A method and system for generating thermal coupling deicing simulation data

Abstract

The application relates to the technical field of data simulation, and specifically discloses a heat coupling anti-icing simulation data generation method and system. The method comprises the following steps: receiving and configuring simulation parameters; based on a preset round-robin timing configuration, generating control signals for driving multiple independent working partition time sequence work; based on the simulation parameters and the control signals, running the coupled electric heating sub-model and the vibration excitation sub-model to respectively calculate the skin temperature data and the vibration response data; in the simulation process, at least one key model parameter in the electric heating sub-model or the vibration excitation sub-model is dynamically modified according to a preset fault parameter; multiple noise signals conforming to preset statistical characteristics are injected into the data; simulation is performed based on the fault and noise injection to obtain multi-channel simulation data, and the multi-channel simulation data is output as a time-stamped standardized data file. The application can efficiently and controllably generate high-quality simulation data sets with physical mechanism authenticity and measurement disturbance diversity.

Description

Technical Field

[0001] This application relates to the technical field of data simulation, and in particular to a method and system for generating thermally coupled anti-icing and de-icing simulation data. Background Technology

[0002] In the research and development and management of aviation anti-icing systems, obtaining high-quality and diverse real-world fault data has long been a serious challenge. On the one hand, the occurrence of faults in de-icing systems during actual flight is a low-probability event, and for safety reasons, it is impossible to actively and frequently induce various faults in real aircraft to collect data. This results in extremely scarce fault samples under natural conditions, and the acquisition cycle is long and costly. On the other hand, relying solely on limited laboratory bench tests is insufficient to fully reproduce the complex and ever-changing real-world flight conditions (such as extreme temperatures, airflow, and vibration coupling environments). The data obtained often has limitations in fault type coverage, noise accuracy, and operating condition diversity, and cannot fully support subsequent data-driven state assessment and fault prediction model training.

[0003] In the prior art, for example, prior art document CN115723961A proposes an integrated anti-icing device with fault prediction and health management. This device realizes real-time temperature / strain distribution imaging through fiber optic temperature and strain sensors deployed on the surface of the structure, and performs fault diagnosis and health management based on the comparison of real-time data with historical data or thresholds, thereby adjusting the system power control or issuing early warnings.

[0004] However, this technical solution has the following limitations: First, it relies on physical sensors and hardware systems, lacks the ability to generate large-scale fault data flexibly and customizablely, and is difficult to efficiently construct labeled datasets covering multi-dimensional fault modes and evolution processes, which is not conducive to the training and verification of data-driven algorithms (such as deep learning); Second, the system functions focus on online monitoring and real-time control, and are insufficient in terms of the interpretability of fault prediction mechanisms, the ability to reproduce degradation paths, and the synthesis of extreme or rare fault scenarios; Third, although its sensing dimensions include temperature and strain, it cannot achieve comprehensive extraction and analysis of multi-modal fault features. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for generating thermally coupled anti-icing and de-icing simulation data, which partially solves or alleviates the above-mentioned shortcomings in the prior art, and can efficiently and controllably generate high-quality simulation datasets that combine the authenticity of physical mechanisms with the diversity of measurement disturbances.

[0006] To solve the aforementioned technical problems, the present invention specifically adopts the following technical solution:

[0007] A first aspect of the present invention is to provide a method for generating thermally coupled anti-icing and de-icing simulation data, the method comprising:

[0008] S1: Receive and configure simulation parameters, including system geometric parameters, material property parameters, electric heating parameters, vibration excitation parameters, environmental condition parameters, and noise parameters; S2: Based on a preset cycle timing configuration, generate control signals to drive multiple independent working zones to work sequentially in a time-sharing manner, wherein each working zone executes the electric heating stage and the vibration excitation stage sequentially within one working cycle; S3: Based on the simulation parameters and the control signals, run the coupled electric heating sub-model and the vibration excitation sub-model to calculate the skin temperature data and vibration response data respectively; wherein the electric heating sub-model obtains the skin temperature data based on the thermal balance equation, and the vibration excitation sub-model obtains the vibration response data based on the single-degree-of-freedom vibration equation;

[0009] S4: During the simulation, at least one key model parameter in the electric heating sub-model or the vibration excitation sub-model is dynamically modified according to preset fault parameters to simulate system performance degradation or functional failure. The preset fault parameters include fault type, fault occurrence time, fault duration, and degradation evolution mode. The fault type includes failure-type faults and abnormal-type faults. Failure-type faults include open-circuit faults, and abnormal-type faults include faults caused by abnormal parameter changes, such as excessively high / low heating temperature or excessive / insufficient vibration. The degradation evolution mode includes linear evolution mode and exponential evolution mode. When using linear evolution mode, the degree of degradation of the key model parameters is proportional to the number of working cycles, and the degradation process increases uniformly with the number of cycles. When using exponential evolution mode, the degree of degradation of the key model parameters is exponentially related to the number of working cycles, and the degradation process shows slow initial growth followed by accelerated deterioration.

[0010] S5: Inject various noise signals conforming to preset statistical characteristics into the skin temperature data, vibration response data, and corresponding current data, including: injecting Gaussian white noise and quantization error into the skin temperature data; injecting Gaussian white noise, 1 / f low-frequency drift noise, environmental vibration harmonic interference noise, and transient mechanical shock noise into the vibration response data; and injecting power frequency interference noise, Gaussian white noise, quantization error, and slow drift noise into the current data.

[0011] S6: Based on steps S4 and S5, perform simulation to obtain multi-channel simulation data, encapsulate the multi-channel simulation data according to the structure corresponding to multiple working partitions and multiple sensor types, and output it as a standardized data file with timestamps.

[0012] Furthermore, in step S4, a fault simulation is triggered by determining whether the current simulation time is within a preset fault time window. The fault time window is determined based on the fault occurrence time and fault duration configured by the user.

[0013] Furthermore, in step S4, the fault parameters also include a maximum degradation threshold. When the change in the key model parameters calculated through the degradation evolution mode reaches the maximum degradation threshold, the key model parameters are maintained at the threshold level to simulate the fault development to a stable saturation state.

[0014] Furthermore, in step S4, when the fault type is an open circuit fault, the equivalent resistance in the electric heating sub-model is modified to infinity, or the electromagnetic driving force in the vibration excitation sub-model is modified to zero.

[0015] Furthermore, in step S5, when injecting the environmental vibration harmonic interference noise into the vibration response data, one or more sinusoidal signals with preset amplitudes are generated and superimposed according to the preset fundamental frequency and harmonic order to simulate the structural background vibration caused by the operation of the aircraft engine.

[0016] Furthermore, in step S5, the generation of the slow drift noise is simulated using a random walk process, and the drift amount is randomly accumulated within each simulation step according to a preset step standard deviation to simulate the slow time-varying characteristics of the sensor measurement baseline.

[0017] Furthermore, in step S6, the standardized data file is stored in HDF5 format, and the internal data structure of the standardized data file contains at least groups named according to partition numbers. Each group is divided into datasets according to sensor type, and each dataset contains one-dimensional time series data and corresponding attribute information.

[0018] Furthermore, in step S3, the electric heating sub-model and the vibration excitation sub-model run synchronously in simulation time, and the two are time-coupled through a shared simulation step size and partition activation signal, so that the skin temperature data calculated in the electric heating stage can be used as the initial or boundary condition for the simulation operation in the vibration excitation stage within the same partition.

[0019] Furthermore, the configuration of the simulation parameters is accomplished through a graphical user interface, which provides independent input panels for different parameter categories and saves the entire set of parameter settings as a configuration file or loads them from the configuration file to achieve the reproduction and batch management of simulation scenarios.

