Fault handling method and system for ice protection simulation system

Abstract

The present application belongs to the technical field of data processing, and particularly relates to a fault processing method and system of an anti-icing simulation system, aiming to solve the problem of low warning accuracy and easy occurrence of processing instruction mismatch in the prior art. The method proposed in the present application comprises: acquiring multi-source simulation data of the anti-icing simulation system; determining the temperature drop rate and the humidity change rate according to the environmental simulation data, and determining the dynamic regulation factor in combination with the system characteristic data; acquiring valve timing data of the anti-icing simulation system, inputting the valve timing data and the dynamic regulation factor into a pre-trained timing prediction model for fault prediction to obtain a fault prediction probability; taking the fault prediction probability as a query vector, searching for a preplan node in a pre-constructed knowledge graph and generating a predictive instruction; and executing the predictive instruction when the fault prediction probability reaches a preset threshold. Based on the method, the warning accuracy and instruction adaptability in the fault processing stage are effectively improved.

Description

Technical Field

[0001] This invention belongs to the technical field of data processing, and specifically relates to a fault handling method and system for an anti-icing simulation system. Background Technology

[0002] Aircraft anti-icing system simulation typically includes multiple subsystems such as wing anti-icing, engine anti-icing, windshield heating, and pitot tube heating. These subsystems work together to simulate real-world flight conditions (such as temperature, humidity, and flight speed) to evaluate the anti-icing system's performance. For fault handling scenarios in anti-icing simulation systems, the relevant technologies mainly employ the following two approaches to achieve fault warning and handling:

[0003] One approach involves triggering alarms using predefined environmental parameter thresholds (e.g., temperature ≤ 0℃, humidity ≥ 85%). However, this method tends to overlook the impact of dynamic environmental changes on fault sensitivity, reducing the accuracy of fault warnings. Specifically, using a constant threshold can lead to missed detections due to icing caused by response delays in sudden temperature drops, while false alarms are more likely to occur when humidity fluctuations are gradual. The second approach relies on mapping pre-defined handling schemes to a historical fault database. However, these schemes often use discrete probability intervals (e.g., "high / medium / low risk"), lacking refined analysis of real-time predicted probabilities, leading to mismatches in processing instructions. Specifically, when the predicted result is a boundary probability, incorrect scheme types are easily triggered, resulting in a mismatch between the final execution instructions and the actual risk.

[0004] Therefore, it is necessary to propose a fault handling method for the anti-icing simulation system to solve the problems of low early warning accuracy and mismatch of processing instructions in related technologies when performing fault handling. Summary of the Invention

[0005] To address the aforementioned problems in the prior art, namely, the low accuracy of early warnings and the susceptibility to mismatched processing commands during fault handling, this invention proposes a fault handling method for an anti-icing simulation system in its first aspect, the method comprising:

[0006] Acquire multi-source simulation data of the anti-icing simulation system, wherein the multi-source simulation data includes environmental simulation data and system characteristic data;

[0007] Based on the environmental simulation data, the temperature drop rate and humidity change rate are determined, and dynamic control factors are determined in combination with the system characteristic data. The dynamic control factors include valve response threshold and fault detection sensitivity. The valve response threshold is negatively correlated with the humidity change rate, and the fault detection sensitivity is positively correlated with the temperature drop rate.

[0008] The valve timing data of the anti-icing simulation system is obtained, and the valve timing data and the dynamic control factor are input into the pre-trained timing prediction model to perform fault prediction and obtain the fault prediction probability. The valve timing data is used as the timing feature input of the timing prediction model, and the timing prediction model determines the fault prediction boundary according to the dynamic control factor.

[0009] Using the fault prediction probability as a query vector, a contingency plan node is retrieved in a pre-constructed knowledge graph, and a predictive instruction is generated. The knowledge graph includes a static topology network constructed based on historical fault cases. The static topology network includes multiple nodes, and each node is used to store at least one mapping relationship between fault prediction probabilities and execution instructions.

