Low-altitude aircraft state prediction and health management method and system

Abstract

The invention provides a low-altitude aircraft state prediction and health management method and system. The method comprises the following steps: acquiring sensor simulation data streams corresponding to a plurality of subsystems of a low-altitude aircraft; predicting the residual life of the target subsystem according to the sensor simulation data flow to obtain residual life prediction time, and confirming health state type information of the sensor simulation data flow; performing comprehensive analysis on the health state of the sensor simulation data flow by adopting the target diagnosis model according to the health state type information to obtain a comprehensive analysis result; and performing maintenance decision on the target subsystem according to the residual life prediction time and the comprehensive analysis result to obtain a maintenance decision result. According to the method, the health state evaluation accuracy of the low-altitude aircraft in the whole life cycle is improved, and health management and state analysis of the overall structure of the low-altitude aircraft in the whole life cycle are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of low-altitude aircraft health management, and particularly relates to a low-altitude aircraft state prediction and health management method and system. BACKGROUND

[0002] The PHM (Prognostics and Health Management) system of the low-altitude aircraft can monitor and diagnose the health status of each component and sensor of the low-altitude aircraft, especially in the sudden working condition or critical task scene, not only can realize early fault diagnosis and residual life prediction, but also can show the maintenance strategy suggestion for the user, and provide data support for task dynamic scheduling. Therefore, it has practical significance to scientifically and whole life cycle health management and state analysis of the low-altitude aircraft.

[0003] In the related art, the existing fault prediction and health management method focuses on intelligent diagnosis and analysis of the health status of a certain type of structure of the low-altitude aircraft in the current life cycle, resulting in inaccurate overall health status evaluation results of the low-altitude aircraft in the whole life cycle health, and it is difficult to meet the demand of whole life cycle health management and state analysis of the low-altitude aircraft. SUMMARY

[0004] The present application provides a low-altitude aircraft state prediction and health management method and system, which solves the defect that the prior art focuses on intelligent diagnosis and analysis of the health status of a certain type of structure of the low-altitude aircraft in the current life cycle, resulting in inaccurate overall health status evaluation results of the low-altitude aircraft in the whole life cycle health; the method improves the health status evaluation accuracy of the low-altitude aircraft in the whole life cycle, and realizes the health management and state analysis of the overall structure of the low-altitude aircraft in the whole life cycle.

[0005] The present application provides a low-altitude aircraft state prediction and health management method, comprising: Obtaining sensor simulation data streams corresponding to a plurality of subsystems of a low-altitude aircraft respectively; According to the sensor simulation data stream, the residual life of the target subsystem is predicted to obtain a residual life prediction time, and the health status type information of the sensor simulation data stream is confirmed; wherein the health status type information is used to represent the health degree or abnormal degree of the sensor simulation data stream, and the target subsystem belongs to the plurality of subsystems; adopting a target diagnosis model to comprehensively analyze the health state of the sensor simulation data stream according to the health state type information, and obtaining a comprehensive analysis result; wherein the target diagnosis model comprises one of a mixed diagnosis model cluster and a mixed prediction model, the mixed diagnosis model cluster is used for fault analysis on abnormal data stream, and the mixed prediction model is used for fault prediction on healthy data stream; making a maintenance decision on the target subsystem according to the residual life prediction time and the comprehensive analysis result, and obtaining a maintenance decision result.

[0006] According to the low-altitude aircraft state prediction and health management method provided by the application, the target diagnosis model is a mixed diagnosis model cluster. After the comprehensive analysis result is obtained, the method further comprises: querying the comprehensive analysis result for fault cause information corresponding to the comprehensive analysis result by using a fault mode analysis table constructed based on FMEA; calculating a failure severity index according to the fault cause information.

[0007] According to the low-altitude aircraft state prediction and health management method provided by the application, after the health state type information of the sensor simulation data stream is confirmed, the method further comprises: extracting a plurality of index parameters from the health state type information; calculating the weight corresponding to each index parameter according to a relationship importance degree judgment matrix constructed by index and expert knowledge, and calculating a comprehensive performance index according to the weight and each index parameter.

[0008] According to the low-altitude aircraft state prediction and health management method provided by the application, after the plurality of subsystems of the low-altitude aircraft are obtained, the method further comprises: performing a preprocessing operation on the sensor simulation data stream to obtain a preprocessed data stream; wherein the preprocessing operation comprises at least one of data cleaning, removing outliers and standardization processing.

[0009] According to the low-altitude aircraft state prediction and health management method provided by the application, after the maintenance decision result is obtained, the method further comprises: inputting the comprehensive performance index corresponding to the health state type information, the failure severity index corresponding to the comprehensive analysis result output by the mixed diagnosis model cluster and the maintenance decision result into a management module based on B / S architecture to perform at least one of the following operations: interactive management of the functions corresponding to each subsystem; visual display.