[0020] Secondly, this application also discloses a thermally coupled anti-icing and de-icing simulation data generation system, the system comprising: a simulation parameter configuration module configured to receive and configure simulation parameters, the simulation parameters including system geometric parameters, material property parameters, electric heating parameters, vibration excitation parameters, environmental condition parameters, and noise parameters; a polling timing configuration module configured to generate control signals for driving multiple independent working sub-divisions to work sequentially in a time-sharing manner based on a preset polling timing configuration, wherein each working sub-division sequentially executes an electric heating stage and a vibration excitation stage within one working cycle; and a data acquisition module, which, based on the simulation parameters and the control signals, runs a coupled electric heating sub-model. The simulation includes two sub-models: an electric heating sub-model and a vibration excitation sub-model. The electric heating sub-model obtains the skin temperature data based on the heat balance equation, while the vibration excitation sub-model obtains the vibration response data based on the single-degree-of-freedom vibration equation. A fault injection module is configured to dynamically modify at least one key model parameter in either the electric heating sub-model or the vibration excitation sub-model during the simulation process, based on preset fault parameters, to simulate system performance degradation or functional failure. The preset fault parameters include fault type, fault occurrence time, fault duration, and degradation evolution mode. The fault type includes failure-type faults and anomaly-type faults. The faults include open-circuit faults and abnormal faults, such as excessively high / low heating temperature or excessive / insufficient vibration caused by abnormal parameter changes. The degradation evolution modes include linear and exponential evolution modes. In the linear evolution mode, the degradation degree of key model parameters is proportional to the number of working cycles, and the degradation process increases uniformly with the number of cycles. In the exponential evolution mode, the degradation degree of key model parameters is exponentially related to the number of working cycles, and the degradation process shows slow initial growth followed by accelerated deterioration. A noise injection module is configured to inject noise into the skin temperature data, vibration response data, and corresponding electrical signals. In the streaming data, various noise signals conforming to preset statistical characteristics are injected, including: Gaussian white noise and quantization error are injected into the skin temperature data; Gaussian white noise, 1 / f low-frequency drift noise, environmental vibration harmonic interference noise, and transient mechanical shock noise are injected into the vibration response data; power frequency interference noise, Gaussian white noise, quantization error, and slow drift noise are injected into the current data; the simulation data generation module is configured to perform simulation based on the fault injection module and the noise injection module to obtain multi-channel simulation data, and encapsulate the multi-channel simulation data according to the structure corresponding to multiple working partitions and multiple sensor types, and output it as a standardized data file with timestamps.

[0021] Beneficial Technical Effects: The high-fidelity mathematical and physical simulation model of the thermo-coupled anti-icing and de-icing system constructed in this application accurately characterizes the heat conduction balance equation of the electric heating subsystem and the single-degree-of-freedom dynamic equation of the vibration excitation subsystem. It deeply integrates multiple physical mechanisms such as material degradation, electrical anomalies, and structural failures, achieving digital mapping and dynamic simulation of the wing anti-icing and de-icing system's operation under almost all conditions. This model not only possesses high modularity and configurable parameters, allowing users to flexibly set geometric dimensions, material properties, environmental conditions, and noise parameters, but also innovatively integrates partitioned time-sequential control, linear and exponential degradation fault injection, and a multi-sensor composite noise model covering Gaussian white noise, power frequency interference, quantization error, and low-frequency drift. This enables the systematic generation of multi-channel simulation data with time-series annotations covering normal, abnormal, and failure states. Simulation results show that the data generated by the model exhibits high consistency with real experimental data in key characteristics such as current transient response, temperature change trends, and vibration acceleration spectrum, verifying the accuracy of its physical mechanism and its engineering applicability.

[0022] This application uses a rigorous mathematical and physical model to simulate the fault mechanism, which has reliable interpretability of the fault mechanism. Furthermore, it incorporates fault degradation modes, making it possible to reproduce the fault degradation path and thus providing more comprehensive and reliable data support for subsequent data analysis.

[0023] This application constructs a virtual data generation platform that combines physical realism, functional completeness, and user-friendliness. It can generate high-quality fault datasets for training and validating machine learning algorithms in batches at extremely low cost and in a repeatable manner. This fundamentally solves or greatly alleviates the common problems of scarce fault samples, high risks in actual testing, and difficulties in data annotation for high-reliability systems in the aerospace field. It provides a solid data foundation and an efficient verification environment for the research and development and application of advanced fault prediction and health management technologies.

[0024] Furthermore, compared to the existing technology that uses a combination of hardware and software to simulate faults and generate fault data, the solution in this application consumes significantly less computing resources; at the same time, by adjusting the input parameters, various fault state data can be generated, which greatly reduces the cost and cycle of data acquisition. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. The elements or parts in the drawings are not necessarily drawn to scale. Obviously, the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0026] Figure 1 This is a flowchart of a thermally coupled anti-icing and de-icing simulation data generation method according to this application.

[0027] Figure 2 This is a diagram of the overall architecture of the simulation model in one embodiment of this application.

[0028] Figure 3 This is a schematic diagram of a partitioned round-robin control system in one embodiment of this application.

[0029] Figure 4 This is a schematic diagram of a fault control system in one embodiment of this application.

[0030] Figure 5 This is a schematic diagram of the working partition in one embodiment of this application.

[0031] Figure 6 This is a schematic diagram of the internal structure of the electric heating system in one embodiment of this application.

[0032] Figure 7 This is a schematic diagram of the internal structure of the heating core functional module in one embodiment of this application.

[0033] Figure 8 This is a schematic diagram of the internal structure of the fault injection module in one embodiment of this application.

[0034] Figure 9 This is a schematic diagram of a resistor degradation fault simulation module in one embodiment of this application.

[0035] Figure 10 This is a schematic diagram of a voltage fault simulation module in one embodiment of this application.

[0036] Figure 11 This is a schematic diagram of the internal structure of the vibration excitation system in one embodiment of this application.

[0037] Figure 12 This is a schematic diagram of the internal structure of the vibration core functional module in one embodiment of this application.

[0038] Figure 13 This is a schematic diagram of the internal structure of the fault injection module in one embodiment of this application.

[0039] Figure 14 This is a schematic diagram of the internal structure of the voltage fault simulation module in one embodiment of this application.

[0040] Figure 15 This is a schematic diagram of a resistor degradation fault simulation module in one embodiment of this application.

[0041] Figure 16 This is a schematic diagram of a current control system in one embodiment of this application.

[0042] Figure 17 This is a schematic diagram of the module structure of a thermally coupled anti-icing and de-icing simulation data generation system in one embodiment of this application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In this document, suffixes such as "module," "component," or "unit" used to represent elements are only for the purpose of illustrative purposes and have no specific meaning in themselves. Therefore, "module," "component," or "unit" can be used interchangeably. In this document, the terms "upper," "lower," "inner," "outer," "front," "rear," "one end," "the other end," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. In this document, unless otherwise expressly specified and limited, the terms "installed," "equipped with," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. "And / or" in this document includes any and all combinations of one or more of the listed related items. "A plurality" in this document means two or more, i.e., it includes two, three, four, five, etc.

[0044] Figure 1A flowchart of a thermally coupled anti-icing and de-icing simulation data generation method according to this application is shown. (Refer to...) Figure 1 The method specifically includes the following steps: S1: Receive and configure simulation parameters, including system geometric parameters, material property parameters, electric heating parameters, vibration excitation parameters, environmental condition parameters, and noise parameters. S2: Based on a preset polling timing configuration, generate control signals to drive multiple independent working sub-divisions to work sequentially at different times, wherein each working sub-division executes the electric heating stage and the vibration excitation stage sequentially within one working cycle. S3: Based on the simulation parameters and the control signals, run the coupled electric heating sub-model and the vibration excitation sub-model to calculate the skin temperature data and vibration response data, respectively. The electric heating sub-model obtains the skin temperature data based on the thermal balance equation, and the vibration excitation sub-model obtains the vibration response data based on the single-degree-of-freedom vibration equation. S4: During the simulation process, dynamically modify at least one key model parameter in the electric heating sub-model or the vibration excitation sub-model according to preset fault parameters to simulate system performance degradation or functional failure.

[0045] The preset fault parameters include fault type, fault occurrence time, fault duration, and degradation evolution mode. S5: Inject various noise signals conforming to preset statistical characteristics into the skin temperature data, vibration response data, and corresponding current data, including: injecting Gaussian white noise and quantization error into the skin temperature data; injecting Gaussian white noise, 1 / f low-frequency drift noise, environmental vibration harmonic interference noise, and transient mechanical shock noise into the vibration response data; and injecting power frequency interference noise, Gaussian white noise, quantization error, and slow drift noise into the current data. S6: Perform simulation based on steps S4 and S5 to obtain multi-channel simulation data. Encapsulate the multi-channel simulation data according to a structure corresponding to multiple working partitions and various sensor types, and output it as a standardized data file with timestamps.