[0010] When the fault prediction probability reaches a preset threshold, the predictive instruction is executed.

[0011] In some preferred embodiments, determining the temperature drop rate and humidity change rate based on the environmental simulation data includes:

[0012] Based on the environmental simulation data, determine the time-series data of environmental temperature;

[0013] According to the preset sampling duration, a time window consisting of the current moment and the target historical moment is determined based on the environmental temperature time series data, wherein the time interval between the target historical moment and the current moment satisfies the sampling duration;

[0014] Determine the difference in ambient temperature between the first and last moments of the time window;

[0015] The temperature drop rate is determined based on the difference between the sampling duration and the ambient temperature.

[0016] In some preferred embodiments, determining the temperature drop rate and humidity change rate based on the environmental simulation data includes:

[0017] Based on the environmental simulation data, the dew point temperature time series data is determined;

[0018] The time series data of dew point temperature is subjected to moving average filtering, and the dew point temperature values ​​of the two sampling points closest to the current time node are determined from the filtered time series data of dew point temperature.

[0019] Based on the dew point temperature values ​​of the two most recent sampling points, determine the dew point temperature increment per unit time.

[0020] The humidity change rate is determined based on the dew point temperature increment.

[0021] In some preferred embodiments, determining the dynamic control factor by combining the system characteristic data includes:

[0022] Extract valve structural parameters from the system feature data, including opening / closing delay time and steady-state leakage rate.

[0023] Based on the opening and closing delay time, determine the valve's response delay compensation coefficient;

[0024] Based on the steady-state leakage rate, determine the leakage compensation rate of the valve;

[0025] Based on the humidity change rate, determine the negative correlation adjustment coefficient;

[0026] The valve response threshold is determined by combining the response delay compensation coefficient, the leakage compensation rate, and the negative correlation adjustment coefficient.

[0027] In some preferred embodiments, determining the dynamic control factor by combining the system characteristic data includes:

[0028] Based on the system characteristic data, determine the degradation parameters of the system heating unit;

[0029] The basic sensitivity gain is determined based on the degradation parameters of the heating unit;

[0030] Based on the stated temperature drop rate, determine the positively correlated enhancement factor;

[0031] The fault detection sensitivity is determined based on the baseline sensitivity gain and the positive correlation enhancement factor.

[0032] In some preferred embodiments, the valve response threshold is calculated using the following formula:

[0033] ;

[0034] In the formula, V th V is the valve response threshold. base Based on the response threshold, ϵ represents the valve opening / closing delay time, and ϵ is a preset safety parameter. This represents the average valve opening and closing delay time. k is the delay compensation coefficient. η For leakage compensation rate, V h η is the humidity change rate, σ is the environmental modulation intensity coefficient, β is the steepness factor of the humidity change rate modulation curve, and η is the steady-state leakage rate.

[0035] In some preferred embodiments, the fault detection sensitivity is calculated using the following formula:

[0036] ;

[0037] In the formula, S det For fault detection sensitivity, S base The preset reference sensitivity is given, ΔR is the system resistance aging rate, δ is the root mean square deviation of heat flux distribution, α is the aging effect gain coefficient, and V is the system resistance aging rate. c γ is the rate of temperature drop, V0 is the threshold for temperature drop enhancement initiation, and γ is the temperature drop rate. min γ max is the extreme value of the system safety constraint parameter, and k is the preset coefficient of the control gain growth rate.

[0038] In some preferred embodiments, the step of using the fault prediction probability as a query vector to retrieve contingency plan nodes in a pre-constructed knowledge graph and generating predictive instructions includes:

[0039] Calculate the absolute difference between the target fault probability value in the fault prediction result and the fault probability weight stored in each node of the knowledge graph;

[0040] Based on the preset topological distance weight coefficient and probability tolerance threshold, a set of candidate nodes that meet the conditions is selected in the knowledge graph. The absolute interpolation between each node in the candidate node set and the target fault probability value is less than the probability tolerance threshold.

