General aero-engine fault simulation and architecture management platform and method

Abstract

The invention provides a general aero-engine fault simulation and architecture management platform and method. The method comprises the following steps: defining a fault diagnosis requirement of an aero-engine system, subdividing the aero-engine system into a plurality of subsystems, performing functional analysis and behavior modeling on each subsystem, and constructing an aero-engine system architecture; establishing an aero-engine system physical simulation model, establishing a link between the system physical simulation model and the aero-engine system architecture, associating each physical simulation model with a fault diagnosis demand, and deploying the associated aero-engine system physical simulation model to a target hardware platform to perform a simulation test; and inputting a simulation test result output by the aero-engine system physical simulation model into the aero-engine fault diagnosis model to obtain an aero-engine fault diagnosis result. The aero-engine fault diagnosis and isolation efficiency and accuracy are improved, the maintenance cost is reduced, and the flight safety is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engine simulation, testing and maintenance, and in particular to a universal aircraft engine fault simulation and architecture management platform and method. Background Art

[0002] As the core power source of aircraft, the reliability and safety of aircraft engine operation are directly related to the safety of the entire flight. Given the complexity of the engine structure and the harshness of the operating environment, various faults are difficult to avoid. These faults will not only weaken the engine performance, but may also cause serious safety accidents. During operation, aircraft engines may encounter a variety of faults, including but not limited to sensor failure, gas path failure, compressor surge, combustion chamber atomization failure, and tail nozzle thrust reduction. If these potential faults are not monitored and handled in a timely manner, they will pose a major risk to flight safety.

[0003] In the past, aircraft engine fault diagnosis mainly relied on technicians' experience and ground testing, but these methods were inefficient and difficult to fully cover all possible failure modes. With the continuous advancement of computer technology, model-based simulation methods have gradually emerged and become an important means of studying aircraft engine failures. As a powerful simulation tool, Matlab / Simulink can easily build aircraft engine models and simulate a variety of failure scenarios, providing strong technical support for fault diagnosis and prediction.

[0004] However, most of the existing research focuses on the fault diagnosis of specific types of faults or specific components, lacking a general and systematic aero-engine fault simulation and architecture management platform, which cannot meet the needs of different types of engines and different fault diagnosis algorithms. In recent years, with the widespread application of model-based system engineering (MBSE) methods in the field of aero-engine architecture design, the combination of MBSE ideas and Matlab / Simulink simulation can further improve the efficiency and accuracy of the aero-engine fault simulation and architecture management platform. The MBSE method can run through the entire product development cycle by building a model-centric traceable information integration framework, thereby providing a strong platform support for the forward design of complex products. Applying the MBSE idea to the aero-engine fault simulation and architecture management platform can achieve rapid construction and verification of fault models, as well as optimization and iterative design of the architecture.

[0005] In summary, the research on general-purpose aircraft engine fault simulation and architecture management platform has important theoretical significance and engineering practical value. Summary of the invention

[0006] In view of this, the embodiments of the present application provide a universal aircraft engine fault simulation and architecture management platform and method to meet the needs of different types of engines and different fault diagnosis algorithms, so as to improve the efficiency and accuracy of aircraft engine fault diagnosis and isolation, reduce maintenance costs, and ensure flight safety.

[0007] The embodiment of the present application provides the following technical solution: a general-purpose aircraft engine fault simulation and architecture management method, comprising:

[0008] Define the fault diagnosis requirements of the aircraft engine system, subdivide the aircraft engine system into multiple subsystems, perform functional analysis and behavior modeling on each subsystem, and build the aircraft engine system architecture;

[0009] Establishing a physical simulation model of an aircraft engine system, establishing a link between the physical simulation model of the aircraft engine system and the aircraft engine system architecture, associating each physical simulation model with the fault diagnosis requirement, and deploying the associated physical simulation model of the aircraft engine system on a target hardware platform for simulation testing;

[0010] The simulation test results output by the aero-engine system physical simulation model are input into an aero-engine fault diagnosis model to obtain an aero-engine fault diagnosis result.

[0011] According to an embodiment of the present application, an aircraft engine system architecture is constructed, including:

[0012] Quantify the key parameters of the engine and establish a demand tracking matrix. The key parameters include: engine performance indicators, safety standards, and maintenance intervals;

[0013] According to the structural composition and functional division of the aircraft engine system, the engine system is divided into multiple subsystems;

[0014] Functional analysis and behavioral modeling are performed on each of the subsystems, and logical relationships between functions are established, and the aircraft engine system architecture is constructed based on distributed design principles.