[0046] The fault types include failure-type faults and abnormal faults. Failure-type faults include open-circuit faults, and abnormal faults include faults caused by abnormal parameter changes, such as excessively high / low heating temperature or excessive / insufficient vibration. The degradation evolution modes include linear evolution modes and exponential evolution modes. When using the linear evolution mode, the degree of degradation of key model parameters is directly proportional to the number of working cycles, and the degradation process shows a uniform increase with the number of cycles. When using the exponential evolution mode, the degree of degradation of key model parameters is exponentially related to the number of working cycles, and the degradation process shows slow initial growth followed by accelerated deterioration.

[0047] In one specific embodiment, the following disclosure is applicable to a specific embodiment of a thermally coupled anti-icing and de-icing simulation data generation method of this application, specifically including the following steps:

[0048] S101. Mathematical and Physical Modeling Assumptions for the Thermally Coupled Anti-icing and De-icing System: 1) System Working Principle: The thermally coupled anti-icing and de-icing system of this invention adopts a collaborative working mechanism of electric heating and vibration excitation (see announcement number CN120716939B for details). The system is divided into multiple independent zones, each zone working in a cyclical manner according to the sequence of "heating first, then vibration": first, electric heating is applied to reduce the adhesion force of the ice layer, and then mechanical vibration is applied to promote the peeling of the ice layer. This zone-by-zone cyclic control method improves energy utilization efficiency and achieves fault isolation. 2) Geometric and Material Simplification: The three-dimensional skin structure is simplified to a two-dimensional flat plate model, ignoring the influence of curvature. It is assumed that the thermal properties (density, thermal conductivity, specific heat capacity, elastic modulus, damping coefficient) of the material (aluminum alloy, PI film, etc.) are constant in the working range of -40℃ to 200℃. The temperature dependence can be adjusted through weak coupling posterior and is not explicitly considered in the model.

[0049] 3) Key parameter assumptions: (1) The heat transfer between the front surface of the skin and the environment (air or ice layer) is mainly convective heat transfer, using a constant h conv (1) Approximate, without explicit analysis of heat transfer changes in turbulent or icing states; (2) The coupling between the vibration exciter and the local skin is simplified to an equivalent mass-damping-stiffness model (not changing with temperature), and the output displacement, velocity, and acceleration responses represent local vibration behavior; (3) The electro-thermal conversion efficiency η of the electric heating film is regarded as a constant value, and its decay can be simulated by adding a drift function when there is a fault or aging. (4) External working condition settings: The ambient temperature is configurable. Ground experiments are usually in the range of -10 to 20℃, and real high-altitude flight can reach -40℃ or lower; The system is driven by pulsed DC voltage / current; The sensor has a preset sampling rate (temperature 1-10Hz, current / vibration 1-10kHz); Integrated overcurrent protection and other Boolean logic protection mechanisms.

[0050] S102, Design of Mathematical and Physical Model for Anti-icing and De-icing: The thermo-coupled anti-icing and de-icing system consists of two subsystems: (1) Electric heating subsystem: The fuselage skin is heated by current to reduce the adhesion between the ice layer and the skin. (2) Vibration excitation subsystem: The vibration exciter is driven by electro-electro-mechanical conversion to apply mechanical vibration to the weakened ice layer, causing it to fall off. The overall idea of ​​this model is: to describe the heating process of the electric heating film on the skin by combining convection heat dissipation with the heat conduction equation; and to describe the current-force conversion of the vibration exciter and its electromechanical response to the skin by the single-degree-of-freedom vibration dynamics equation. At the same time, fault models and noise models are designed for different fault modes and measurement noise / disturbances to generate data that closely resembles the real flight environment.

[0051] 1) Mathematical and Physical Description of the Electro-Heating Model: The skin is considered as a homogeneous flat plate structure with a finite thickness, whose material density, specific heat capacity, and thermal conductivity remain consistent within the plate. Since the skin thickness is typically much smaller than its in-plane dimensions, and the heating film is tightly attached to its inner surface, the temperature gradient along the thickness direction rapidly tends to equilibrium under thermal inertia. Therefore, the skin can be approximated as an equivalent lumped-parameter heat capacity, with its characteristic temperature denoted as T(t), representing the average temperature of the skin in the vertical direction. There is no significant thermal contact resistance between the heating film and the inner surface of the skin, allowing the electro-heating power to be directly input into the skin as a surface heat source. The outer surface of the skin is exposed to the external flow field, undergoing forced convection heat transfer with the environment, forming a continuous heat dissipation process along the outer surface. According to the law of conservation of heat, the electro-heating power added to the block minus the convection power dissipated from the block equals the energy required for the block temperature to rise. ；

[0052] c p Specific heat capacity of the skin material (unit: J / (kg·℃)); Mass of the bulk material: ρ is the density of the metallic material (unit: kg / m³). A is the volume of the skin, A is the area of ​​the skin (unit: m²), and h is the thickness of the skin (unit: m).

[0053] Assuming the equivalent resistance of the heating film is R (unit: Ω) and the current flowing through it is I(t) (unit: A), then the electric heating power generated under Ohm's law is: .

[0054] In reality, not all electrical power results in a temperature rise in the bulk material; some heat is lost through the edges of the heating film or from the bottom surface. A thermal efficiency coefficient η∈(0,1] is introduced to represent this loss.

[0055] Convection cooling refers to the process by which the skin surface carries away heat. Newton's law of cooling is commonly used: h conv (Unit: W / (m²·K)) is the convective heat transfer coefficient, which depends on the external airflow velocity, air temperature, etc.; T ∞ A represents ambient temperature; A still represents the skin area. Let t be the temperature at time t.

[0056] Finally, substituting all the above quantities into the heat conservation formula, we get:

[0057] ；Given an initial temperature T(0) and a current curve I(t), the change of T(t) over time can be calculated step by step through numerical integration; η∈(0,1] is the thermal efficiency coefficient, for example, taken as 0.95, where 5% of the heat is power loss and only 95% of the power is manifested as a temperature rise; The temperature is at the current moment.

[0058] 2) Mathematical and Physical Description of the Vibration Excitation Model: The vibration exciter is made of polyimide (PI) film and copper foil wound together. The PI film serves as the insulating and structural support layer, while the copper foil serves as the conductive functional layer, together forming an integrated composite layered structure. When energized / voltage is applied, it generates a reciprocating force acting on the skin or attached structure, causing the entire structure to vibrate. This creates shear force between the ice layer and the skin, ultimately causing it to detach. Consider it as a mass... -damping -Stiffness A single-degree-of-freedom vibration system. The equation of motion for a single-degree-of-freedom system can be written as: ； x(t) is the vibration displacement, which represents the displacement of the vibrator (or excitation point) in a certain direction; and These are the first and second time derivatives, corresponding to velocity and acceleration, respectively; m is the equivalent mass, c is the damping coefficient, and k is the equivalent stiffness. F(t) is the external excitation force, originating from the current I. v (t). For a common electromagnetic exciter, the current I flowing through it is... v (t) will generate a force proportional to the current in the magnetic field. B is the magnetic flux density, using Perform calculations. is the permeability (unit: H / m), L is the effective length of the vibration exciter in the magnetic field, N is the number of copper foil layers, and d is the interlayer spacing between each copper foil layer.

[0059] Substituting into the equation of motion, we get: ； The left side represents inertia ( ), damping ( The reaction force on the motion consists of three parts: the elastic restoring force (kx), the elastic restoring force (kx), and the reaction force on the motion. The right side represents the input current I. v (t) An external excitation force generated at one end drives the system to move.

[0060] 3) Fault type and parameter changes:

[0061] (1) Fault types of electric heating film: Normal operation: During the system design phase, the normal operating current parameter has been preset, denoted as I no m(Normal operating current). In the system simulation environment, the heating module follows preset control logic, operating for a specified duration according to a predetermined timing sequence within each simulation cycle. During the simulation, the system temperature changes according to the following pattern: Initially, the system temperature is the preset initial ambient temperature T. e (Initial temperature); During the operation of the heating module, the system temperature changes from T... e The temperature gradually increases; after the heating module reaches its preset operating time and stops operating, the system temperature will gradually decrease to near T as the environment dissipates heat. e The stable range.