[0041] The node with the smallest topological distance from the candidate node set is selected as the matching plan node, and the preset instruction set mapped by the plan node is called as the predictive instruction set.

[0042] In some preferred embodiments, the candidate node set satisfies:

[0043] ;

[0044] In the formula, V candidate For the candidate node set, Let V be the target fault probability value in the fault prediction result, and let V be all the nodes in the knowledge graph. A node V in a knowledge graph i Stored historical failure probability weights, This is the preset probability tolerance threshold;

[0045] The matched plan nodes satisfy:

[0046] ;

[0047] In the formula, For the matched plan nodes, λ is the topology weight coefficient. For the current node V current To V iThe sum of edge weights of the shortest path, argmin() is the minimum value traversal function.

[0048] In a second aspect, the present invention also proposes a fault handling system for an anti-icing simulation system, comprising:

[0049] The data acquisition module is used to acquire multi-source simulation data of the anti-icing simulation system, including environmental simulation data and system characteristic data.

[0050] The parameter calculation module is used to determine the temperature drop rate and humidity change rate based on the environmental simulation data, and to determine the dynamic control factor in combination with the system characteristic data. The dynamic control factor includes the valve response threshold and the fault detection sensitivity. The valve response threshold is negatively correlated with the humidity change rate, and the fault detection sensitivity is positively correlated with the temperature drop rate.

[0051] The fault prediction module is used to acquire valve timing data of the anti-icing simulation system, and based on the valve timing data and the dynamic control factor, to perform fault prediction using a pre-trained timing prediction model to obtain the fault prediction probability. The valve timing data is used as the input data of the timing prediction model, and the dynamic control factor is used as the prediction control parameter of the timing prediction model.

[0052] The instruction generation module is used to match contingency plan nodes in a pre-constructed knowledge graph based on the fault prediction probability and generate predictive instructions. The knowledge graph includes a static topology network constructed based on historical fault cases of the anti-icing simulation system. The static topology network includes multiple nodes, and each node is used to store at least one mapping relationship between fault prediction probability and execution instruction.

[0053] The fault response module is used to respond to the fault trigger signal of the anti-icing simulation system and execute the predictive instructions.

[0054] The beneficial effects of this invention are:

[0055] (1) Based on the method proposed in this invention, a negative / positive correlation between valve response threshold, fault detection sensitivity and environmental parameters is established by dynamically adjusting the factor. When the ambient humidity rises rapidly, the system automatically lowers the valve response threshold, so that the anti-icing device can start earlier, avoiding the response lag problem caused by the fixed threshold in the traditional solution. When the temperature drops sharply, the system actively improves the fault detection sensitivity, enhances the ability to capture the risk of icing, and eliminates the protection blind spot in the sudden temperature drop scenario.

[0056] (2) Based on the method proposed in this invention, valve time series data and dynamic control factors are synchronously input into the time series prediction model, so that the time series prediction model dynamically adjusts the probability boundary according to environmental parameters, and the prediction mechanism evolves from single data-driven to dual-track coupling of environment and equipment, so that fault early warning has meteorological elasticity and equipment adaptability. Attached Figure Description

[0057] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0058] Figure 1 This is a flowchart illustrating a fault handling method for an anti-icing simulation system proposed in an embodiment of the present invention.

[0059] Figure 2 This is a schematic diagram of the structure of a fault handling system for an anti-icing simulation system proposed in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the structure of a computer system proposed in an embodiment of the present invention. Detailed Implementation

[0061] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0062] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0063] Please refer to Figure 1 The first embodiment of this application provides a fault handling method for an anti-icing simulation system, the method comprising:

[0064] Step S10: Obtain multi-source simulation data of the anti-icing simulation system, wherein the multi-source simulation data includes environmental simulation data and system characteristic data;

[0065] In this embodiment, the multi-source simulation data of the anti-icing simulation system essentially refers to a heterogeneous dataset that integrates dynamic environmental simulation features and system operating status. Specifically, environmental simulation data mainly refers to real-time environmental parameters and their derived indicators generated by the sensor network or meteorological model of the anti-icing simulation system. System feature data mainly refers to the operating parameters of the anti-icing simulation system regarding its real-time operating status and historical performance. Based on different data attributes, they can be roughly divided into two categories: static equipment attributes and dynamic operating parameters. Static equipment attributes include, but are not limited to, system valve type, heating power, pipeline layout topology, etc., while dynamic operating parameters include, but are not limited to, historical valve time series data, historical fault records, etc.

