degradation information of each component of the helicopter based on simulation software to obtain simulation values;

[0042] Step 302: Compare the simulated value with the actual value to obtain an error value between the simulated value and the actual value to determine a correction factor.

[0043] In one embodiment, the improved grey GM(1,1) prediction model is trained based on the following correction factor formula:

[0044] ;

[0045] in, is the correction factor, To develop gray numbers, b To control the ash number, represents the data output by the improved grey GM(1,1) prediction model after first-order processing, is the characteristic parameter of the performance degradation information after first-order processing.

[0046] In a specific embodiment, the Elman neural network is a typical dynamic feedback neural network, and the nonlinear expression of Elman is:

[0047]

[0048] Where: l Indicates the moment; y , x , u and x c Respectively represent the network output, hidden layer output, external input of the network and output of the receiving layer; b 1 and b 2are the thresholds of the input layer and hidden layer respectively; , are the connection weight matrices from the receiving layer to the hidden layer, the input layer to the hidden layer, and the hidden layer to the output layer respectively; is the hidden layer transfer function, Transfer function for the output layer.

[0049] in, Take the sigmoid function:

[0050]

[0051] Take the linear function, that is:

[0052]

[0053] The input and output data of Elman are normalized. The input layer converts the data into values ​​in the interval [0, 1] so that larger inputs can fall where the gradient of the transfer function is large. The output layer denormalizes the prediction results to obtain the final fault prediction value.

[0054] The normalized data is then input into the network for training, and the weights and thresholds corresponding to each node in the network are obtained.

[0055] In one embodiment, after outputting the fault prediction result, the method further includes:

[0056] Compare the fault prediction results with the actual fault results, and calculate the root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) based on the comparison results;

[0057] The root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE) are used as evaluation indicators of fault prediction results to determine the accuracy of the fault prediction results.

[0058] In a specific embodiment, after the fault prediction result is output, the fault prediction result is compared with a set threshold: if the prediction result is less than the set threshold, no alarm is generated; if the prediction result is greater than the set threshold, an alarm is triggered, and based on the information of the relevant components of the alarm system, the historical maintenance records of the component are automatically searched, and maintenance suggestions are given in combination with the historical maintenance records, expert experience library and corresponding maintenance manuals, and the corresponding chapters of the maintenance manual are automatically called out and a maintenance work card is generated.

[0059] The present invention also provides a helicopter fault prediction device, as described in the following embodiments. Since the principle of the device to solve the problem is similar to that of the helicopter fault prediction method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0060] Figure 4 Schematic diagram of a helicopter fault prediction device according to an embodiment of the present invention, the device includes:

[0061] The status information acquisition module 401 is used to obtain the status information of the helicopter to be tested; the status information includes: normal status data and fault status data;

[0062] The fault prediction result output module 402 is used to input the state information of the helicopter to be tested into the helicopter fault prediction model and output the fault prediction result; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values ​​and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults.

[0063] In one embodiment, the fault prediction result output module 402 is further configured to:

[0064] Simulating the previously acquired performance degradation information of each helicopter component based on simulation software to obtain simulation values;

[0065] The simulation value is compared with the actual value to obtain the error value between the simulation value and the actual value to determine the correction factor.

[0066] In one embodiment, the fault prediction result output module 402 is specifically configured to:

[0067] The improved grey GM(1,1) prediction model is trained based on the following correction factor formula:

[0068] ;

[0069] in, is the correction factor, To develop gray numbers, b To control the ash number, represents the data output by the improved grey GM(1,1) prediction model after first-order processing, is the characteristic parameter of the performance degradation information after first-order processing.

[0070] In one embodiment, an evaluation index calculation module is further included, specifically configured to:

[0071] After outputting the fault prediction results, the fault prediction results are compared with the actual fault results, and the root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) are calculated based on the comparison results.

[0072] The root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE) are used as evaluation indicators of fault prediction results to determine the accuracy of the fault prediction results.

[0073] In one embodiment, the fault prediction result output module 402 is specifically configured to:

[0074] A helicopter fault prediction model is constructed using the improved grey GM (1, 1) prediction model as the first layer training network and the Elman neural network model as the second layer training network.

[0075] An embodiment of the present invention further provides a computer device, Figure 5 Schematic diagram of a computer device in an embodiment of the present invention. The computer device 500 includes a memory 510, a processor 520, and a computer program 530 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 530, the above-mentioned helicopter fault prediction method is implemented.