[0062] Excessive temperature: During system operation, the heating film may experience an abnormal decrease in its equivalent resistance R due to localized changes in material properties or deterioration in heat dissipation. In modeling, R can be expressed as a function that changes with time t, for example: ;in This refers to the rate of change of resistance over time (positive for increased resistance, negative for decreased resistance). In the event of an overheating fault, the resistance of the electric heating film may decrease by a certain percentage due to localized changes in material properties or other reasons. For example, if the resistance decreases by 30%, then... ; This represents the initial resistance. Under these conditions, with the same voltage or current input, the actual power generated by the heating film will exceed the design range, or the temperature will rise rapidly with a slight increase in current. Data shows that even when the control current is within the normal range during operation, the temperature in this zone is still significantly higher than expected; or the current required to maintain the target temperature is lower than normal, while both the rate of temperature rise and the steady-state temperature are higher than the design level.

[0063] Low temperature: Over time, the heating film material may age, causing a change in its equivalent resistance R. In the model, R can be set to gradually increase with the usage time t, for example: ;in The current increases slowly over time, and is taken as a positive value. This means that with the same voltage or current input, less heating power is generated, or a larger current is needed to reach the original temperature. In terms of data, to maintain the temperature rise during use, the current needs to gradually increase beyond the normal value; or the current remains constant, but the temperature rises more and more slowly, and the steady-state temperature is lower than before.

[0064] Open circuit: An open circuit fault occurs when the heating film completely loses its conductive path due to material breakage, connection detachment, or circuit interruption. At this time, the equivalent resistance... It can be considered to approach infinity. In the model, it can be described as: In this state, no voltage is applied, no current can be generated, and the heating power is zero. Sensor data shows that the current sensor reads zero for this zone, while the temperature sensor collects the same temperature value as the ambient temperature, which remains unchanged regardless of control commands. The system completely loses its heating capability in this zone.

[0065] (2) Vibration exciter fault type: Normal operation: The vibration exciter of each zone will work once in each cycle. Each time it will be powered on several times in a short period of time, generating a corresponding number of vibration acceleration peaks.

[0066] Excessive vibration: During the operation of a vibration system, the exciter's equivalent driving force parameters may change due to variations in the performance of internal components or changes in external load conditions. An abnormal increase. The vibration exciter resistance can be adjusted during modeling. Represented as over time A function that gradually decreases, for example: ;in The vibration gradually increases over time. Under these circumstances, with the same input signal conditions, the actual vibration output generated by the exciter will exceed the design range, or the input current required to achieve the target vibration level will be lower than the normal value. Data shows that even when the control signal is within the rated range during operation, the acceleration measured by the vibration sensor in this zone is still significantly higher than expected, and both the vibration response sensitivity and steady-state amplitude exceed the normal design thresholds.

[0067] Insufficient vibration: As the system operates for an extended period, the exciter's equivalent driving force parameters may decrease due to factors such as permanent magnet demagnetization, coil aging, or mechanical obstruction. It gradually decreases. The resistance can be represented in the model. Set to use for a certain period of time A function that gradually increases, for example: ;in The amplitude of vibration acceleration increases slowly over time. This change causes the amplitude of vibration acceleration output by the exciter to decrease under the same input current conditions, or the current required to maintain the target vibration intensity to gradually increase above the normal value. Correspondingly, in sensor data, with the drive current remaining constant, the vibration acceleration increases slowly and the steady-state amplitude is low; to reach the original vibration level, the input current needs to be gradually increased.

[0068] Open circuit: An open circuit fault occurs when the exciter's circuit is completely interrupted due to coil breakage, loose wiring, or power drive failure. At this time, the equivalent driving force... Reducing to zero can be expressed in the model as: In this state, the applied voltage cannot generate an effective current, and the exciter outputs zero vibration. Sensor data shows that the current sensor reading for this zone is zero, and the vibration sensor detects that the vibration acceleration signal remains at the background noise level and does not change with control commands. The system completely loses its vibration output capability in this zone.

[0069] 4) Noise Types: In practical applications, the following types of noise or interference are most common, and they will be superimposed on the three main sensor signals: current, temperature, and vibration: sensor measurement noise, including current sensor noise, temperature sensor noise, and vibration sensor noise; process noise or operating condition disturbance, including ambient temperature fluctuations, airflow convection coefficient disturbances, power supply voltage fluctuations, exciter drive waveform distortion, etc.; fault-specific noise or interference, including open circuit / short circuit switching transients, random drift caused by aging, vibration mechanical friction noise, etc.

[0070] (1) Noise model of current sensor:

[0071] Gaussian white noise: Assuming the measured current I 测 (t) under the real current I 真 A zero-mean Gaussian white noise is superimposed on (t). :

[0072] ； White noise standard deviation σ I It can be determined by the sensor's accuracy, such as 0.5% of full scale.

[0073] Power frequency interference: External power supply interference may occur at a certain frequency. f Coupling at the point of contact generates sinusoidal interference noise:

[0074] ; The amplitude at 50Hz This represents the initial phase corresponding to the power frequency interference frequency;

[0075] An additional 100Hz or 150Hz harmonic term can be added to simulate the residue after non-ideal filtering.

[0076] Quantization error: Assuming an analog-to-digital converter (ADC) has a resolution of 12 bits and a full-scale range of 0–10A, then each LSB (Least Significant Bit, the bit with the smallest weight in the digital encoding) is approximately equivalent to The quantization step. The quantization error can be considered as being uniformly distributed across... Random numbers between .

[0077] Slow drift or bias: Sensor drift, baseline bias, etc., can be represented by a random walk that changes slowly over time.

[0078] ； Where the standard deviation σ b Very small values, such as 0.001 A / hour, exhibit near-linear drift in simulations over 0–100 seconds. As the baseline value, This refers to the drift or bias.

[0079] Adding the above values ​​together, we obtain the final measured current:

[0080]

[0081] (2) Temperature sensor noise model:

[0082] Gaussian white noise: Assuming the temperature sensor has an accuracy of ±0.5°C within the range of -40 to 150°C, a zero-mean normal distribution can be set.

[0083] ; The standard deviation of white noise;

[0084] Ambient temperature fluctuations: The actual ambient temperature will change slowly. A random walk or low-frequency noise can be added to it. ,in, , The standard deviation of ambient temperature noise can be taken as 0.5℃; Let t be the actual ambient temperature. For the random drift duration, This is the drift value.

[0085] Quantization error: If the temperature signal passes through a 12-bit ADC, the range is... 50-150°C, then each quantization step size = Quantization error can be considered as... The uniform distribution between them is denoted as .

[0086] The final measured temperature signal can be written as:

[0087] .

[0088] (3) Noise model of vibration sensor:

[0089] Gaussian white noise: The white noise of an accelerometer can generally be expressed as... ,in Let B be the noise density and B be the effective analog bandwidth. If the sampling frequency B is 1kHz, then approximately the standard deviation will be generated. Gaussian white noise of magnitude [order];

[0090] Low-frequency drift (1 / f noise): Mechanical temperature changes or aging of electronic components can cause a slow drift in the sensor output baseline, commonly represented by a 1 / f spectrum. A 1 / f noise sequence can be generated in the simulation and then superimposed with Gaussian white noise.

[0091] Environmental vibration interference: The aircraft structure itself will introduce certain reference vibrations, such as the 150Hz fundamental frequency and several harmonics caused by engine speed, which can be represented by a single or multiple sine waves:

[0092] ;

[0093] in 1 = 150Hz 2 = 300Hz The reference phase corresponding to the vibration frequency, amplitude At 0.01 0.1 between.

[0094] Mechanical shock noise: Occasional minor structural impacts (such as the impact of tiny stone chips on the fin surface) can be simulated using a series of randomly spaced impact pulses.

[0095] ;

[0096] t j The impact moment is randomly generated, A j It is the impact amplitude, τ j It is the exponential decay coefficient, which simulates the rapid decay of mechanical vibrations after an impact.

[0097] Final vibration sensor output:

[0098] .