[0066] It is easy to understand that this embodiment achieves dual closed-loop feedback between environment and system by simultaneously collecting environmental simulation data and system characteristic data. In subsequent steps, the coupling relationship between the two types of data is established through dynamic adjustment factors to overcome the response lag of static threshold to environmental changes (such as valve delay when humidity rises sharply), while effectively avoiding the instruction jump of discrete plans at critical probability points (such as a sharp increase in trigger power when the probability is 59%→61%).

[0067] Step S20: Based on the environmental simulation data, determine the temperature drop rate and humidity change rate, and combine the system characteristic data to determine the dynamic control factor. The dynamic control factor includes the valve response threshold and the fault detection sensitivity. The valve response threshold is negatively correlated with the humidity change rate, and the fault detection sensitivity is positively correlated with the temperature drop rate.

[0068] In this step, the cooling rate specifically refers to the rate at which the temperature of the anti-icing simulation system decreases per unit time, expressed in minutes. Therefore, the unit of the cooling rate is °C / min. Based on this cooling rate, the rate of heat exchange imbalance within the system can be determined. This rate of heat exchange imbalance directly relates to the phase change efficiency of liquid water into ice crystals, which further serves as the basis for analyzing the icing process.

[0069] In this step, the humidity variation rate specifically refers to the percentage change in humidity of the anti-icing simulation system per unit time, expressed in minutes, with the unit being % / min. This humidity variation rate effectively reflects the intensity of water vapor transport within the system, thereby determining the ice growth rate.

[0070] It is easy to understand that by further analyzing the rate of temperature drop and the rate of humidity change, a dynamic control factor is obtained that couples environmental simulation data and system characteristic data. This dynamic control factor consists of the valve response threshold and the fault detection sensitivity. Based on this, the valve response threshold and the fault detection sensitivity are used as the basis for early warning detection in subsequent steps, which can effectively improve the accuracy of fault early warning.

[0071] Step S30: Obtain valve timing data of the anti-icing simulation system, input the valve timing data and the dynamic control factor into the pre-trained timing prediction model to perform fault prediction and obtain the fault prediction probability. The valve timing data is used as the timing feature input of the timing prediction model, and the timing prediction model determines the fault prediction boundary according to the dynamic control factor.

[0072] Among them, valve timing data is a millisecond-level record reflecting the dynamic behavior of the actuator of the anti-icing system. It includes at least: timestamp sequence of valve opening and closing, state curve of opening percentage change, and action-related indicators (such as response delay, i.e., the time difference between command issuance and action execution; and oscillation frequency, i.e. the number of starts and stops per unit time).

[0073] In this embodiment, valve time-series data is used as the main feature input, and time-series statistics (mean, variance, autocorrelation, etc.) are extracted through necessary feature engineering. At the same time, a dynamic adjustment factor is used to define the prediction boundary, so that the time-series prediction model uses different parameters to make predictions under different conditions. For example, when the temperature drop rate increases, the time-series prediction model will compress the confidence interval using the dynamic adjustment factor, making it more accurate in high-risk scenarios. When the humidity change rate increases, the time-series prediction model will expand the fault probability judgment range to respond in advance to the risk of a sudden increase in humidity.

[0074] In this embodiment, the time series prediction model can adopt various feasible frameworks such as LSTM (Long Short-Term Memory) and RNN (Recurrent Neural Network). Regarding the specific configuration, training process, parameter configuration, etc. of the time series prediction model, those skilled in the art can set them as needed, and this embodiment does not limit them.