[0015] According to an embodiment of the present application, after associating each physical simulation model with the fault diagnosis requirement, the method further includes:

[0016] Generate a test case set, establish a mapping relationship between the diagnostic requirements of different engine fault types and the test cases according to the requirement tracking matrix, obtain the simulation test results corresponding to each test case based on the mapping relationship, and verify and confirm the associated physical simulation model of the aircraft engine system according to the simulation test results and the execution of the test cases.

[0017] According to one embodiment of the present application, the physical simulation model of the aircraft engine system is established using the Simscape modeling tool under Simulink.

[0018] According to one embodiment of the present application, the aircraft engine fault diagnosis model is an adaptive neural network fault diagnosis model constructed by integrating the Kalman filter algorithm and the LSTM deep learning network.

[0019] According to an embodiment of the present application, the method further includes: updating and optimizing the physical simulation model of the aircraft engine system according to changes in the fault diagnosis requirements, the simulation test results and the aircraft engine fault diagnosis results.

[0020] According to one embodiment of the present application, the method further includes: classifying the faults and abnormal operating conditions in the aircraft engine fault diagnosis results according to the degree of impact, and sending them to aircraft equipment in the form of graded alarms; and taking corresponding fault countermeasures based on the fault diagnosis results.

[0021] The present application also provides a general-purpose aircraft engine fault simulation and architecture management platform, including:

[0022] The model building and architecture management module is used to define the fault diagnosis requirements of the aircraft engine system, subdivide the aircraft engine system into multiple subsystems, perform functional analysis and behavior modeling on each of the subsystems, and build the aircraft engine system architecture;

[0023] The model building and architecture management module is further used to establish a physical simulation model of the aircraft engine system, establish a link between the physical simulation model of the aircraft engine system and the aircraft engine system architecture, associate each physical simulation model with the fault diagnosis requirement, and deploy the associated physical simulation model of the aircraft engine system on a target hardware platform for simulation testing;

[0024] The fault diagnosis algorithm module is used to input the simulation test results output by the physical simulation model of the aircraft engine system into the aircraft engine fault diagnosis model to obtain the aircraft engine fault diagnosis results.

[0025] According to one embodiment of the present application, the model construction and architecture management module is also used to generate a test case set after each physical simulation model is associated with the fault diagnosis requirement, establish a mapping relationship between the diagnostic requirements and test cases of different engine fault types according to the requirement tracking matrix, obtain the simulation test results corresponding to each test case based on the mapping relationship, and verify and confirm the associated physical simulation model of the aircraft engine system according to the simulation test results and the execution of the test case.

[0026] According to one embodiment of the present application, the fault diagnosis algorithm module is also used to classify the faults and abnormal operating conditions in the aircraft engine fault diagnosis results according to the degree of impact, and send them to the aircraft equipment in the form of graded alarms; and take corresponding fault countermeasures according to the fault diagnosis results.

[0027] Compared with the prior art, the beneficial effects that can be achieved by at least one of the above technical solutions adopted in the embodiments of this specification include at least:

[0028] (1) Systematic model and architecture construction:

[0029] The embodiment of the present invention is based on the MBSE design concept, and the method can be applied throughout the entire life cycle of an aircraft engine from the scheme design stage to the service stage. Through Matlab's Simulink modeling tool and System Composer architecture tool, the efficient construction of an aircraft engine simulation model and system architecture is achieved. This systematic approach not only improves the accuracy and efficiency of model design, but also adopts the distributed architecture design principle to subdivide the engine system into multiple subsystems, improves the readability, development efficiency, traceability, testability and reusability of the model, and reduces the complexity of model development and maintenance.

[0030] (2) General and Scalable MBD Algorithm Strategy:

[0031] The embodiment of the present invention integrates a variety of MBD algorithm strategies, and uses Matlab's signal processing and machine learning toolboxes to facilitate platform algorithm expansion, and can develop new aircraft engine fault diagnosis algorithms that can monitor engine status in real time, warn of potential problems, and achieve more accurate fault diagnosis and more accurate fault isolation. The simulation model can adapt to different engine models, diagnostic requirements, and operating conditions by calling algorithms, adjusting parameters, and configurations. At the same time, the simulation model is easy to adjust and optimize, which can shorten the diagnostic isolation cycle, reduce downtime and maintenance costs.