[0076] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the helicopter fault prediction method is implemented.

[0077] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned helicopter fault prediction method is implemented.

[0078] In an embodiment of the present invention, by obtaining the state information of a helicopter to be tested, the state information includes normal state data and fault state data; constructing a helicopter fault prediction model, the helicopter fault prediction model is constructed based on a trained improved grey GM (1, 1) prediction model and an Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault actual values, and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and historical helicopter fault prediction values; the state information of the helicopter to be tested is input into the helicopter fault prediction model, and a fault prediction result is output. In the above process, the embodiment of the present invention adds a correction factor during the training process of the improved grey GM (1, 1) prediction model to improve the GM (1, 1) model's anti-interference ability against random factors, and combines the trained improved grey GM (1, 1) prediction model with the Elman neural network model, thereby improving the accuracy of the helicopter fault prediction result through the model combination.

[0079] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, 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 magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0081] 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 work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0083] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A helicopter fault prediction method, characterized in that: include: Get the test status information of the helicopter; Status information includes: normal status data and fault status data; The state information of the helicopter to be tested is input into the helicopter fault prediction model, and the fault prediction result is output; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values, and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults; The previously acquired performance degradation information of each helicopter component is simulated to obtain the correction factors, including: Simulating the previously acquired performance degradation information of each helicopter component based on simulation software to obtain simulation values; Compare the simulation value with the actual value to obtain the error value between the simulation value and the actual value to determine the correction factor; The improved grey GM(1,1) prediction model is trained based on the following correction factor formula: ； in, is the correction factor, To develop gray numbers, b To control the ash number, represents the data output by the improved grey GM(1,1) prediction model after first-order processing, is the characteristic parameter of the performance degradation information after first-order processing.

2. The method according to claim 1, wherein After outputting the fault prediction results, it also includes: Compare the fault prediction results with the actual fault results, and calculate the root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) based on the comparison results; The root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE) are used as evaluation indicators of fault prediction results to determine the accuracy of the fault prediction results.

3. The method according to claim 1, wherein Based on the trained improved grey GM (1, 1) prediction model and Elman neural network model, a helicopter fault prediction model is constructed, including: A helicopter fault prediction model is constructed using the improved grey GM (1, 1) prediction model as the first layer training network and the Elman neural network model as the second layer training network.

4. A helicopter fault prediction device, characterized in that: include: A status information acquisition module is used to obtain the status information of the helicopter to be tested; Status information includes: normal status data and fault status data; The fault prediction result output module is used to input the state information of the helicopter to be tested into the helicopter fault prediction model and output the fault prediction result; the helicopter fault prediction model is constructed based on the trained improved grey GM (1, 1) prediction model and the Elman neural network model; the trained improved grey GM (1, 1) prediction model is trained based on historical state information, historical helicopter fault prediction values ​​and correction factors; the correction factors are obtained by simulating the performance degradation information of each helicopter component obtained in advance; the trained Elman neural network model is trained based on the weight values ​​of the trained improved grey GM (1, 1) prediction model and the actual values ​​of historical helicopter faults; The fault prediction result output module is also used to: Simulating the previously acquired performance degradation information of each helicopter component based on simulation software to obtain simulation values; Compare the simulation value with the actual value to obtain the error value between the simulation value and the actual value to determine the correction factor; The fault prediction result output module is specifically used to: The improved grey GM(1,1) prediction model is trained based on the following correction factor formula: ； in, is the correction factor, To develop gray numbers, b To control the ash number, represents the data output by the improved grey GM(1,1) prediction model after first-order processing, is the characteristic parameter of the performance degradation information after first-order processing.

5. The device according to claim 4, characterized in that It also includes an evaluation index calculation module, which is specifically used to: After outputting the fault prediction results, the fault prediction results are compared with the actual fault results, and the root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) are calculated based on the comparison results. The root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE) are used as evaluation indicators of fault prediction results to determine the accuracy of the fault prediction results.

6. The device according to claim 4, characterized in that The fault prediction result output module is specifically used to: A helicopter fault prediction model is constructed using the improved grey GM (1, 1) prediction model as the first layer training network and the Elman neural network model as the second layer training network.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 3 is implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

9. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images