[0075] Step S40: Using the fault prediction probability as a query vector, retrieve the contingency plan node in the pre-constructed knowledge graph and generate a predictive instruction. The knowledge graph includes a static topology network constructed based on historical fault cases. The static topology network includes multiple nodes, and each node is used to store at least one mapping relationship between fault prediction probabilities and execution instructions.

[0076] Contingency plan nodes are the smallest decision-making units in the static topology network of the knowledge graph. They are built based on historical failure cases and are used to store the precise mapping relationship between failure prediction probabilities and execution instructions.

[0077] Among them, the knowledge graph is a multi-dimensional relationship network built based on historical failure cases, and its core structure includes:

[0078] Nodes include, but are not limited to: fault probability nodes: used to store continuous probability values; instruction nodes: used to bind and execute commands (such as "valve opening 85% + heating power 5kW"); environmental constraint nodes: used to record the meteorological condition boundaries that trigger the instruction.

[0079] The topological relationships between nodes are such that each node stores at least 256 sets of historical case mapping relationships, forming a continuous mapping space from probability to instruction.

[0080] Step S50: When the fault prediction probability reaches a preset threshold, execute the predictive instruction.

[0081] In this embodiment, the preset threshold is a boundary value used to evaluate the probability of fault prediction. The probability of fault prediction obtained in the previous steps is compared with this boundary value. When it meets the boundary value, it is determined that the fault risk has reached a certain level and the system should be called to execute the pre-stored instruction set.

[0082] In one feasible implementation, the preset threshold is not a fixed value, but a dynamic boundary that is adjusted in real time according to environmental risks. Its calculation formula is as follows:

[0083] ;

[0084] In the formula, T dynamic For this preset threshold, T base Based on the threshold (which can be determined by equipment operating standards), R h R represents the rate of change in humidity. c This represents the rate of temperature drop.

[0085] Furthermore, in the above embodiments, determining the temperature drop rate and humidity change rate based on the environmental simulation data includes:

[0086] Based on the environmental simulation data, determine the environmental temperature time series data; according to the preset sampling duration, determine the time window formed by the current moment and the target historical moment based on the environmental temperature time series data, wherein the time interval between the target historical moment and the current moment satisfies the sampling duration; determine the difference in environmental temperature corresponding to the first moment and the last moment in the time window; determine the temperature drop rate based on the sampling duration and the difference in environmental temperature.

[0087] Furthermore, in the above embodiments, determining the temperature drop rate and humidity change rate based on the environmental simulation data includes:

[0088] Based on the environmental simulation data, the time series data of dew point temperature is determined; the time series data of dew point temperature is subjected to moving average filtering, and the dew point temperature values ​​of the two sampling points closest to the current time node are determined in the filtered time series data of dew point temperature; based on the dew point temperature values ​​of the two closest sampling points, the dew point temperature increment within a preset unit time is determined; based on the dew point temperature increment, the humidity change rate is determined.

[0089] Furthermore, in the above embodiments, determining the dynamic control factor by combining the system characteristic data includes:

[0090] The valve's structural parameters are extracted from the system feature data. These parameters include opening / closing delay time and steady-state leakage rate. Based on the opening / closing delay time, a response delay compensation coefficient is determined. Based on the steady-state leakage rate, a leakage compensation rate is determined. Based on the humidity change rate, a negative correlation adjustment coefficient is determined. Combining the response delay compensation coefficient, the leakage compensation rate, and the negative correlation adjustment coefficient, a valve response threshold is determined.

[0091] Furthermore, determining the dynamic control factor by combining the system characteristic data includes:

[0092] Based on the system characteristic data, the degradation parameters of the system heating unit are determined; based on the degradation parameters of the heating unit, the basic sensitivity gain is determined; based on the temperature drop rate, the positive correlation enhancement factor is determined; based on the basic sensitivity gain and the positive correlation enhancement factor, the fault detection sensitivity is determined.