[0032] (3) Comprehensive simulation and verification process:

[0033] In the model simulation and verification stage, the embodiment of the present invention adopts a comprehensive physical simulation process, and uses the Simscape modeling tool under Simulink to perform multi-physics field simulation on the aircraft engine to verify the system function, performance, reliability and other indicators. At the same time, by generating test cases, automatically executing simulation tests and recording test results, the stability and reliability of the system under different working conditions and fault types are ensured. This process provides strong support for the design and verification of aircraft engines.

[0034] (4) Flexible architecture management and update mechanism:

[0035] The embodiment of the present invention shows great flexibility in architecture management and updating. System Composer is used to manage the model architecture, including resource management, performance evaluation, and modeling standard checking, to ensure the standardization and consistency of the model and support model reuse and sharing. In addition, according to changes in system requirements, the platform can quickly update and optimize the model, and ensure that the updated model meets the requirements through re-simulation and testing. This mechanism provides a strong guarantee for the long-term use of aircraft engine models and the development of new models. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a schematic diagram of a general-purpose aircraft engine fault simulation and architecture management method flow chart according to an embodiment of the present invention;

[0038] Figure 2 1. It is a schematic diagram of the architecture of a general-purpose aircraft engine fault simulation and architecture management platform according to an embodiment of the present invention;

[0039] Figure 3 It is a flow chart of MBD algorithm design that integrates Kalman filtering algorithm and LSTM deep learning network proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0041] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0042] like Figure 1 As shown, an embodiment of the present invention provides a general-purpose aircraft engine fault simulation and architecture management method, including:

[0043] S101. Define the fault diagnosis requirements of the aircraft engine system, subdivide the aircraft engine system into multiple subsystems, perform functional analysis and behavior modeling on each of the subsystems, and construct the aircraft engine system architecture;

[0044] S102. Establishing a physical simulation model of an aircraft engine system, establishing a link between the physical simulation model of the aircraft engine system and the aircraft engine system architecture, and associating each physical simulation model with the fault diagnosis requirement, and deploying the associated physical simulation model of the aircraft engine system on a target hardware platform for simulation testing;

[0045] S103. Input the simulation test results output by the aero-engine system physical simulation model into an aero-engine fault diagnosis model to obtain an aero-engine fault diagnosis result.

[0046] The embodiment of the present invention is based on the Matlab development environment, and adopts an engine simulation method based on a physical model, a system architecture design concept based on MBSE, and an MBD algorithm strategy. Based on the MBSE design concept and using the System Composer architecture tool, it will be responsible for building, storing, and managing the simulation model and system architecture of the aircraft engine; using Matlab's Simulink modeling tool, the engine simulation model is efficiently designed and optimized to simulate its operation under different working conditions. At the same time, it also supports redundant configuration and allocation of software and hardware resources at the micro and macro levels of the system, builds a system architecture and flexibly configures and manages it to ensure the reusability and scalability of the model, and realizes a comprehensive simulation of potential faults of aircraft engines and efficient management of the engine system architecture, aiming to improve the accuracy and effectiveness of engine fault diagnosis and isolation, shorten maintenance cycles, reduce maintenance costs, and ensure flight safety.

[0047] In specific implementation, S101 includes: defining the requirements of the aero-engine system and the overall architecture construction, aiming to solve the design and planning problems of the model, ensure that the model can truly reflect the needs of users, and provide a basis for subsequent model-based product development and simulation. First, the engine's performance indicators, safety standards, maintenance intervals and other key parameters are quantified in detail through the requirements management tool, and a requirements tracking matrix is ​​established to ensure the consistency and traceability of the requirements. Then, according to the structural composition and functions of the system, the engine system is subdivided into subsystems such as the combustion chamber, turbine, and compressor to achieve modular management and fine analysis. Next, the physical model is established using the system modeling language, the functional performance and faults of the subsystem are deeply analyzed, and the interface relationship between functions is established to ensure the coordination of the internal functions of the system and the integration of the structure. Finally, based on the distributed architecture design principles, a system architecture including data channels, communication protocols, and synchronization mechanisms between hardware, software, and components is constructed to improve the readability, development efficiency, traceability, testability, and reusability of the system.