[0099] 5) Degradation model:

[0100] By introducing a degradation mechanism into the Simulink simulation model, the performance degradation process of the system changes with the operating cycle, so as to generate more representative time series data to support the remaining lifetime modeling research.

[0101] The main purpose of adding a degradation model to the simulation model is to simulate the performance degradation trend of key system components during long-term operation, so that the generated signal exhibits reasonable degradation changes over time, thus more closely resembling the characteristics of real data.

[0102] This application employs two typical degradation models: linear degradation model and exponential degradation model. In the linear degradation model, the system parameters increase at a constant rate with the number of operating cycles n. T The gradual, degenerate form can be represented as

[0103] ;

[0104] Where P(n) T ) represents the degradation parameter (such as resistance or gain) that changes with the number of runs, P0 is the initial value, and λ is the linear degradation rate. This model is suitable for simulating slow, stable degradation processes, such as uniformly aged conductive film materials.

[0105] In the exponential degradation model, the system performance deteriorates at an accelerated rate, and the degradation parameters follow the following relationship:

[0106] ;

[0107] Where β is the exponential degradation rate factor. This model is more suitable for reflecting degradation behaviors with nonlinear characteristics, such as high-temperature fatigue and material crack propagation. Both models are driven by a running counter and are linked to the time process in the simulation model, facilitating dynamic control.

[0108] In the temperature signal channel, the equivalent resistance R of the electric heating film is set to vary with n. T The increase in current causes the current response to gradually decrease, thus indirectly affecting the heating rate and steady-state temperature. Taking a constant voltage U power supply as an example, the actual operating current over time can be expressed as:

[0109] ;

[0110] This causes the temperature rise process to slow down, resulting in a decrease in thermal efficiency.

[0111] In vibration signal modeling, the output amplitude or damping coefficient of the exciter is adjusted according to... Changes can simulate the attenuation of vibration response due to structural aging; for example, the gain can be modeled as follows:

[0112] ;

[0113] Given the initial current value, the electrical performance change model can be achieved by controlling the relationship between the equivalent load resistance and the excitation current in the current path, and the current degradation trajectory can be constructed in conjunction with the system operation cycle counting.

[0114] S104. Construction of the Anti-icing and De-icing System: The anti-icing and de-icing system model consists of three parts: Top-level system: According to the working cycle set by the user input parameters, the five sections (section 1–5) work in turn in each working cycle. When each section works, the electric heating subsystem and the vibration excitation subsystem are activated in sequence; Electric heating subsystem: Simulates the heating and heat conduction process of the electric heating film, and embeds aging and fault triggering logic; Vibration excitation subsystem: Uses the mass-damping-spring model to simulate the vibration excitation process.

[0115] In some embodiments, the anti-icing and de-icing system disclosed in this application is described in detail below:

[0116] 1) System Overall Design (1) Overall Architecture: The overall architecture of the anti-icing and de-icing system is as follows Figure 2 As shown. The system uses simulation time as the driving source, and generates partition working signals and fault signals with the help of the partitioned round-robin control system and the fault control system. These signals are input to the five partitioned working subsystems to realize the control of the working status of each partition and the fault setting. The simulation data generated by each partition is collected and flows into the data storage system. It is grouped and stored according to the categories of temperature sensor, electric heating current sensor, vibration sensor and vibration current sensor, and finally stored in HDF5 format file. (2) Partitioned round-robin control system: The working logic flow of the partitioned round-robin control module is as follows Figure 3 As shown, its function is to generate working signals for each partition, indicating that the five partitions work in sequence in each cycle. This module receives input parameters such as the current simulation time, working cycle, working time of the electric heating system, current pulse frequency, current pulse count, and current pulse width, and calculates to form the working signal T of the heating partition and the working signal V of the vibration partition. The specific working method of this module is as follows: First, the current simulation time is used to take the remainder of the working cycle to calculate the offset within the cycle; then the offset within the cycle is divided by the working time of a single partition to calculate the working partition; finally, the offset within the cycle is taken as the remainder of the working time of a single partition to obtain the time offset within the partition, and the working system is judged to determine whether the current output is the working signal (T) of the electric heating system or the working signal (V) of the vibration excitation system. (3) Fault control system: The workflow of the fault control module is as follows Figure 4As shown, this module's function is to generate corresponding fault control signals based on user-input fault parameters, controlling the generation of faults in the heating and vibration systems of each zone of the anti-icing and de-icing system. The fault control module can be divided into a heating fault control module and a vibration fault control module. The two modules are consistent in their overall architecture, each independently responsible for the fault state management and control of its corresponding heating / vibration system. Specifically, the heating fault control module is mainly responsible for outputting heating fault type codes to simulate and inject heating system faults; correspondingly, the vibration fault control module outputs vibration fault type codes to control the generation and triggering of vibration system faults. The heating / vibration fault control module performs real-time conditional judgments based on input variables: if the current simulation time is within the fault time and the current zone system has a fault, the corresponding output port will activate and output the specified fault type code, thereby injecting and maintaining the heating / vibration fault in the corresponding zone; conversely, if it is not within the fault time or the current zone system has no fault, the port will always output a logical value representing the system's normal state (set to 1), ensuring that the heating / vibration system is not affected by faults during this period and operates normally under predetermined conditions. (4) Working partitions: The system contains five working partitions, and the structure of a single working partition is as follows: Figure 5 As shown. Each zone receives the following control signals: heating operation signal, vibration operation signal, electric heating fault signal, and vibration fault signal. Each zone consists of a heating module and a vibration module, and outputs multiple physical quantity monitoring signals corresponding to the number of sensors, specifically including: heating current, vibration current, two vibration acceleration signals, and three temperature signals. The operating zone module internally contains a heating module and a vibration module. The heating module receives the heating operation signal and heating fault signal inputs, performs simulation calculations of the electric heating system, and outputs the heating current and three temperature sensor data. The vibration system receives the vibration operation signal and vibration fault signal, performs vibration system simulation calculations, and outputs the vibration current and two vibration sensor data.

[0117] 2) Modeling of the electric heating system: The internal structure of the heating module is as follows: Figure 6 As shown, this module implements the simulation of a heating system in a single zone. It consists of a fault injection module, a core function module, a current noise injection module, and a temperature noise injection module. It receives heating operation signals and heating fault signals as inputs, and generates heating current data and data from three temperature sensors as outputs.

[0118] The heating module's workflow begins with two key control signals: a heating operation signal and a heating fault signal. These signals control the heating module's operating state and activate the fault injection module, respectively. Upon receiving the signals, the fault injection module generates current and resistance values. First, the current signal enters the current noise injection module, generating noisy heating current data output. Simultaneously, the noisy current and resistance are input to the core functional module for thermodynamic simulation, thereby calculating the theoretical temperature. Finally, this temperature value is passed to the temperature noise injection module to add noise, ultimately generating simulated data corresponding to the three temperature sensors.

[0119] (1) Heating core functional module: The internal structure of the heating core functional module is as follows Figure 7 As shown, the module receives current I and resistance R as input, simulates the temperature evolution of the system based on the principle of thermodynamic equilibrium, and outputs the calculated temperature. This module is based on the formula... A model is constructed to simulate the temperature change process of the heating system. User input variables include: ambient temperature T. ∞ (Environmental fluctuation noise was included), skin area A, skin thickness h, convective heat transfer coefficient h conv Heating efficiency η, density ρ, specific heat capacity c p wait.

[0120] The simulation process first converts the input current and resistance into heating power based on heating efficiency. Simultaneously, it introduces ambient temperature with added fluctuation noise, calculates the temperature difference between this ambient temperature and the current system temperature, and then uses the skin area and convective heat transfer coefficient to derive heat dissipation loss. The module uses the difference between heating power and heat dissipation loss (net power) divided by the thermal inertia parameter composed of skin density, skin area, skin thickness, and specific heat capacity of the skin material to calculate the rate of temperature change. Finally, through integration, it obtains the real-time temperature output, which is then fed back into the heat dissipation calculation stage, forming a continuous dynamic closed-loop simulation.

[0121] (2) Fault Injection Module: The internal structure of the fault injection module is as follows: Figure 8 As shown in the diagram, this module receives heating fault signals as input and generates resistance and current as outputs. It consists of a resistance degradation fault simulation module and a voltage fault simulation module, simulating faults such as overheating, underheating, and open circuits based on simulation data.