[0010] The application further provides a low-altitude aircraft state prediction and health management system, comprising: a data stream acquisition module configured to acquire sensor simulation data streams corresponding to a plurality of subsystems of a low-altitude aircraft respectively; a health assessment module configured to predict a remaining life of a target subsystem according to the sensor simulation data streams, obtain a remaining life prediction time, and confirm health state type information of the sensor simulation data streams; wherein the health state type information is used to represent a health degree or an abnormality degree of the sensor simulation data streams, and the target subsystem belongs to the plurality of subsystems; a diagnosis and prediction module configured to comprehensively analyze a health state of the sensor simulation data streams according to the health state type information by using a target diagnosis model, and obtain a comprehensive analysis result; wherein the target diagnosis model comprises one of a mixed diagnosis model cluster and a mixed prediction model, the mixed diagnosis model cluster is used to analyze faults of abnormal data streams, and the mixed prediction model is used to predict faults of healthy data streams; a first decision module configured to make a maintenance decision for the target subsystem according to the remaining life prediction time and the comprehensive analysis result, and obtain a maintenance decision result.

[0011] According to the low-altitude aircraft state prediction and health management system provided by the application, the target diagnosis model is a mixed diagnosis model cluster; the system further comprises: a second decision module configured to query fault cause information corresponding to the comprehensive analysis result by using a fault mode analysis table constructed based on FMEA after the comprehensive analysis result is obtained; and calculate a failure severity index according to the fault cause information.

[0012] According to the low-altitude aircraft state prediction and health management system provided by the application, the system further comprises: a third decision module configured to extract a plurality of index parameters from the health state type information after the health state type information of the sensor simulation data streams is confirmed; calculate corresponding weights of the index parameters according to a relationship importance degree judgment matrix constructed by using indexes and expert knowledge, and calculate a comprehensive performance index according to the weights and the index parameters.

[0013] The application further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the low-altitude aircraft state prediction and health management method according to any one of the above-mentioned methods when executing the computer program.

[0014] The application further provides a non-transitory computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the low-altitude aircraft state prediction and health management method.

[0015] The application further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the low-altitude aircraft state prediction and health management method.

[0016] The low-altitude aircraft state prediction and health management method and system provided by the application can predict the remaining life of a target subsystem through sensor simulation data flow, obtain a remaining life prediction time, confirm health state type information of the sensor simulation data flow, comprehensively analyze the health state of the sensor simulation data flow according to the health state type information by using a target diagnosis model, obtain a comprehensive analysis result, and finally make a maintenance decision for the target subsystem according to the remaining life prediction time and the comprehensive analysis result, thereby obtaining a maintenance decision result, improving the health state evaluation accuracy of the low-altitude aircraft in the whole life cycle, and realizing health management and state analysis of the overall structure of the low-altitude aircraft in the whole life cycle. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort.

[0018] Figure 1 is a flowchart of the low-altitude aircraft state prediction and health management method provided by the application.

[0019] Figure 2 is one of structural schematic diagrams of the low-altitude aircraft state prediction and health management device provided by the application.

[0020] Figure 3 is another structural schematic diagram of the low-altitude aircraft state prediction and health management device provided by the application.

[0021] Figure 4 is a third structural schematic diagram of the low-altitude aircraft state prediction and health management device provided by the application.

[0022] Figure 5 is a fourth structural schematic diagram of the low-altitude aircraft state prediction and health management device provided by the application.

[0023] Figure 6 is a structural schematic diagram of an electronic device provided by the application. DETAILED DESCRIPTION

[0024] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0025] The low-altitude aircraft state prediction and health management method and system of the present application will be described below. Figures 1-5 The low-altitude aircraft state prediction and health management method and system of the present application will be described below.

[0026] Figure 1 The flowchart of the low-altitude aircraft state prediction and health management method provided by the present application is shown in FIG. 1, which comprises the following steps: Figure 1 Step 110: obtaining sensor simulation data streams corresponding to a plurality of subsystems of a low-altitude aircraft respectively.

[0027] In this step, the plurality of subsystems include a power system, an avionics system, a navigation control, etc.

[0028] In this embodiment, the low-altitude aircraft sensor parameter configuration module is configured to configure operating parameters of IMU sensors, motors and other components of the low-altitude aircraft, and output sensor performance parameters; the simulation condition parameter configuration module is configured to configure simulation time length, influence factors and other parameters, and output simulation data generation parameters; finally, the low-altitude aircraft sensor data generation module based on Simulink runs the constructed sensor model according to the sensor performance parameters and the simulation data generation parameters, to obtain corresponding sensor simulation data streams; different subsystems correspond to different sensor simulation data streams.