[0048] In the specific implementation, S102 includes: establishing an accurate physical simulation model of the aircraft engine system, simulating the behavior of the engine at different working stages, and covering a variety of fault injection methods and fault scenarios by developing test cases, automating the simulation test and evaluation system's fault response, aiming to solve the verification and confirmation problems of the model and ensure that the model can accurately reflect the system's faults and functional performance. First, the aircraft engine is divided into various functional subsystems, and its key indicators such as function, performance and reliability are constructed through physical modeling to ensure compliance with predetermined standards. The simulation process can simulate the operating conditions of the engine under different operating conditions, which is convenient for identifying and solving possible faults. Subsequently, the physical model is connected to the system architecture to ensure the consistency of the model with the requirements, so as to facilitate the rapid update of the model when the requirements change. It also includes generating a test case set, establishing a mapping relationship between the diagnostic requirements and test cases of different engine fault types according to the requirement tracking matrix, obtaining the simulation test results corresponding to each test case based on the mapping relationship, and verifying and confirming the physical simulation model of the aircraft engine system after association according to the simulation test results and the execution of the test case. A series of test cases are developed according to the test requirements, covering a variety of fault injection methods and work task stages, and automating the simulation test and evaluation system's response and recording the test results. Finally, the system verifies the satisfaction of design requirements and the stable performance of aircraft engines under different operating conditions, including thrust output, fuel efficiency and emission standards, through simulation operation results, thereby improving the reliability and stability of the system and laying the foundation for practical application.

[0049] In some embodiments, S102 also includes: model implementation and deployment. This stage involves the software implementation of the model and its final deployment to the target hardware platform, aiming to solve the executability and practical application problems of the model, ensuring that the model can effectively run in the target environment and achieve the expected functions. First, the platform performs software architecture design, clarifies components, interfaces and connectors, and ensures that the software system structure is clear and easy to maintain. Then, a code conversion tool is used to automatically generate C code or other target language code from the simulation model, and integrate and test it to ensure that the software functions normally. Finally, the platform deploys the verified and tested software system to the target hardware platform to realize the practical application of the model.

[0050] In some embodiments, S102 also includes: architecture management and update. This stage involves the architecture management and model update of the platform, including resource management, performance analysis, design solutions, maintenance suggestions, etc., to ensure the standardization and accuracy of the model, and to update and optimize it according to changes in system requirements, simulation tests and fault diagnosis results, aiming to solve the sustainability and maintainability of the model, ensure the standardization and consistency of the model, and update and optimize it according to changes in system requirements, so as to provide guarantee for the long-term use and development of the model. The platform uses the System Composer tool to comprehensively manage the system architecture, covering resource allocation, performance analysis and modeling specification verification, etc., to ensure the standardization and uniformity of the model system. In response to the evolution of system requirements, the platform has the ability to optimize and iteratively upgrade the model in a targeted manner, and ensure that the adjusted model meets the latest requirements by running the simulation and testing process again. In addition, the platform has also established a model version management mechanism to facilitate tracking of model change history and backtracking, and provide support for continuous improvement of the model.

[0051] In specific implementation, the aircraft engine fault diagnosis model in S103 can support and provide a variety of aircraft engine fault diagnosis (MBD) algorithms based on simulation models, including but not limited to Kalman filtering, neural networks, support vector machines, etc., to achieve rapid detection, accurate isolation and effective prediction of aircraft engine faults, and provide reference for diagnostic demand updates, software system deployment and architecture optimization within the platform. Users can select appropriate algorithms based on actual needs and engine types, and configure algorithm parameters to adapt to different fault diagnosis needs.

[0052] Specifically, it includes the following functions and features:

[0053] (1) Algorithm support:

[0054] The aircraft engine fault diagnosis model supports multiple mainstream fault diagnosis algorithms such as Kalman filtering, neural networks, support vector machines, etc. Users can select appropriate algorithms according to actual needs and engine types, and configure algorithm parameters, such as the covariance matrix of the Kalman filter, the number of hidden layers and nodes of the neural network, to adapt to different fault diagnosis scenarios.

[0055] (2) Fault diagnosis:

[0056] The aircraft engine fault diagnosis model receives sensor data from the engine model, including parameters such as temperature, pressure, vibration, and speed, as input for fault diagnosis. Then, the input data is processed using the selected algorithm, its characteristics are analyzed, and it is determined whether the engine has a fault. If a fault is detected, the module will further analyze the data to determine the component or subsystem where the fault occurred and achieve fault isolation. At the same time, based on historical data and current status, possible future engine faults can be predicted to provide a basis for preventive maintenance.