[0093] Furthermore, the formula for calculating the valve response threshold mentioned above satisfies:

[0094] ;

[0095] In the formula, Vth is the valve response threshold, Vbase is the basic response threshold, Tdesign is the valve opening and closing delay time, ϵ is the preset safety parameter, Tactual is the valve average opening and closing delay time, kT is the delay compensation coefficient, kη is the leakage compensation rate, Vh is the humidity change rate, σ is the environmental modulation intensity coefficient, β is the steepness factor of the humidity change rate modulation curve, and η is the steady-state leakage rate.

[0096] Meanwhile, the calculation formula for the aforementioned fault detection sensitivity satisfies:

[0097] ;

[0098] In the formula, Sdet For fault detection sensitivity, S base The preset reference sensitivity is given, ΔR is the system resistance aging rate, δ is the root mean square deviation of heat flux distribution, α is the aging effect gain coefficient, and V is the system resistance aging rate. c γ is the rate of temperature drop, V0 is the threshold for temperature drop enhancement initiation, and γ is the temperature drop rate. min γ max is the extreme value of the system safety constraint parameter, and k is the preset coefficient of the control gain growth rate.

[0099] Furthermore, the step of matching contingency plan nodes in a pre-constructed knowledge graph based on the dynamic failure probability to generate a predictive instruction set containing hierarchical control logic includes:

[0100] Calculate the absolute difference between the target fault probability value in the fault prediction result and the fault probability weight stored in each node of the knowledge graph; based on the preset topological distance weight coefficient and probability tolerance threshold, filter the candidate node set that meets the conditions in the knowledge graph, wherein the absolute interpolation between each node in the candidate node set and the target fault probability value is less than the probability tolerance threshold; select the node with the smallest topological distance from the candidate node set as the matching plan node, and call the preset instruction set mapped by the plan node as the predictive instruction set.

[0101] The candidate node set satisfies:

[0102] ;

[0103] In the formula, V candidate For the candidate node set, Let V be the target fault probability value in the fault prediction result, and let V be all the nodes in the knowledge graph. A node V in a knowledge graph i Stored historical failure probability weights, This is the preset probability tolerance threshold;

[0104] The matched plan nodes satisfy:

[0105] ;

[0106] In the formula, For the matched plan nodes, λ is the topology weight coefficient. For the current node V current To V i The sum of edge weights of the shortest path, argmin() is the minimum value traversal function.

[0107] Please refer to Figure 2 The second embodiment of this application provides a fault handling system for an anti-icing simulation system, comprising:

[0108] Data acquisition module 210 is used to acquire multi-source simulation data of the anti-icing simulation system, wherein the multi-source simulation data includes environmental simulation data and system characteristic data;

[0109] The parameter calculation module 220 is used to determine the temperature drop rate and humidity change rate based on the environmental simulation data, and to determine the dynamic control factor in combination with the system characteristic data. The dynamic control factor includes a valve response threshold and a fault detection sensitivity. The valve response threshold is negatively correlated with the humidity change rate, and the fault detection sensitivity is positively correlated with the temperature drop rate.

[0110] The fault prediction module 230 is used to acquire valve timing data of the anti-icing simulation system, and based on the valve timing data and the dynamic control factor, to perform fault prediction using a pre-trained timing prediction model to obtain the fault prediction probability. The valve timing data is used as the input data of the timing prediction model, and the dynamic control factor is used as the prediction control parameter of the timing prediction model.

[0111] The instruction generation module 240 is used to match contingency plan nodes in a pre-constructed knowledge graph based on the fault prediction probability and generate predictive instructions. The knowledge graph includes a static topology network constructed based on historical fault cases of the anti-icing simulation system. The static topology network includes multiple nodes, and each node is used to store at least one mapping relationship between fault prediction probability and execution instruction.

[0112] The fault response module 250 is used to respond to the fault trigger signal of the anti-icing simulation system and execute the predictive instructions.

[0113] A third embodiment of this application also proposes an electronic device, comprising:

[0114] At least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor to implement the method as described in the first embodiment.

