Satellite fault diagnosis method and system based on multi-source information fusion

Abstract

The invention provides a satellite fault diagnosis method and system based on multi-source information fusion. The satellite fault diagnosis method comprises the steps that firstly, a multi-source monitoring data set of platform system state data, load system working data and measurement and control link interaction data is generated in the satellite testing process based on an emergency fast response satellite testing system; then performing multi-source information association extraction processing on the multi-source monitoring data set to generate an association feature set reflecting a coupling relationship between systems; then inputting the associated feature set into a pre-trained satellite fault diagnosis model for fault mode analysis processing to obtain a fault feature description containing a fault trigger node and an influence link; a preset fault knowledge base is matched based on the fault feature description, and a fault diagnosis result containing a fault type identifier and a propagation path is generated; and finally, generating a satellite fault handling strategy according to a fault diagnosis result. The method can comprehensively and accurately diagnose the satellite fault and improve the satellite reliability.

Description

Technical Field

[0001] The present application relates to the field of aerospace technology, and in particular to a satellite fault diagnosis method and system based on multi-source information fusion. Background Art

[0002] Modern satellite systems are increasingly complex and highly integrated, operating in orbit for extended periods in the harsh space environment (e.g., extreme temperatures, high-energy particle radiation, and vacuum microgravity). Their reliability and safety are directly linked to the safety of valuable assets and the success of critical space missions. Ground-based environmental simulations during the development phase and comprehensive pre-launch testing provide a true picture of the operating status and health of satellite subsystems, individual units, and even components. This data is typically multi-dimensional, high-frequency, time-series, and large in volume, containing a wealth of status information and potential fault indicators.

[0003] Currently, satellite fault diagnosis primarily relies on a single type of data monitoring and analysis method. For example, only the status data of the platform system is monitored, or only the operating data of the payload system is focused on. However, satellite systems are highly complex coupled systems with close interactions and dependencies between subsystems. Fault diagnosis methods based on a single data source often fail to fully and accurately reflect the actual operating status of the satellite system and struggle to capture potential faults caused by the coupling relationships between systems. Furthermore, existing fault diagnosis methods also have shortcomings in fault mode analysis and fault knowledge base matching, making it impossible to quickly and accurately determine the fault type and propagation path. This results in a lack of targeted and effective fault handling solutions, seriously impacting the reliability and service life of the satellite. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a satellite fault diagnosis method and system based on multi-source information fusion.

[0005] According to a first aspect of the present application, a satellite fault diagnosis method based on multi-source information fusion is provided, the method comprising:

[0006] A multi-source monitoring data set measured during the satellite test process based on the emergency quick response satellite test system, wherein the multi-source monitoring data set includes platform system status data, payload system operating data, and measurement and control link interaction data;

[0007] Performing multi-source information correlation extraction processing on the multi-source monitoring data set to generate a correlation feature set reflecting the coupling relationship between systems;

[0008] Inputting the associated feature set into a pre-trained satellite fault diagnosis model for fault mode analysis to obtain a fault feature description including the fault triggering node and the affected link;

[0009] Based on the fault feature description, a preset fault knowledge base is matched to generate a fault diagnosis result including a fault type identifier and a propagation path;

[0010] A satellite fault handling strategy is generated according to the fault diagnosis result.

[0011] According to the second aspect of the present application, a satellite fault diagnosis system based on multi-source information fusion is provided. The satellite fault diagnosis system based on multi-source information fusion includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the satellite fault diagnosis system based on multi-source information fusion implements the aforementioned satellite fault diagnosis method based on multi-source information fusion.

[0012] According to a third aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed, the aforementioned satellite fault diagnosis method based on multi-source information fusion is implemented.

[0013] According to any of the above aspects, the technical effects of this application are:

[0014] By acquiring multi-source monitoring data sets during satellite operation, including platform system status data, payload system operating data, and measurement and control link interaction data, and performing multi-source information correlation extraction and processing, a comprehensive and in-depth analysis of the coupling relationships between satellite system subsystems is achieved, generating a correlation feature set that accurately reflects the inter-system correlation characteristics. This correlation feature set is then fed into a pre-trained satellite fault diagnosis model for fault mode analysis, accurately identifying the fault triggering node and impacting links, and obtaining a detailed fault feature description. Based on this fault feature description and matching it with a pre-set fault knowledge base, a fault diagnosis result is rapidly generated, including fault type identification and propagation path, providing an accurate basis for fault handling. Finally, a satellite fault handling plan, including positioning coordinates and handling strategies, is generated based on the fault diagnosis results. This achieves an integrated process from fault diagnosis to handling, significantly improving the accuracy and timeliness of satellite fault diagnosis, effectively enhancing satellite reliability and operational efficiency, and reducing losses caused by satellite failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 A flowchart of a satellite fault diagnosis method based on multi-source information fusion provided in an embodiment of the present application.

[0017] Figure 2 A schematic diagram of the component structure of a satellite fault diagnosis system based on multi-source information fusion provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below based on the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of illustration and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate the operations implemented in some embodiments of the embodiments of the present application. It should be understood that the operations of the flowcharts can be implemented out of sequence, and steps without logical context can be reversed or implemented simultaneously. In addition, those skilled in the art can add multiple other operations to the flowcharts or delete multiple operations from the flowcharts under the guidance of the contents of this application.