[0029] Step 120: predicting the remaining life of a target subsystem according to the sensor simulation data streams, obtaining a remaining life prediction time, and confirming health state type information of the sensor simulation data streams; wherein the health state type information is used to represent the health degree or abnormality degree of the sensor simulation data streams, and the target subsystem belongs to the plurality of subsystems.

[0030] In this step, the sensor simulation data streams contain multi-dimensional features, such as motor temperature, rotating speed, vibration frequency, etc., which are extracted from the low-altitude aircraft sensor parameter configuration module (such as IMU sensor parameters) and generated based on simulation condition parameters (such as influence factors); the present embodiment can determine the data streams as healthy or abnormal according to a plurality of sensor data features.

[0031] ​In this embodiment, the sensor simulation data stream is input into the remaining life prediction model, and the remaining life prediction time is output.

[0032] In this embodiment, the remaining life prediction model can be trained by a bidirectional long short-term memory neural network according to sample time parameters; the remaining life prediction model can provide remaining life prediction for each system unit of the low-altitude aircraft to realize overall remaining life prediction of the low-altitude aircraft.

[0033] Step 130, comprehensively analyzing the health state of the sensor simulation data stream according to the health state type information by using a target diagnosis model to obtain a comprehensive analysis result; wherein the target diagnosis model includes one of a mixed diagnosis model cluster and a mixed prediction model, the mixed diagnosis model cluster is used for fault analysis on abnormal data streams, and the mixed prediction model is used for fault prediction on healthy data streams.

[0034] In this step, the target diagnosis model can be a dual-mode intelligent diagnosis and prediction module, and the use scenarios of the model include: (a) When the health state type information of the sensor simulation data stream is abnormal, activate the mixed diagnosis model cluster composed of support vector machines, random forests and deep belief networks, and introduce a decision-level fusion algorithm based on expert judgment to obtain a comprehensive analysis result for root cause positioning of the fault.

[0035] (b) When the health state type information of the sensor simulation data stream is normal, start the mixed prediction model of the fusion LSTM, and introduce a decision-level fusion algorithm based on expert judgment to obtain a comprehensive prediction analysis result for predicting the time of fault occurrence.

[0036] Step 140, making a maintenance decision for the target subsystem according to the remaining life prediction time and the comprehensive analysis result to obtain a maintenance decision result.

[0037] In this step, a maintenance decision module based on machine learning can be used to receive data of the comprehensive analysis result output from the diagnosis model cluster or the mixed prediction model, and the remaining life prediction time output from the remaining life prediction model, and input these result data into a random forest model to obtain a maintenance decision result.

[0038] The low-altitude aircraft state prediction and health management method provided in the embodiment of the application predicts the residual life of a target subsystem through sensor simulation data flow, obtains a residual life prediction time, confirms health state type information of the sensor simulation data flow, comprehensively analyzes the health state of the sensor simulation data flow according to the health state type information by using a target diagnosis model, obtains a comprehensive analysis result, and finally makes a maintenance decision for the target subsystem according to the residual life prediction time and the comprehensive analysis result, thereby obtaining a maintenance decision result, improving the health state evaluation accuracy of the low-altitude aircraft in the whole life cycle, and realizing health management and state analysis of the overall structure of the low-altitude aircraft in the whole life cycle.

[0039] In some embodiments, the target diagnosis model is a mixed diagnosis model cluster; after the comprehensive analysis result is obtained, the method further includes: querying the failure cause information corresponding to the comprehensive analysis result by using a failure mode analysis table constructed based on FMEA; and calculating the failure severity index according to the failure cause information.

[0040] In this embodiment, the FMEA analysis table contains key fields such as failure mode code (for example, weight range 1-10, which is predefined by domain experts according to historical failure data), failure phenomenon description, root cause, and influence subsystem.

[0041] A feasible embodiment includes: the root cause of the power system “motor overheating” failure mode is mapped to: radiator blockage (weight S=8), winding insulation aging (weight S=9).

[0042] In this embodiment, the historical failure data in the FMEA analysis table can be derived from low-altitude aircraft design documents, maintenance records, and simulation test results.

[0043] In this embodiment, the failure severity index is calculated by the following formula: Failure severity index=S×O×D; Wherein, S is severity, indicating the influence degree of the failure on the system function; O is occurrence probability, indicating the frequency of the failure in the historical data; and D is detectability, indicating the detection difficulty of the system for the failure.