[0115] The fourth embodiment of this application also proposes a computer-readable storage medium storing computer instructions for execution by the computer to implement the method described in the first embodiment.

[0116] The following is for reference. Figure 3 It shows a schematic diagram of the structure of a computer system suitable for using a server to implement the methods, systems, and apparatus embodiments of this application. Figure 3 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0117] like Figure 3 As shown, the computer system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 302 or programs loaded from storage section 308 into Random Access Memory (RAM) 303. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0118] The following components are connected to I / O interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0119] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof.

[0120] More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0121] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0122] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0123] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0124] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0125] The technical solution of the present invention has now been described in conjunction with the preferred embodiments shown in the accompanying drawings.

[0126] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A fault handling method for an anti-icing simulation system, characterized in that, The method includes: Acquire multi-source simulation data of the anti-icing simulation system, wherein the multi-source simulation data includes environmental simulation data and system characteristic data; Based on the environmental simulation data, the temperature drop rate and humidity change rate are determined, and dynamic control factors are determined in combination with the system characteristic data. The dynamic control factors include valve response threshold and fault detection sensitivity. The valve response threshold is negatively correlated with the humidity change rate, and the fault detection sensitivity is positively correlated with the temperature drop rate. The valve timing data of the anti-icing simulation system is obtained, and the valve timing data and the dynamic control factor are input into the pre-trained timing prediction model to perform fault prediction and obtain the fault prediction probability. The valve timing data is used as the timing feature input of the timing prediction model, and the timing prediction model determines the fault prediction boundary according to the dynamic control factor. Using the fault prediction probability as a query vector, a contingency plan node is retrieved in a pre-constructed knowledge graph, and a predictive instruction is generated. The knowledge graph includes a static topology network constructed based on historical fault cases. The static topology network includes multiple nodes, and each node is used to store at least one mapping relationship between fault prediction probabilities and execution instructions. When the fault prediction probability reaches a preset threshold, the predictive instruction is executed.

2. The fault handling method for the anti-icing simulation system according to claim 1, characterized in that, The step of determining the temperature drop rate and humidity change rate based on the environmental simulation data includes: Based on the environmental simulation data, determine the time-series data of environmental temperature; According to the preset sampling duration, a time window consisting of the current moment and the target historical moment is determined based on the environmental temperature time series data, wherein the time interval between the target historical moment and the current moment satisfies the sampling duration; Determine the difference in ambient temperature between the first and last moments of the time window; The temperature drop rate is determined based on the difference between the sampling duration and the ambient temperature.

3. The fault handling method for the anti-icing simulation system according to claim 1, characterized in that, The step of determining the temperature drop rate and humidity change rate based on the environmental simulation data includes: Based on the environmental simulation data, the dew point temperature time series data is determined; The time series data of dew point temperature is subjected to moving average filtering, and the dew point temperature values ​​of the two sampling points closest to the current time node are determined from the filtered time series data of dew point temperature. Based on the dew point temperature values ​​of the two most recent sampling points, determine the dew point temperature increment per unit time. The humidity change rate is determined based on the dew point temperature increment.

4. The fault handling method for the anti-icing simulation system according to claim 1, characterized in that, The step of determining the dynamic control factor by combining the system characteristic data includes: Extract the valve's structural parameters from the system feature data. The valve's structural parameters include opening / closing delay time and steady-state leakage rate. Based on the opening and closing delay time, determine the valve's response delay compensation coefficient; Determine the leakage compensation rate of the valve based on the steady-state leakage rate; Based on the humidity change rate, determine the negative correlation adjustment coefficient; The valve response threshold is determined by combining the response delay compensation coefficient, the leakage compensation rate, and the negative correlation adjustment coefficient.

5. The fault handling method for the anti-icing simulation system according to claim 1, characterized in that, The step of determining the dynamic control factor by combining the system characteristic data includes: Based on the system characteristic data, determine the degradation parameters of the system heating unit; The basic sensitivity gain is determined based on the degradation parameters of the heating unit; Based on the stated temperature drop rate, determine the positively correlated enhancement factor; The fault detection sensitivity is determined based on the baseline sensitivity gain and the positive correlation enhancement factor.