[0057] (3) Classification warning and fault countermeasures:

[0058] The diagnosed faults and abnormal operating conditions are classified according to the degree of impact and sent to the aircraft equipment in the form of graded alarms. The pilots and aircraft equipment will deal with them according to the situation. According to the diagnosis results, corresponding fault countermeasures are taken, such as channel switching, parameter optimization, emergency shutdown and maintenance operations, to ensure the safe operation of the engine.

[0059] The embodiment of the present invention aims at the problems existing in the current fault diagnosis of aircraft engines, such as excessive reliance on experience, lack of objective parameter data evolution law, low test efficiency, incomplete coverage of fault modes and lack of general platform. The embodiment of the present invention is based on the Matlab development environment, and integrates the simulation technology based on physical models, the model-based system engineering (MBSE) architecture design concept, and the model-based fault diagnosis (MBD) algorithm strategy to achieve comprehensive simulation of aircraft engine fault behavior and flexible management of system architecture. The method mainly includes two main processes: model construction and architecture management and fault diagnosis; model construction and architecture management is to use Matlab's Simulink modeling tool and System Composer architecture tool to build, optimize and manage aircraft engine simulation models, and support flexible configuration and management of simulation architecture to ensure the universality and scalability of the model. Fault diagnosis includes an integrated MBD strategy, which uses Matlab's signal processing and machine learning toolbox to provide a variety of fault diagnosis algorithms to achieve rapid detection, accurate isolation and effective prediction of aircraft engine faults, while supporting algorithm expansion and customization to improve the adaptability of platform fault diagnosis algorithms. Theoretical analysis shows that the platform can help improve the efficiency and accuracy of aircraft engine fault diagnosis and isolation, reduce maintenance costs, ensure flight safety, and promote the development of model-based aircraft engine fault diagnosis technology.

[0060] like Figure 2 As shown, the embodiment of the present invention also provides a general-purpose aircraft engine fault simulation and architecture management platform, including:

[0061] The model building and architecture management module is used to define the fault diagnosis requirements of the aircraft engine system, subdivide the aircraft engine system into multiple subsystems, perform functional analysis and behavior modeling on each of the subsystems, and build the aircraft engine system architecture;

[0062] The model building and architecture management module is further used to establish a physical simulation model of the aircraft engine system, establish a link between the physical simulation model of the aircraft engine system and the aircraft engine system architecture, associate each physical simulation model with the fault diagnosis requirement, and deploy the associated physical simulation model of the aircraft engine system on a target hardware platform for simulation testing;

[0063] The fault diagnosis algorithm module is used to input the simulation test results output by the physical simulation model of the aircraft engine system into the aircraft engine fault diagnosis model to obtain the aircraft engine fault diagnosis results.

[0064] A universal aircraft engine fault simulation and architecture management platform according to an embodiment of the present invention includes two core modules: a model construction and architecture management module and a fault diagnosis algorithm module. The model construction and architecture management module is responsible for building, optimizing and managing the aircraft engine simulation model and system architecture, while the fault diagnosis algorithm module provides a variety of MBD algorithms to achieve rapid detection, accurate isolation and effective prediction of aircraft engine faults, providing technical support for the research and development, testing and maintenance of aircraft engines.

[0065] In specific implementation, the model building and architecture management module mainly includes:

[0066] (1) Requirements analysis and architecture:

[0067] Define system requirements: Use the Requirements Toolbox to itemize and quantify the requirements of stakeholders of aircraft engines, including engine performance indicators, safety standards, maintenance intervals, etc. Establish a requirements tracking matrix for verification, validation, and change control of test requirements.

[0068] System decomposition: Based on the design characteristics and functional divisions of aircraft engines, aircraft engine systems are decomposed into smaller and more manageable components and subsystems, such as combustion chambers, turbines, compressors, fuel systems, control systems, etc., to improve readability, concurrent development, traceability, unit testing, and reusability.

[0069] Functional analysis and behavioral modeling: Use System Composer's use case diagram, activity diagram, state machine diagram and other models to conduct detailed analysis and behavioral modeling of each subsystem's functions, and establish logical relationships between functions to ensure the overall coordination of the system. For example, analyze the fuel combustion process in the combustion chamber, the rotational dynamics of the turbine, and the compression efficiency of the compressor.