[0044] The low-altitude aircraft state prediction and health management method provided in the embodiment of the application queries the failure cause information corresponding to the comprehensive analysis result by using a failure mode analysis table constructed based on FMEA; calculates the failure severity index according to the failure cause information, realizes accurate positioning and severity quantification of the failure cause, and provides reliable judgment basis for auxiliary decision-making.

[0045] In some embodiments, after confirming the health state type information of the sensor simulation data stream, the low-altitude aircraft state prediction and health management method further comprises: extracting a plurality of index parameters from the health state type information; calculating the corresponding weight of each index parameter according to a relationship importance judgment matrix constructed by the index and expert knowledge, and calculating a comprehensive performance index according to the weight and each index parameter.

[0046] In this embodiment, a plurality of multi-level evaluation indexes (i.e., index parameters) are extracted from the health state type information, wherein the index parameters can be divided into three levels: system-level indexes, subsystem-level indexes, and component-level indexes.

[0047] In this embodiment, the system-level indexes include the overall health degree of the low-altitude aircraft; the subsystem-level indexes include power system stability or navigation control accuracy, etc.; and the component-level indexes include motor temperature deviation or IMU sensor error rate, etc.

[0048] In this embodiment, the importance of two indexes in the same level can be compared according to expert knowledge, a relationship importance judgment matrix is constructed, and the maximum eigenvalue and the corresponding eigenvector of the matrix are solved by using the eigenvector method, the weight set is obtained after normalization, then the consistency ratio CR is calculated, and finally the failure severity index is calculated according to the corresponding weight of each index parameter and the corresponding standard value parameter value of the index parameter by using a weighted aggregation formula.

[0049] The low-altitude aircraft state prediction and health management method provided by the embodiment of the application realizes the quantitative evaluation of the overall effectiveness of the low-altitude aircraft by extracting a plurality of index parameters from the health state type information, calculating the corresponding weight of each index parameter according to a relationship importance judgment matrix constructed by the index and expert knowledge, and calculating a comprehensive performance index according to the weight and each index parameter.

[0050] The embodiment realizes intelligent health state diagnosis based on data driving and three-in-one comprehensive decision evaluation; the intelligent health state diagnosis based on data driving is realized by constructing a deep learning-based fault diagnosis & prediction and residual life prediction model, and the health state of each subsystem component of the low-altitude aircraft at the current and future time is calculated and analyzed; the three-in-one comprehensive decision evaluation is realized by constructing a machine learning-based maintenance decision, an effectiveness evaluation based on analytic hierarchy process, and a reliability analysis method based on FMEA, realizing multi-dimensional comprehensive decision evaluation of the health state effectiveness of the low-altitude aircraft; in combination with the above two parts, the health diagnosis and comprehensive health state analysis of the low-altitude aircraft in the whole life cycle are effectively solved.

[0051] In some embodiments, after obtaining the sensor simulation data streams corresponding to the plurality of subsystems of the low-altitude flying vehicle respectively, the low-altitude flying vehicle state prediction and health management method further comprises: performing a preprocessing operation on the sensor simulation data streams to obtain preprocessed data streams, wherein the preprocessing operation comprises at least one of data cleaning, outlier rejection, and standardization processing.

[0052] In this embodiment, the multi-source data processing and storage module is used to realize format alignment of the sensor simulation data streams corresponding to the plurality of subsystems respectively by using a plurality of preprocessing algorithms, and to complete data cleaning, outlier rejection, and standardization preprocessing, output uniform-dimension data, and mixed storage of structured data such as flight parameters and other unstructured data.

[0053] In this embodiment, the multi-source data storage and processing module comprises a data storage and processing module based on a data lake and a big data component, which is used to receive the sensor simulation data streams, then build a data transmission channel through the fusion of components such as a message queue system (for example, Kafka) and a streaming processing framework (for example, flink), and realize preprocessing operations such as data labeling, normalization, and outlier rejection on the multi-sensor data streams, to realize unified cleaning and interval division of data.

[0054] In this embodiment, by constructing a high-performance distributed computing architecture, using a technical solution of deep fusion of a data lake storage system and a big data computing framework, and based on distributed computing components, multi-dimensional processing of million-level sensor data is realized, and data retrieval, feature calculation, data aggregation, and dynamic splitting in a high-concurrency scenario are supported.

[0055] The low-altitude flying vehicle state prediction and health management method provided in the embodiments of the present application realizes preprocessing of sensor simulation data streams, and realizes storage and computing processing of unstructured data by using a data lake and a big data streaming processing technology, so that the system can store a large amount of historical offline data and store and output simulated real-time data, that is, the system can be used for offline non-real-time processing of big data, and can also be used as a real-time processing and storage platform for real-time health management testing of a low-altitude flying vehicle.