6. The fault handling method for the anti-icing simulation system according to claim 4, characterized in that, The valve response threshold is calculated using the following formula: ； In the formula, For valve response threshold, Based on the response threshold, ϵ represents the valve opening / closing delay time, and ϵ is a preset safety parameter. This represents the average valve opening and closing delay time. k is the delay compensation coefficient. η For leakage compensation rate, V h η is the humidity change rate, σ is the environmental modulation intensity coefficient, β is the steepness factor of the humidity change rate modulation curve, and η is the steady-state leakage rate.

7. The fault handling method for the anti-icing simulation system according to claim 5, characterized in that, The fault detection sensitivity is calculated using the following formula: ； In the formula, S det For fault detection sensitivity, S base The preset reference sensitivity is given, ΔR is the system resistance aging rate, δ is the root mean square deviation of heat flux distribution, α is the aging effect gain coefficient, and V is the system resistance aging rate. c γ is the rate of temperature drop, V0 is the threshold for temperature drop enhancement initiation, and γ is the temperature drop rate. min γ max is the extreme value of the system safety constraint parameter, and k is the preset coefficient of the control gain growth rate.

8. The fault handling method for the anti-icing simulation system according to claim 1, characterized in that, The step of using the fault prediction probability as a query vector to retrieve contingency plan nodes in a pre-constructed knowledge graph and generating predictive instructions includes: Calculate the absolute difference between the target fault probability value in the fault prediction result and the fault probability weight stored in each node of the knowledge graph; Based on the preset topological distance weight coefficient and probability tolerance threshold, a set of candidate nodes that meet the conditions is selected in the knowledge graph. The absolute interpolation between each node in the candidate node set and the target fault probability value is less than the probability tolerance threshold. The node with the smallest topological distance from the candidate node set is selected as the matching plan node, and the preset instruction set mapped by the plan node is called as the predictive instruction.

9. The fault handling method for the anti-icing simulation system according to claim 8, characterized in that, The candidate node set satisfies: ； In the formula, V candidate For the candidate node set, Let V be the target fault probability value in the fault prediction result, and let V be all the nodes in the knowledge graph. A node V in a knowledge graph i Stored historical failure probability weights, This is the preset probability tolerance threshold; The matched plan nodes satisfy: ； In the formula, For the matched plan nodes, λ is the topology weight coefficient. For the current node V current To V i The sum of edge weights of the shortest path, argmin() is the minimum value traversal function.

10. A fault handling system for an anti-icing simulation system, characterized in that, The system includes: The data acquisition module is used to acquire multi-source simulation data of the anti-icing simulation system, including environmental simulation data and system characteristic data. The parameter calculation module is used to determine the temperature drop rate and humidity change rate based on the environmental simulation data, and to determine the dynamic control factor in combination with the system characteristic data. The dynamic control factor includes the valve response threshold and the fault detection sensitivity. The valve response threshold is negatively correlated with the humidity change rate, and the fault detection sensitivity is positively correlated with the temperature drop rate. The fault prediction module is used to acquire valve timing data of the anti-icing simulation system, and based on the valve timing data and the dynamic control factor, to perform fault prediction using a pre-trained timing prediction model to obtain the fault prediction probability. The valve timing data is used as the input data of the timing prediction model, and the dynamic control factor is used as the prediction control parameter of the timing prediction model. The instruction generation module is used to match contingency plan nodes in a pre-constructed knowledge graph based on the fault prediction probability and generate predictive instructions. The knowledge graph includes a static topology network constructed based on historical fault cases of the anti-icing simulation system. The static topology network includes multiple nodes, and each node is used to store at least one mapping relationship between fault prediction probability and execution instruction. The fault response module is used to respond to the fault trigger signal of the anti-icing simulation system and execute the predictive instructions.

Images