[0070] System architecture design: Use System Composer's block diagram, state diagram, sequence diagram and other models to design the system architecture of aircraft engines, including hardware (such as engine body, sensors, actuators), software (such as control algorithms, monitoring programs), communication networks (such as data buses, wireless communications), etc. Establish the connection relationship between independent components, the hierarchical relationship and interaction of system functions, the hierarchical relationship between structural elements and the mapping relationship between structure, function and structure, and form a distributed system architecture.

[0071] (2) Model simulation and verification:

[0072] Model simulation: Use Simscape modeling tools under Simulink to simulate the electrical, chemical, material, mechanical, and other physical systems of the aircraft engine model, and verify whether the system function, performance, reliability, testability, and other indicators meet the requirements. Through simulation, the operating status of the engine can be observed intuitively and potential problems such as incomplete combustion, reduced turbine efficiency, and compressor surge can be discovered.

[0073] Model and requirement allocation: Based on the port and interface allocation of the architecture components, the Simulink physical model is linked to the System Composer architecture, which facilitates direct access and running of simulations at the architecture level. At the same time, each simulation model is associated with the requirements to achieve traceability of the requirements and quickly locate and update related models when the requirements change.

[0074] Test case generation: Use Simulink Test Manager to generate a test case set. According to the requirement tracking matrix, establish a mapping between the diagnostic requirements and test cases of different engine fault types (drift fault, out-of-range fault, correlation fault, etc.), track the test results and evaluate the test coverage. The test case set will include engine operation in different mission stages (takeoff, cruise, landing, etc.), system response in different fault modes (sensor fault, actuator fault, component wear, etc.). Automatically execute the simulation test set, and the recorded test results will be used for the platform's fault diagnosis algorithm analysis.

[0075] System verification and validation: Use the Simulink Check toolbox to verify and validate the system based on simulation results and test case execution. Ensure that the system meets all requirements and has stable and reliable performance, such as engine thrust output, fuel efficiency, and emission standards.

[0076] (3) Model implementation and deployment:

[0077] Software architecture design: Use System Composer to design the aircraft engine software architecture, clarify elements such as components, interfaces and connectors, and ensure that the software system has a clear structure and is easy to maintain. The software architecture includes engine control systems, fault diagnosis systems, data acquisition and monitoring systems, etc.

[0078] Code generation and platform integration: Use the Simulink Coder toolbox to automatically generate C code or other target language code from the Simulink model. Integrate the generated code into the software system for testing to ensure that the software functions properly, and deploy the verified and tested software system to the target hardware platform, such as engine control unit, sensor node, etc.

[0079] (4) Architecture management and update:

[0080] Architecture management: Use System Composer to manage resources such as model architecture development, requirements management, engineering management, data management, performance evaluation, modeling standard checks, and implement impact and coverage analysis. Determine the best architecture design solution through redundancy methods and system architecture evaluation and trade-offs, and generate a detailed system design report, which includes the engine's architecture division, performance analysis, fault diagnosis results, maintenance recommendations, etc.

[0081] Model update: Update and optimize the model based on changes in system requirements, simulation test results, and fault diagnosis reports, and re-simulate and test. For example, with the introduction of new sensors, the model needs to be updated to include these new data sources. Or if the engine model is changed, a completely new simulation and architecture system needs to be established.

[0082] In specific implementation, the fault diagnosis algorithm module mainly includes:

[0083] The engine simulation operation data in the fault test case set is the basis of the MBD algorithm, providing important data support for algorithm training, testing and model verification. The algorithm module includes functions such as aircraft engine electronic controller fault diagnosis, sensor fault diagnosis, actuator fault diagnosis, and engine status monitoring. Figure 3 As shown in the figure, taking the fusion of Kalman filter algorithm and LSTM deep learning network to build an adaptive neural network fault diagnosis model as an example, the implementation process of the MBD algorithm is as follows:

[0084] Data preprocessing: Clean and standardize the engine simulation model data to eliminate the influence of noise and outliers. Organize the data according to time series and select the appropriate time window size for data standardization. According to the fault type and diagnosis requirements, select appropriate features for extraction, such as vibration signals, temperature signals, pressure signals, etc.

[0085] Kalman filter design: Define the state variables and observation variables of the engine and establish the state space model of the engine. Set the state estimation and error covariance matrix of the Kalman filter, use the Kalman filter to perform state estimation on real-time operation data, and analyze the residuals to detect abnormal conditions.