[0056] In some embodiments, after obtaining the maintenance decision result, the low-altitude flying vehicle state prediction and health management method further comprises: inputting the comprehensive performance index corresponding to the health state type information, the failure severity index corresponding to the comprehensive analysis result output by the mixed diagnosis model cluster, and the maintenance decision result into a management module based on a B / S architecture, to perform at least one of the following operations: interactive management of functions corresponding to the subsystems; and visual display.

[0057] In this embodiment, a multi-dimensional data integration visualization and interaction platform is constructed to receive all element information (including the aforementioned comprehensive performance indicators, failure severity indicators, and maintenance decision results) processed and calculated from the data storage service terminal and the intelligent health analysis service terminal. This platform integrates and visualizes these low-altitude aircraft PHM full life cycle elements and has management functions such as information interaction operations, thereby realizing integrated monitoring and management of data, status, and analysis results.

[0058] In this embodiment, the multidimensional data integration visualization and interaction platform includes a database module based on a B / S architecture, an application system function interaction module, and a PHM data display module. The functions of each sub-module are as follows: (a) The database module based on the B / S architecture receives all the data from the preceding calculation process, including raw sensor data of multiple components of the low-altitude aircraft, fault diagnosis and prediction results, life prediction results, and intelligent decision evaluation data, providing an integrated database that can be called by the visualization system.

[0059] (b) The application system function interaction module realizes convenient interactive management of the B / S architecture-based visualization system through the constructed software, and has interactive operation functions such as system management and database data operation management.

[0060] (c) The PHM data display module includes multiple interactive interface visualization component modules, which display the health status data of the entire life cycle in the database module through a variety of front-end visualization components.

[0061] The low-altitude aircraft status prediction and health management method provided in this invention inputs comprehensive performance indicators, failure severity indicators, and maintenance decision results into a management module based on a B / S architecture to perform interactive management and visualization of the corresponding functions of each subsystem. This method takes into account both user interaction experience and the need for user-friendly visualization of multiple elements. It effectively solves the problems of slow calculation speed, difficult retrieval, and slow display speed caused by massive data in the visualization management system, and provides full life-cycle health status management visualization support for the dynamic flight process of low-altitude aircraft.

[0062] Figure 2 This is one of the structural schematic diagrams of the low-altitude aircraft state prediction and health management device provided by the present invention. Figure 2In the illustrated embodiment, the low-altitude aircraft status prediction and health management system includes a low-altitude aircraft simulation server for acquiring sensor simulation data streams corresponding to multiple subsystems of the low-altitude aircraft; a data storage service server for preprocessing the sensor simulation data streams to obtain preprocessed data streams; and an intelligent health analysis server for predicting the remaining lifetime of a target subsystem based on the sensor simulation data streams, obtaining the remaining lifetime prediction time, and confirming the health status type information of the sensor simulation data streams; using a target diagnostic model to perform a comprehensive analysis of the health status of the sensor simulation data streams based on the health status type information to obtain a comprehensive analysis result; making maintenance decisions for the target subsystem based on the remaining lifetime prediction time and the comprehensive analysis result to obtain a maintenance decision result; using a failure mode analysis table built based on FMEA to query the failure cause information corresponding to the comprehensive analysis result; calculating the failure severity index based on the failure cause information; extracting multiple indicator parameters from the health status type information; calculating the weights corresponding to each indicator parameter based on the relationship importance judgment matrix constructed through indicators and expert knowledge; and calculating a comprehensive performance index based on the weights and each indicator parameter.

[0063] Figure 3 This is the second structural schematic diagram of the low-altitude aircraft state prediction and health management device provided by the present invention. Figure 3 In the illustrated embodiment, the low-altitude aircraft simulation server includes a data dynamic simulation module; the data storage service includes a multi-source data processing and storage module; the intelligent health analysis server includes a health assessment module, a dual-modal intelligent diagnosis and prediction module, a remaining lifespan prediction and calculation module, and an intelligent decision evaluation module; and the user interaction server includes a multi-dimensional data visualization and interactive operation module.