[0122] The module is equipped with overcurrent protection detection. When the current calculated using voltage and resistance exceeds 30A, the current fault signal is forcibly set to open circuit fault.

[0123] Resistance degradation fault simulation module: Resistance degradation fault simulation module (such as...) Figure 9(As shown) This module is mainly used to simulate the performance degradation process of a resistor under heating fault conditions. It takes a heating fault signal as input, performs a series of logical calculations, and finally outputs the degraded resistor, thus reflecting the aging state of the resistor over time.

[0124] The module's calculation process begins with the fault signal detection stage. Once the system detects a valid vibration fault input, it reads three parameters: the degradation start period, the degradation duration period, and the current simulation period. The duration of the fault is then determined through a degradation period count calculation stage. Subsequently, the module judges the degradation type based on preset conditions and, combined with the input degradation rate, obtains a preliminary degradation value in the degradation rate calculation stage.

[0125] To ensure the validity of the simulation data, the initially calculated degradation rate needs to be included in a limiting range step. In this step, the module strictly limits the degradation rate to the range of [1 ± threshold] based on the degradation threshold. Finally, the module calculates the final resistance by combining the corrected degradation rate with the initial resistance, thereby obtaining and outputting the final degraded resistance value.

[0126] Voltage fault simulation module: The voltage fault simulation module is as follows Figure 10 As shown, this module receives a heating fault signal as input and first checks for an open circuit. If a fault is detected, the module sets the voltage to 0; if not, the system introduces an initial voltage to maintain normal operation. Ultimately, regardless of the path, the module will output a uniform voltage to achieve precise control of the vibration system.

[0127] (3) Current noise injection module:

[0128] The current noise injection module injects noise into the system simulation current data to simulate the characteristics of current sensor data collected in real-world scenarios, and superimposes the following types of noise onto the input current:

[0129] Measuring white noise: ( (Configurable parameters for the model)

[0130] ；

[0131] Power frequency interference noise: (Power frequency amplitude A is a configurable parameter of the model)

[0132] ；

[0133] ；

[0134] Quantization error: (Quantization error step size Q is a configurable parameter of the model)

[0135] ;

[0136] Slow drift noise: (standard deviation) (Configurable parameters for the model)

[0137] ;

[0138] The final output is a noisy current, and the model provides variables to adjust the magnitude of the noise.

[0139] (4) Temperature noise injection module: The temperature noise injection module injects noise into the temperature data calculated in the system simulation to simulate the characteristics of temperature sensor data collected in real-world scenarios, and superimposes the following types of noise onto the input temperature:

[0140] Gaussian white noise: (standard deviation) (Configurable parameters for the model)

[0141] ;

[0142] Quantization error: (Q is a configurable parameter of the model):

[0143] ;

[0144] After generating the noisy temperature, in order to differentiate between the three temperature sensors, different values ​​of quantization and white noise were added to the noisy temperature.

[0145] 3) Vibration excitation system modeling: The internal structure of the vibration excitation module is as follows: Figure 11 As shown, this module receives vibration operating signals and vibration fault signals as inputs, and generates vibration current data and two vibration acceleration sensor data as outputs, realizing the simulation of a single-zone vibration system. It consists of a fault injection module, a vibration core function module, a current noise injection module, and a vibration noise injection module. The vibration operating signal controls the start / stop state of the vibration exciter, while the vibration fault signal controls whether the fault injection module is active. Regardless of whether the vibration fault signal is active, the fault injection module outputs a corresponding current. This current is input to the noise injection module, where it undergoes noise enhancement to form a noisy current, which serves as both the vibration current data output and the input to the vibration core function module, used to calculate acceleration. This acceleration signal is further noise-enhanced by the vibration noise injection module, and the final output is the observation data from the two vibration sensors.

[0146] (1) Vibration core functional module:

[0147] The internal structure of the vibration core functional module is as follows: Figure 12As shown, this module receives current I as input, performs mechanical simulation based on the input data and parameters, and outputs acceleration a. This module is based on the formula...

[0148] ;

[0149] A model is constructed to simulate the acceleration change process of the vibration system. The variables provided by the model include: equivalent mass m, damping coefficient c, equivalent stiffness k, and permeability. Coil width L, number of coil turns N, film thickness d.

[0150] First, the module calculates the electromagnetic driving force based on the input current and preset physical parameters such as coils and thin films. This force enters the resultant force calculation stage, and the instantaneous acceleration of the system is solved by combining the equivalent mass.

[0151] The acceleration data is used to derive the corresponding velocity and displacement through an integral calculation process. These are then combined with the damping coefficient and equivalent stiffness to generate damping force and elastic force, respectively. These two forces serve as feedback quantities, returning in real-time to the resultant force node for superposition and correction with the electromagnetic driving force. Through this continuous dynamic cycle, the module achieves accurate simulation of the system's acceleration variation process.

[0152] (2) Fault injection module: Vibration excitation fault injection module such as Figure 13 As shown, this module receives vibration operating signals and vibration fault signals as inputs, and generates current as output. It simulates faults such as excessive vibration, insufficient vibration, and open circuit based on simulation data, while simultaneously controlling the shape of the pulse current. It includes a voltage fault simulation module, a resistance degradation fault simulation module, and a current control module.

[0153] Voltage fault simulation module: The voltage fault simulation module is as follows Figure 14 As shown, this module receives a vibration fault signal as input and first determines whether there is an open circuit fault. If the fault is detected, the module sets the voltage to 0; if not, the system introduces an initial voltage to maintain normal operation. Ultimately, regardless of the path, the module will generate a uniform voltage output to achieve precise control of the vibration system.

[0154] Resistance degradation fault simulation module: Resistance degradation fault simulation module (such as...) Figure 15 (As shown) This module is mainly used to simulate the performance degradation process of resistors under vibration fault conditions. It takes a vibration fault signal as input, performs a series of logical calculations, and finally outputs the degraded resistor, thus reflecting the aging state of the resistor over time.

[0155] The module's calculation process begins with the fault signal detection stage. Once the system detects a valid vibration fault input, it reads three parameters: the degradation start period, the degradation duration period, and the current simulation period. The duration of the fault is then determined through a degradation period count calculation stage. Subsequently, the module judges the degradation type based on preset conditions and, combined with the input degradation rate, obtains a preliminary degradation value in the degradation rate calculation stage.

[0156] To ensure the validity of the simulation data, the initially calculated degradation rate needs to be included in a limiting range step. In this step, the module strictly limits the degradation rate to the range of [1 ± threshold] based on the degradation threshold. Finally, the module calculates the final resistance by combining the corrected degradation rate with the initial resistance, thereby obtaining and outputting the final degraded resistance value.

[0157] Current control module: Current control module (such as...) Figure 16 (As shown) This module is mainly used to reproduce the experimental characteristics of the gradually decreasing current amplitude in a vibration system. It receives the vibration operating signal as input and generates a current attenuation coefficient through internal logic operations. Multiplying this coefficient by the original current simulates the decrease in current intensity as the number of energization cycles increases.

[0158] The module's operation begins with real-time monitoring of the signal's falling edge. Each detected falling edge triggers a counter to increment, obtaining a count value n. The system then calculates and outputs the current attenuation coefficient based on a linearly decreasing function. Simultaneously, to maintain the cyclic operation, the module checks whether the count value n has reached a threshold of 3 (i.e., the number of times the vibration exciter is energized during each operation): if it has, the counter is immediately reset to zero; otherwise, the system continues to wait for the next trigger signal, thus forming a closed-loop control.

[0159] S105, User Interface Development and Implementation.

[0160] 1) Development Architecture and Operating Environment: The simulation platform described in this invention integrates the simulation kernel and the front-end interactive interface through a hybrid architecture. Specifically, a cross-platform graphical user interface (GUI) is built using the PyQt5 framework, serving as a unified entry point for user control and parameter input. The underlying simulation kernel is built using MATLAB / Simulink, and bidirectional data communication and control command transmission between the two are achieved through MATLAB Engine for Python. The platform clearly defines and encapsulates the operating environment and dependent library versions (including the Python interpreter, numerical computation libraries, packaging tools, etc.) to ensure the repeatability and consistency of simulation tasks in different computing environments. The code and resources are organized using a modular directory structure, achieving effective separation and management of simulation scripts, model files, configuration parameters, and output data.