[0086] In the prediction stage, the next state and the corresponding error covariance are predicted based on the state transition model:

[0087]

[0088] in, is the state prediction value of the kth step, is the state estimation value of the k-1th step, A is the state transfer matrix (describing the dynamic behavior of the system), B is the control matrix (describing the influence of the control input on the state), u k is the control input for the kth step.

[0089] P k|k-1 =AP k-1 A T +Q

[0090] Where P k|k-1 is the covariance prediction value of the kth step, P k-1 is the covariance estimate of the k-1th step, and Q is the process noise covariance matrix (describing the uncertainty in the prediction process).

[0091] In the update phase, the newly acquired observation data is used to update the state estimate and the error covariance matrix, thereby continuously optimizing the performance of the filter. The gain calculation of the Kalman filter is:

[0092] K k =P k|k-1 H T (HP k|k-1 H T +R) -1

[0093] Among them, K k is the Kalman gain, H is the observation matrix (describing the relationship between the measurement value and the state), and R is the measurement noise covariance matrix (describing the uncertainty of the measurement value).

[0094] Perform state update and covariance update:

[0095]

[0096] P k =(IK k H)P k|k-1

[0097] Among them, x k Update the state value for step k, y k is the actual measured value of step k, P k is the covariance update value of the kth step, and I is the identity matrix.

[0098] Finally, fault detection and processing are performed by analyzing the residuals:

[0099]

[0100] Among them, r k is the residual, S k is the weighted sum of squares of the residuals, R k is the weight matrix of the residual.k When it is greater than the set threshold, it is judged that a fault has occurred.

[0101] LSTM network construction: Use a deep learning framework (such as TensorFlow or PyTorch) to build an LSTM network. Design the structure of the network and determine the number and connection method of neurons in the input layer, LSTM layer, and output layer. Specify an appropriate activation function (such as tanh or ReLU) for the LSTM layer and select a loss function suitable for classification or regression tasks (such as cross entropy or mean squared error).

[0102] Model training: The engine operation data under normal and fault conditions are used as training sets and validation sets. The training set data is used to train the LSTM network, and the network weights are adjusted through the back propagation algorithm. After the training is completed, the validation set data is used to evaluate the network performance and adjust the network parameters as needed.

[0103] Fusion strategy: Design a weighted fusion strategy based on the confidence of the Kalman filter residual and the confidence of the LSTM network prediction results. Use the residual to evaluate the confidence of the Kalman filter C KF , the probability distribution of the output layer is used to evaluate the confidence C of the LSTM network prediction results LSTM , then the fusion result is expressed as:

[0104] R fused =αC KF +βC LSTM

[0105] Among them, α and β are weight coefficients, and α+β=1. In practical applications, the weight coefficients α and β need to be determined based on experimental data and specific application requirements to achieve the best fault diagnosis effect. Based on the fusion results, the decision logic is designed to judge the state of the engine, including normal, potential fault or clear fault.

[0106] Fault diagnosis: The real-time collected engine simulation operation data is input into the trained LSTM network as a test set. The LSTM network outputs the fault time, fault type (normal, clear fault or potential fault), fault probability, and gives graded warnings and fault countermeasures for clear faults.

[0107] Platform optimization: Based on the diagnostic result report, the platform will further perform operations such as demand update, software deployment and architecture optimization to ensure the continued stable operation of the simulation system and platform optimization and upgrade.

[0108] Based on MBSE ideas and physical simulation technology, the present invention constructs a clear, scalable and easy-to-maintain aircraft engine model framework in the Matlab environment, realizing demand-oriented and change management, modular and hierarchical structure design, standardized and efficient model construction, model traceability and reusability, platform targeted optimization and real-time update and other functions. The designed fault diagnosis algorithm module integrates multiple MBD algorithm strategies to detect, isolate and predict various fault types in the aircraft engine simulation model, realize the scalability of the algorithm module, comprehensive coverage of fault modes, real-time monitoring of engine status and adaptive optimization of the platform, and realize comprehensive simulation of various fault types of aircraft engines and flexible management of system architecture. In summary, the present invention not only provides a standardized and universal construction idea for an aircraft engine fault simulation platform, but also helps to improve the efficiency and accuracy of aircraft engine fault diagnosis, reduce maintenance costs, ensure flight safety, and promote the development of aircraft engine fault diagnosis technology, but also provides comprehensive and solid technical support for the research and development, testing and maintenance of aircraft engines.