[0064] Specifically, the data dynamic simulation module generates a full-dimensional simulation data stream encompassing all subsystems of the low-altitude aircraft, including the power system and navigation control, providing a configurable dynamic verification data source for the system; the multi-source data processing and storage module uses various preprocessing algorithms to align data on demand, completes data cleaning, outlier removal, and standardization preprocessing, outputs data with unified dimensions, and stores structured data such as flight parameters mixed with other unstructured data; the health assessment module determines the health or abnormal rating of the data stream based on the characteristics of multi-sensor data; the dual-modal intelligent diagnosis and prediction module constructs a dual-modal intelligent diagnosis and prediction system; and the remaining lifetime prediction calculation module calculates the remaining lifetime based on the established... The system employs a deep learning-based Remaining Useful Life (RUL) prediction model, outputting multi-sensor time series prediction results; an intelligent decision evaluation module, used to construct a three-in-one intelligent decision support system, enabling the generation of maintenance decision suggestions, calculation of overall system performance evaluation scores, and reliability analysis results; and a multi-dimensional data visualization and interactive operation module, used to construct a multi-dimensional data integration visualization and interactive platform, receiving all element information processed and calculated from the data storage service terminal and the intelligent health analysis service terminal, integrating and visualizing these low-altitude aircraft PHM full life cycle elements, and possessing information interaction and operation management functions, realizing integrated monitoring and management of data, status, and analysis results.

[0065] Figure 4 This is the third schematic diagram of the low-altitude aircraft state prediction and health management device provided by the present invention. Figure 4In the illustrated embodiment, the data dynamic simulation module comprises three modules, 101 to 103. The low-altitude aircraft sensor parameter configuration module 101 configures the operating parameters of components such as the low-altitude aircraft IMU sensor and motors, and outputs sensor performance parameters. The simulation condition parameter configuration module 102 configures parameters such as simulation time length and influence factors, and outputs simulation data generation parameters. The Simulink-based low-altitude aircraft sensor data generation module 103 receives the configuration parameters from modules 101 and 102, runs the constructed sensor model, and outputs the configured low-altitude aircraft simulation sensor data stream. The multi-source data storage and processing module mainly includes... The system includes a data storage and processing module 201 based on data lake and big data components. This module receives multi-sensor data streams from the data generation module 103 and performs preprocessing operations such as data labeling, normalization, and outlier removal to achieve unified data cleaning and interval partitioning. The health assessment module mainly includes a status health determination module 301. This module receives data from the storage and processing module 201, performs health discrimination classification based on data features, and outputs data to a multi-model-based fault diagnosis module 401 when the input batch of data is determined to be abnormal. Conversely, when the input batch of data is determined to be healthy, the data is output to a multi-model-based fault prediction module 401. The testing module 402; the dual-modal intelligent diagnostic prediction module includes a multi-model-based fault diagnosis module 401 and a multi-model-based fault prediction module 402, wherein the diagnosis module 401 receives data segments belonging to anomaly markers, inputs them into the constructed fault diagnosis model, and outputs the predicted possible future fault time results; the prediction module 402 receives data segments belonging to health markers, inputs them into the constructed fault prediction model, and outputs the predicted possible future fault time results; the remaining lifetime prediction calculation module includes a deep learning-based remaining lifetime prediction module 501, which, after receiving data segments from module 201, inputs the data into the constructed remaining lifetime prediction model. The remaining life prediction model outputs the life prediction time; the intelligent decision evaluation module includes a comprehensive performance evaluation module 601 based on hierarchical analysis (outputting comprehensive performance indicators, i.e., quantitative results of performance evaluation), a reliability analysis module 602 based on FEMA (outputting failure severity indicators, i.e., reliability analysis results), and a maintenance decision module 603 based on machine learning (outputting maintenance decision results); the multi-dimensional data visualization and interactive operation module includes a database module 701 based on B / S architecture (used for managing the system backend), an application system function interaction module 702 (used for user interaction), and a PHM data display module 703 (used for multi-dimensional data visualization).

[0066] The low-altitude aircraft status prediction and health management system provided by the present invention is described below. The low-altitude aircraft status prediction and health management system described below can be referred to in correspondence with the low-altitude aircraft status prediction and health management method described above.

[0067] Figure 5 This is the fourth structural schematic diagram of the low-altitude aircraft state prediction and health management device provided by the present invention, as shown below. Figure 5 As shown, the low-altitude aircraft status prediction and health management device includes: a data stream acquisition module 510, a health assessment module 520, a diagnosis and prediction module 530, and a first decision module 540.

[0068] The data stream acquisition module 510 is used to acquire sensor simulation data streams corresponding to multiple subsystems of the low-altitude aircraft. The health assessment module 520 is used to predict the remaining lifetime of the target subsystem based on the sensor simulation data stream, obtain the remaining lifetime prediction time, and confirm the health status type information of the sensor simulation data stream; wherein, the health status type information is used to characterize the health level or abnormality level of the sensor simulation data stream, and the target subsystem belongs to multiple subsystems; The diagnosis and prediction module 530 is used to perform a comprehensive analysis of the health status of the sensor simulation data stream based on the health status type information using a target diagnosis model, and obtain a comprehensive analysis result. The target diagnosis model includes one of a hybrid diagnosis model cluster and a hybrid prediction model. The hybrid diagnosis model cluster is used to perform fault analysis on abnormal data streams, and the hybrid prediction model is used to predict faults in healthy data streams. The first decision module 540 is used to make maintenance decisions for the target subsystem based on the remaining useful life prediction time and comprehensive analysis results, and obtain maintenance decision results.