[0161] 2) Parametric Configuration and Management Interface: The platform provides a complete graphical parameter configuration system to achieve fine-grained control of the simulation model. This system includes: 1) A system parameter configuration interface, providing centralized editing and batch loading functions for heating systems (such as skin geometry, material properties, electrothermal parameters), vibration systems (such as equivalent mass, stiffness, electromagnetic parameters), environmental parameters, and noise parameters (noise intensity of each sensor, power frequency interference amplitude, etc.) in a grouped form; 2) A fault simulation configuration interface, supporting users to define fault types (such as open circuit, temperature anomaly, vibration anomaly), trigger cycles, durations, degradation models (linear / exponential), and their evolution rates by partition, through drop-down selection, numerical input, etc., and can save multiple fault scenarios as reusable configuration templates; 3) A simulation task control interface, integrating data acquisition parameter (such as sampling rate of each sensor, total simulation duration, storage path) setting functions, and providing real-time control buttons such as "Start," "Pause / Continue," and "Terminate," as well as a dynamic display window for simulation status and logs.

[0162] 3) Data Generation and Visualization Interaction: The platform provides end-to-end data support from simulation execution to result analysis. In the background, the system automatically schedules the simulation kernel according to user configuration, synchronously collects and timestamps the output data from dozens of sensor channels across five partitions, and encapsulates and stores it in HDF5 standard format files according to a predefined structure (such as grouping by sensor type). At the same time, it records complete simulation metadata and operation logs.

[0163] On the front end, the platform provides a dedicated data visualization module. Users can select historical or real-time generated simulation data files through a file browser and choose specific working partitions, sensor types (such as temperature, current, and acceleration) in the interactive chart interface. They can also customize time intervals and numerical ranges to generate and display multi-dimensional data curves in real time. This module supports graph zooming, viewing local details, and multi-graph comparison functions, making it easy for users to intuitively verify model behavior, analyze fault characteristics, and evaluate data quality.

[0164] S106. Simulation Results and Verification

[0165] 1) Verification Methodology and Data Foundation: This application systematically describes the verification process and methods used to evaluate the effectiveness and practicality of the mathematical physics model described in this invention. The core verification method is the comparative analysis method, which compares the multi-channel time series data generated by the model simulation with the experimental data collected from a real thermo-coupled anti-icing prototype under the same operating conditions and fault conditions. The comparison covers normal operating mode and various set fault modes, and the analysis dimensions include time-domain waveforms, key characteristic parameters (such as peak value, steady-state value, rise time), and overall trend, aiming to comprehensively evaluate the fidelity of the model from both qualitative matching and quantitative deviation perspectives.

[0166] 2) Verification Results of the Electric Heating Subsystem: For the electric heating subsystem, the current and temperature curves generated by the model under normal operating mode showed a high degree of consistency with the experimental data in core dynamic characteristics: the current signal exhibited the characteristic of returning to zero after an instantaneous pulse, and the temperature signal exhibited the pattern of rapid temperature rise followed by slow decay, verifying the accuracy of the basic thermodynamic model. Under fault modes, the model successfully reproduced the preset typical fault characteristics: when simulating an "open circuit" fault, the current output was zero, and the temperature remained consistent with the ambient temperature; when simulating an "under-temperature" fault, the current peak weakened, and the temperature rise rate and steady-state temperature were both lower than normal levels; when simulating an "over-temperature" fault, abnormal current and overheating temperature rise were observed. The trends and physical characteristics of the simulation data for all fault scenarios were consistent with the corresponding experimental data, proving the effectiveness of the fault injection mechanism.

[0167] 3) Validation Results of the Vibration Excitation Subsystem: For the vibration excitation subsystem, in the normal operation mode validation, the current pulse sequence and the corresponding vibration acceleration pulse sequence generated by the model achieved good matching with the experimental data in terms of pulse count, interval timing, and amplitude attenuation trend. In the fault mode validation, the model accurately simulated key fault states: when simulating "open circuit," there was no output of current and acceleration signals; when simulating "insufficient vibration," the acceleration pulse amplitude attenuated significantly; and when simulating "excessive vibration," the acceleration pulse amplitude increased abnormally. The simulation results accurately reflect the unique signal characteristics corresponding to different fault mechanisms, demonstrating the correctness of the vibration dynamics model and fault parameterization method.

[0168] 4) Comprehensive Performance Testing and Evaluation: Based on the aforementioned non-functional requirements, this application conducted rigorous comprehensive performance testing on the model. Execution efficiency testing showed that, on standard computing hardware, the computation time for completing a full simulation task with a physical duration of 600 seconds was significantly lower than the set threshold of 300 seconds, meeting the requirements for efficient simulation. Data quality testing confirmed that all output time-series data were continuous and complete, with no invalid data points (NaN or Inf) found, and timestamps were strictly evenly spaced. Reliability testing, by monitoring the memory usage curve of long-duration simulation tasks, demonstrated that the periodic simulation and data management strategy described in this invention can effectively avoid cumulative memory growth, ensuring the model possesses stability and robustness to support large-scale, batch data generation tasks. All test results met or exceeded the preset performance indicators.

[0169] This application successfully constructs a complete mathematical and physical simulation model of a thermo-coupled anti-icing and de-icing system, realizing a closed-loop mapping from electro-thermal and electro-mechanical physical equations to engineering simulation implementation. The model, through a parameterized fault injection mechanism and a modular noise superposition method, possesses a systematic ability to simulate faults and disturbances. It can controllably generate simulation data covering failure-type and anomaly-type faults, and describes the performance degradation process through linear or exponential evolution models. Simultaneously, the model integrates multiple types of high-fidelity noise, including Gaussian white noise, power frequency interference, low-frequency drift, and random impacts, ensuring that the data possesses the uncertainty characteristics of a real measurement environment. At the engineering implementation level, the model adopts a modular design, possessing high configurability and scalability. Through an integrated graphical user interface, it supports flexible configuration of system parameters, fault sequences, noise intensity, and operating cycles. The model has undergone systematic verification; its generated data is highly consistent with experimental data in terms of trends and characteristics, and meets stringent performance indicators. Therefore, the model constructed in this invention provides a high-quality, customizable data generation platform for research on fault diagnosis algorithms and verification of health management strategies for anti-icing and de-icing systems, possessing significant engineering application value.

[0170] Further reference Figure 17 As an implementation of the above-described method, this application provides an embodiment of a thermally coupled anti-icing and de-icing simulation data generation system, which is similar to... Figure 1 Corresponding to the method embodiments shown, the system can be specifically applied to various electronic devices.

[0171] refer to Figure 17A thermally coupled anti-icing and de-icing simulation data generation system includes: a simulation parameter configuration module 201, configured to receive and configure simulation parameters, including system geometric parameters, material property parameters, electric heating parameters, vibration excitation parameters, environmental condition parameters, and noise parameters; a cycle timing configuration module 202, configured to generate control signals to drive multiple independent working sub-divisions to work sequentially in a time-sharing manner based on a preset cycle timing configuration, wherein each working sub-division sequentially executes an electric heating stage and a vibration excitation stage within one working cycle; and a data acquisition module 203, which, based on the simulation parameters and the control signals, runs the coupled electric heating sub-model and vibration excitation sub-model. The excitation sub-model calculates skin temperature data and vibration response data respectively; wherein, the electric heating sub-model obtains skin temperature data based on the heat balance equation, and the vibration excitation sub-model obtains vibration response data based on the single-degree-of-freedom vibration equation; the fault injection module 204 is configured to dynamically modify at least one key model parameter in the electric heating sub-model or the vibration excitation sub-model according to preset fault parameters during the simulation process, so as to simulate system performance degradation or functional failure; wherein, the preset fault parameters include fault type, fault occurrence time, fault duration, and degradation evolution mode; the fault type includes failure-type faults and abnormal-type faults, and so on. The failure types include open circuit failures, and the abnormal failure types include failures caused by abnormal parameter changes, such as excessively high / low heating temperature or excessive / insufficient vibration. The degradation evolution modes include linear evolution modes and exponential evolution modes. When using the linear evolution mode, the degree of degradation of key model parameters is proportional to the number of working cycles, and the degradation process shows a uniform increase with the number of cycles. When using the exponential evolution mode, the degree of degradation of key model parameters is exponentially related to the number of working cycles, and the degradation process shows a slow initial increase followed by accelerated deterioration. The noise injection module 205 is configured to inject noise into the skin temperature data, vibration response data, and corresponding current data. According to the data, various noise signals conforming to preset statistical characteristics are injected, including: Gaussian white noise and quantization error are injected into the skin temperature data; Gaussian white noise, 1 / f low-frequency drift noise, environmental vibration harmonic interference noise and transient mechanical shock noise are injected into the vibration response data; power frequency interference noise, Gaussian white noise, quantization error and slow drift noise are injected into the current data; the simulation data generation module 206 is configured to perform simulation based on the fault injection module and the noise injection module to obtain multi-channel simulation data, encapsulate the multi-channel simulation data according to the structure corresponding to multiple working partitions and multiple sensor types, and output it as a standardized data file with timestamps.