[0109] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A general-purpose aircraft engine fault simulation and architecture management method, characterized in that: include: Define the fault diagnosis requirements of the aircraft engine system, subdivide the aircraft engine system into multiple subsystems, perform functional analysis and behavior modeling on each subsystem, and build the aircraft engine system architecture; Establishing a physical simulation model of an aircraft engine system, establishing a link between the physical simulation model of the aircraft engine system and the aircraft engine system architecture, associating each physical simulation model with the fault diagnosis requirement, and deploying the associated physical simulation model of the aircraft engine system on a target hardware platform for simulation testing; The simulation test results output by the aero-engine system physical simulation model are input into an aero-engine fault diagnosis model to obtain an aero-engine fault diagnosis result.

2. The general aviation engine fault simulation and architecture management method according to claim 1, characterized in that: Build aircraft engine system architecture, including: Quantify the key parameters of the engine and establish a demand tracking matrix. The key parameters include: engine performance indicators, safety standards, and maintenance intervals; According to the structural composition and functional division of the aircraft engine system, the engine system is divided into multiple subsystems; Functional analysis and behavioral modeling are performed on each of the subsystems, and logical relationships between functions are established, and the aircraft engine system architecture is constructed based on distributed design principles.

3. The general aviation engine fault simulation and architecture management method according to claim 2, characterized in that: After associating each physical simulation model with the fault diagnosis requirement, the method further includes: Generate a test case set, establish a mapping relationship between the diagnostic requirements of different engine fault types and the test cases according to the requirement tracking matrix, obtain the simulation test results corresponding to each test case based on the mapping relationship, and verify and confirm the associated physical simulation model of the aircraft engine system according to the simulation test results and the execution of the test cases.

4. The general aviation engine fault simulation and architecture management method according to claim 1, characterized in that: The physical simulation model of the aero-engine system is established using the Simscape modeling tool under Simulink.

5. The general aviation engine fault simulation and architecture management method according to claim 1, characterized in that: The aircraft engine fault diagnosis model is an adaptive neural network fault diagnosis model constructed by integrating the Kalman filter algorithm and the LSTM deep learning network.

6. The general aviation engine fault simulation and architecture management method according to claim 1, characterized in that: The method further comprises: The physical simulation model of the aero-engine system is updated and optimized according to the changes in the fault diagnosis requirements, the simulation test results and the aero-engine fault diagnosis results.

7. The general aviation engine fault simulation and architecture management method according to claim 1, characterized in that: The method further comprises: The faults and abnormal operating conditions in the aircraft engine fault diagnosis results are classified into levels according to the degree of impact, and sent to aircraft equipment in the form of graded alarms; and corresponding fault countermeasures are taken according to the fault diagnosis results.

8. A general-purpose aircraft engine fault simulation and architecture management platform, characterized in that: include: The model building and architecture management module is used to define the fault diagnosis requirements of the aircraft engine system, subdivide the aircraft engine system into multiple subsystems, perform functional analysis and behavior modeling on each of the subsystems, and build the aircraft engine system architecture; The model building and architecture management module is further used to establish a physical simulation model of the aircraft engine system, establish a link between the physical simulation model of the aircraft engine system and the aircraft engine system architecture, associate each physical simulation model with the fault diagnosis requirement, and deploy the associated physical simulation model of the aircraft engine system on a target hardware platform for simulation testing; The fault diagnosis algorithm module is used to input the simulation test results output by the physical simulation model of the aircraft engine system into the aircraft engine fault diagnosis model to obtain the aircraft engine fault diagnosis results.

9. The universal aircraft engine fault simulation and architecture management platform according to claim 8, characterized in that: The model building and architecture management module is also used to generate a test case set after each physical simulation model is associated with the fault diagnosis requirement, establish a mapping relationship between the diagnostic requirements and test cases of different engine fault types according to the requirement tracking matrix, obtain the simulation test results corresponding to each test case based on the mapping relationship, and verify and confirm the associated physical simulation model of the aircraft engine system according to the simulation test results and the execution of the test case.

10. The universal aircraft engine fault simulation and architecture management platform according to claim 8, characterized in that: The fault diagnosis algorithm module is further used to classify the faults and abnormal operating conditions in the aircraft engine fault diagnosis results according to the degree of impact, and send them to the aircraft equipment in the form of graded alarms; And take corresponding fault countermeasures based on the fault diagnosis results.

Images