[0069] The low-altitude aircraft status prediction and health management system provided in this invention predicts the remaining lifespan of a target subsystem using sensor simulation data streams, obtains the remaining lifespan prediction time, confirms the health status type information of the sensor simulation data streams, and then uses a target diagnostic model to comprehensively analyze the health status of the sensor simulation data streams based on the health status type information to obtain comprehensive analysis results. Finally, maintenance decisions are made for the target subsystem based on the remaining lifespan prediction time and comprehensive analysis results to obtain maintenance decision results. This improves the accuracy of health status assessment throughout the entire life cycle of low-altitude aircraft and realizes health management and status analysis of the overall structure of low-altitude aircraft throughout its entire life cycle.

[0070] In some embodiments, the target diagnostic model is a cluster of hybrid diagnostic models; the low-altitude aircraft state prediction and health management system further includes: a second decision module.

[0071] The second decision module is used to query the failure cause information corresponding to the comprehensive analysis results using the failure mode analysis table built based on FMEA after obtaining the comprehensive analysis results; and to calculate the failure severity index based on the failure cause information.

[0072] The low-altitude aircraft status prediction and health management system provided in this embodiment of the invention queries the fault cause information corresponding to the comprehensive analysis results by using a fault mode analysis table built based on FMEA; calculates the failure severity index according to the fault cause information, realizes the accurate location and severity quantification of the fault cause, and provides a reliable judgment basis for auxiliary decision-making.

[0073] In some embodiments, the low-altitude aircraft state prediction and health management system further includes a third decision module.

[0074] The third decision module is used to extract multiple indicator parameters from the health status type information after confirming the health status type information of the sensor simulation data stream; calculate the weight of each indicator parameter according to the importance judgment matrix constructed by the relationship between the indicators and expert knowledge; and calculate the comprehensive performance index according to the weight and each indicator parameter.

[0075] The low-altitude aircraft status prediction and health management system provided in this embodiment of the invention extracts multiple indicator parameters from health status type information; calculates the corresponding weights of each indicator parameter based on the relationship importance judgment matrix constructed by the indicators and expert knowledge; and calculates the comprehensive performance index based on the weights and each indicator parameter, thereby realizing a quantitative evaluation of the overall effectiveness of the low-altitude aircraft.

[0076] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communications bus 640, wherein the processor 610, the communications interface 620, and the memory 630 communicate with each other through the communications bus 640. The processor 610 can call logic instructions in the memory 630 to execute a low-altitude aircraft state prediction and health management method. This method includes: acquiring sensor simulation data streams corresponding to multiple subsystems of the low-altitude aircraft; predicting the remaining lifetime of the target subsystem based on the sensor simulation data streams to obtain the remaining lifetime prediction time, and confirming the health status type information of the sensor simulation data streams; wherein the health status type information characterizes the health level or abnormality level of the sensor simulation data streams, and the target subsystem belongs to multiple subsystems; using a target diagnostic model to comprehensively analyze the health status of the sensor simulation data streams based on the health status type information to obtain a comprehensive analysis result; wherein the target diagnostic model includes one of a hybrid diagnostic model cluster and a hybrid prediction model, the hybrid diagnostic model cluster being used for fault analysis of abnormal data streams, and the hybrid prediction model being used for fault prediction of healthy data streams; and making maintenance decisions for the target subsystem based on the remaining lifetime prediction time and the comprehensive analysis result to obtain a maintenance decision result.

[0077] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the low-altitude aircraft state prediction and health management method provided by the above methods. The method includes: acquiring sensor simulation data streams corresponding to multiple subsystems of the low-altitude aircraft; predicting the remaining lifetime of the target subsystem based on the sensor simulation data streams to obtain the remaining lifetime prediction time, and confirming the health status type information of the sensor simulation data streams; wherein, the health status type information is used to characterize the health level or abnormality level of the sensor simulation data streams, and the target subsystem belongs to multiple subsystems; using a target diagnostic model to comprehensively analyze the health status of the sensor simulation data streams based on the health status type information to obtain a comprehensive analysis result; wherein, the target diagnostic model includes one of a hybrid diagnostic model cluster and a hybrid prediction model, the hybrid diagnostic model cluster is used for fault analysis of abnormal data streams, and the hybrid prediction model is used for fault prediction of healthy data streams; and making maintenance decisions for the target subsystem based on the remaining lifetime prediction time and the comprehensive analysis result to obtain a maintenance decision result.