[0172] In another aspect, this application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the following... Figure 1 The method shown.

[0173] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0174] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a computer terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0175] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for generating thermal coupling deicing simulation data, the method comprising: determining a plurality of thermal coupling parameters for a plurality of thermal coupling elements; determining a plurality of thermal coupling element locations; and determining a plurality of thermal coupling element orientations. The method includes: S1: Receive and configure simulation parameters, including system geometric parameters, material property parameters, electric heating parameters, vibration excitation parameters, environmental condition parameters, and noise parameters; S2: Based on the preset polling timing configuration, generate control signals to drive multiple independent working zones to work sequentially in a time-sharing manner. Each working zone executes the electric heating stage and the vibration excitation stage sequentially within one working cycle. S3: Based on the simulation parameters and the control signal, run the coupled electric heating sub-model and vibration excitation sub-model to calculate the skin temperature data and vibration response data respectively; The electric heating sub-model obtains skin temperature data based on the thermal balance equation, and the vibration excitation sub-model obtains vibration response data based on the single-degree-of-freedom vibration equation. S4: During the simulation, at least one key model parameter in the electric heating sub-model or the vibration excitation sub-model is dynamically modified according to the preset fault parameters to simulate system performance degradation or functional failure. The preset fault parameters include fault type, fault occurrence time, fault duration, and degradation evolution mode. The fault types include failure faults and abnormal faults. Failure faults include open circuit faults, and abnormal faults include faults caused by abnormal changes in parameters, such as excessively high / low heating temperature or excessive / insufficient vibration. The degradation and evolution modes include linear evolution mode and exponential evolution mode. When the linear evolution mode is adopted, the degree of degradation of the key model parameters is proportional to the number of working cycles, and the degradation process is characterized by a uniform increase with the number of cycles. When the exponential evolution mode is adopted, the degree of degradation of the key model parameters is exponentially related to the number of working cycles, and the degradation process is characterized by slow growth in the early stage and accelerated deterioration in the later stage. S5: Inject various noise signals conforming to preset statistical characteristics into the skin temperature data, vibration response data, and corresponding current data, including: Inject Gaussian white noise and quantization error into the skin temperature data; Gaussian white noise, 1 / f low-frequency drift noise, environmental vibration harmonic interference noise, and transient mechanical shock noise are injected into the vibration response data. Inject power frequency interference noise, Gaussian white noise, quantization error, and slow drift noise into the current data; S6: Based on steps S4 and S5, perform simulation to obtain multi-channel simulation data, encapsulate the multi-channel simulation data according to the structure corresponding to multiple working partitions and multiple sensor types, and output it as a standardized data file with timestamps.

2. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S4, a fault simulation is triggered by determining whether the current simulation time is within a preset fault time window. The fault time window is determined based on the fault occurrence time and fault duration configured by the user.

3. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S4, the fault parameters also include a maximum degradation threshold. When the change in the key model parameters calculated by the degradation evolution mode reaches the maximum degradation threshold, the key model parameters are maintained at the threshold level to simulate the fault development to a stable saturation state.

4. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S4, when the fault type is an open circuit fault, the equivalent resistance in the electric heating sub-model is modified to infinity, or the electromagnetic driving force in the vibration excitation sub-model is modified to zero.

5. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S5, when injecting the environmental vibration harmonic interference noise into the vibration response data, one or more sinusoidal signals with preset amplitudes are generated and superimposed according to the preset fundamental frequency and harmonic order to simulate the structural background vibration caused by the operation of the aircraft engine.

6. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S5, the slow drift noise is generated using a random walk simulation, and the drift amount is randomly accumulated within each simulation step according to a preset step standard deviation to simulate the slow time-varying characteristics of the sensor measurement baseline.

7. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S6, the standardized data file is stored in HDF5 format, and the internal data structure of the standardized data file contains at least groups named according to partition numbers. Each group is divided into datasets according to sensor type, and each dataset contains one-dimensional time series data and corresponding attribute information.

8. The method for generating thermally coupled anti-icing and de-icing simulation data according to claim 1, characterized in that: In step S3, the electric heating sub-model and the vibration excitation sub-model run synchronously in simulation time, and the two are time-coupled through a shared simulation step size and partition activation signal, so that the skin temperature data calculated in the electric heating stage is used as the initial or boundary condition for the simulation operation in the vibration excitation stage within the same partition.

9. A method for generating thermally coupled anti-icing and de-icing simulation data according to any one of claims 1-8, characterized in that: The simulation parameters are configured through a graphical user interface, which provides independent input panels for different parameter categories and saves the entire set of parameter settings as a configuration file or loads them from the configuration file to achieve the reproduction and batch management of simulation scenarios.

10. A thermally coupled deicing simulation data generation system, characterized by, The system includes: The simulation parameter configuration module is configured to receive and configure simulation parameters, which include system geometric parameters, material property parameters, electric heating parameters, vibration excitation parameters, environmental condition parameters, and noise parameters. The polling timing configuration module is configured to generate control signals to drive multiple independent working zones to work sequentially in a time-sharing manner based on a preset polling timing configuration. Each working zone executes the electric heating stage and the vibration excitation stage sequentially within one working cycle. The data acquisition module, based on the simulation parameters and the control signal, runs the coupled electric heating sub-model and vibration excitation sub-model to calculate the skin temperature data and vibration response data respectively. The electric heating sub-model obtains skin temperature data based on the thermal balance equation, and the vibration excitation sub-model obtains vibration response data based on the single-degree-of-freedom vibration equation. The fault injection module is configured to dynamically modify at least one key model parameter in the electric heating sub-model or the vibration excitation sub-model according to preset fault parameters during the simulation process, so as to simulate system performance degradation or functional failure. The preset fault parameters include fault type, fault occurrence time, fault duration, and degradation evolution mode. The fault types include failure faults and abnormal faults. Failure faults include open circuit faults, and abnormal faults include faults caused by abnormal changes in parameters, such as excessively high / low heating temperature or excessive / insufficient vibration. The degradation and evolution modes include linear evolution mode and exponential evolution mode. When the linear evolution mode is adopted, the degree of degradation of the key model parameters is proportional to the number of working cycles, and the degradation process is characterized by a uniform increase with the number of cycles. When the exponential evolution mode is adopted, the degree of degradation of the key model parameters is exponentially related to the number of working cycles, and the degradation process is characterized by slow growth in the early stage and accelerated deterioration in the later stage. The noise injection module is configured to inject various noise signals conforming to preset statistical characteristics into the skin temperature data, vibration response data, and corresponding current data, including: Inject Gaussian white noise and quantization error into the skin temperature data; Gaussian white noise, 1 / f low-frequency drift noise, environmental vibration harmonic interference noise, and transient mechanical shock noise are injected into the vibration response data. Inject power frequency interference noise, Gaussian white noise, quantization error, and slow drift noise into the current data; The simulation data generation module is configured to perform simulations based on the fault injection module and the noise injection module to obtain multi-channel simulation data. The multi-channel simulation data is encapsulated according to the structure corresponding to multiple working partitions and multiple sensor types, and output as a standardized data file with timestamps.

Images