[0079] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0080] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the condition and managing the health of low-altitude aircraft, characterized in that, include: Obtain the sensor simulation data streams corresponding to the multiple subsystems of the low-altitude aircraft; The remaining lifetime of the target subsystem is predicted based on the sensor simulation data stream to obtain the remaining lifetime prediction time, and the health status type information of the sensor simulation data stream is confirmed; wherein, the health status type information is used to characterize the health level or abnormality level of the sensor simulation data stream, and the target subsystem belongs to the multiple subsystems; A target diagnostic model is used to comprehensively analyze the health status of the sensor simulation data stream based on the health status type information, and a comprehensive analysis result is obtained. The target diagnostic model includes one of a hybrid diagnostic model cluster and a hybrid prediction model. The hybrid diagnostic model cluster is used to perform fault analysis on abnormal data streams, and the hybrid prediction model is used to predict faults in healthy data streams. Based on the predicted remaining lifespan and the comprehensive analysis results, maintenance decisions are made for the target subsystem to obtain maintenance decision results.

2. The method for low-altitude aircraft status prediction and health management according to claim 1, characterized in that, The target diagnostic model is a cluster of hybrid diagnostic models; After obtaining the comprehensive analysis results, the method further includes: The fault cause information corresponding to the comprehensive analysis results is queried using the Fault Mode Analysis table built based on FMEA; The failure severity index is calculated based on the fault cause information.

3. The method for low-altitude aircraft condition prediction and health management according to claim 1, characterized in that, After confirming the health status type information of the sensor simulation data stream, the method further includes: Extract multiple indicator parameters from the health status type information; The weights of each indicator parameter are calculated based on the importance judgment matrix constructed by the relationship between the indicators and expert knowledge, and the comprehensive performance index is calculated based on the weights and the parameters of each indicator.

4. The method for low-altitude aircraft condition prediction and health management according to claim 1, characterized in that, After acquiring the sensor simulation data streams corresponding to the multiple subsystems of the low-altitude aircraft, the method further includes: The sensor simulation data stream is preprocessed to obtain a preprocessed data stream; wherein the preprocessing operation includes at least one of data cleaning, outlier removal, and standardization.

5. The method for low-altitude aircraft status prediction and health management according to claim 1, characterized in that, After obtaining the maintenance decision result, the method further includes: The comprehensive performance index corresponding to the health status type information, the failure severity index corresponding to the comprehensive analysis result output by the hybrid diagnostic model cluster, and the maintenance decision result are input into the management module based on the B / S architecture to perform at least one of the following operations: Interactive management of corresponding functions in each subsystem; Visual presentation.

6. A low-altitude aircraft status prediction and health management system, characterized in that, include: The data stream acquisition module is used to acquire sensor simulation data streams corresponding to multiple subsystems of the low-altitude aircraft. A health assessment module is used to predict the remaining lifetime of the target subsystem based on the sensor simulation data stream, obtain the remaining lifetime prediction time, and confirm the health status type information of the sensor simulation data stream; wherein, the health status type information is used to characterize the health level or abnormality level of the sensor simulation data stream, and the target subsystem belongs to the multiple subsystems; The diagnosis and prediction module is used to perform a comprehensive analysis of the health status of the sensor simulation data stream based on the health status type information using a target diagnosis model, and obtain a comprehensive analysis result; wherein, the target diagnosis model includes one of a hybrid diagnosis model cluster and a hybrid prediction model, the hybrid diagnosis model cluster is used to perform fault analysis on abnormal data streams, and the hybrid prediction model is used to perform fault prediction on healthy data streams; The first decision module is used to make maintenance decisions for the target subsystem based on the remaining life prediction time and the comprehensive analysis results, and to obtain maintenance decision results.

7. The low-altitude aircraft state prediction and health management system according to claim 6, characterized in that, The target diagnostic model is a cluster of hybrid diagnostic models; the system also includes: The second decision module is used to query the fault cause information corresponding to the comprehensive analysis results using a fault mode analysis table built based on FMEA after the comprehensive analysis results are obtained. The failure severity index is calculated based on the fault cause information.

8. The low-altitude aircraft status prediction and health management system according to claim 6, characterized in that, The system also includes: The third decision module is used to extract multiple indicator parameters from the health status type information after confirming the health status type information of the sensor simulation data stream. The weights of each indicator parameter are calculated based on the importance judgment matrix constructed by the relationship between the indicators and expert knowledge, and the comprehensive performance index is calculated based on the weights and the parameters of each indicator.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the low-altitude aircraft status prediction and health management method as described in any one of claims 1 to 5.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the low-altitude aircraft state prediction and health management method as described in any one of claims 1 to 5.

Images