[0019] Figure 1 The following is a flow chart of a satellite fault diagnosis method and system based on multi-source information fusion provided by an embodiment of the present application. It should be understood that in other embodiments, the order of some steps in the satellite fault diagnosis method based on multi-source information fusion of this embodiment can be shared according to actual needs, or some steps can be omitted or maintained. The detailed steps of the satellite fault diagnosis method based on multi-source information fusion include:

[0020] Step S110: A multi-source monitoring data set is obtained from a satellite test process based on the emergency quick response satellite test system, wherein the multi-source monitoring data set includes platform system status data, payload system operating data, and measurement and control link interaction data.

[0021] In this embodiment, effective satellite fault diagnosis requires comprehensive collection of all types of data generated during satellite testing. As a satellite performs various missions, its platform system, payload system, and measurement and control link generate a large amount of data related to its operational status. Platform system status data primarily reflects the operational status of the satellite platform itself. Attitude control parameters, for example, are key data describing the satellite's attitude in space. These parameters can be acquired in real time via the satellite's onboard attitude sensors. Power supply output parameters reflect the operating status of the satellite's power supply system, including output voltage and current. These parameters indicate whether the power supply is properly supplying energy to the satellite's various systems. Thermal control parameters record the operating status of the satellite's thermal control system, such as the activation status of thermal control devices and the temperature adjustment range, to ensure that the satellite's internal equipment operates within an appropriate temperature environment.

[0022] Payload system operating data primarily relates to the operating status of satellite payload equipment. The payload operating mode reflects the current operating method of the payload equipment, and different missions may require the payload to adopt different operating modes. The data acquisition rate represents the amount of data collected by the payload equipment per unit time, which directly affects the efficiency of satellite information acquisition. The processing delay parameter, which measures the time it takes for the payload equipment to complete data processing, is crucial for missions that require timely data processing.

[0023] Telemetry and control link interaction data describes the communication status between the satellite and the ground control center. The uplink command response time refers to the time interval between the satellite receiving a command from the ground and executing the corresponding operation, reflecting the satellite's responsiveness to ground commands. The downlink data packet loss rate refers to the proportion of data lost during satellite-to-ground transmission, a parameter that directly affects data transmission reliability. The signal strength fluctuation parameter records the changes in the communication signal strength between the satellite and the ground. The stability of signal strength has a significant impact on communication quality. The satellite's data acquisition module can collect this multi-source monitoring data in real time and integrate it into a multi-source monitoring data set.

[0024] Step S120: performing multi-source information association extraction processing on the multi-source monitoring data set to generate a set of association features reflecting the coupling relationship between systems.

[0025] In this embodiment, after obtaining a multi-source monitoring data set, it is necessary to further process the data to mine the coupling relationship between different systems. The purpose of multi-source information association extraction is to extract features that can reflect the relationship between systems from the multi-source monitoring data.

[0026] Step S121: extracting attitude control parameters, power output parameters and thermal control adjustment parameters from the platform system status data as a platform feature subset.

[0027] Among the platform system status data, attitude control parameters, power supply output parameters, and thermal control parameters are key parameters that reflect the platform system's operational status. Attitude control parameters can be acquired by attitude sensors onboard the satellite, such as gyroscopes and accelerometers. These sensors monitor the satellite's attitude changes in space in real time and transmit this data to the data processing module. Power supply output parameters, including output voltage and current, are collected in real time by the power supply monitoring circuit. Thermal control parameters are acquired by thermal control sensors located at key locations within the satellite. These sensors monitor temperature changes in real time and provide feedback to the thermal control system. These parameters are extracted to form a platform feature subset, enabling subsequent analysis of the platform system's relationships with other systems.

[0028] Step S122: extracting load operation mode, data acquisition rate and processing delay parameters from the load system working data as a load feature subset.

[0029] For payload system operating data, the payload operating mode is determined based on the payload device's control instructions, with different operating modes being set for different tasks. The data acquisition rate can be calculated by recording the amount of data collected by the payload device per unit time. For example, the data acquisition rate can be determined by counting the number of data points collected over a period of time. Processing delay parameters can be obtained by recording the difference between data acquisition time and processing completion time. These parameters are extracted to form a subset of payload characteristics, which can be used to analyze the interactions between the payload system and other systems.

[0030] Step S123: extracting uplink command response time, downlink data packet loss rate and signal strength fluctuation parameters from the measurement and control link interaction data as a link feature subset.

[0031] In the TT&C link interaction data, the uplink command response time can be calculated by recording the time the satellite receives the ground command and the time it begins executing the command. The downlink data packet loss rate can be determined by comparing the amount of data sent by the satellite with the amount of data received by the ground. For example, the packet loss rate can be calculated by calculating the difference between the two and the ratio of the amount of data sent. Signal strength fluctuation parameters can be measured by using a signal strength sensor to monitor changes in signal strength in real time and record information such as the amplitude and frequency of signal strength fluctuations. These parameters are extracted to form a link feature subset, which is used to analyze the coupling relationship between the TT&C link and other systems.

[0032] Step S124: Calculate the dynamic correlation between the platform feature subset and the load feature subset, and generate a first correlation feature reflecting the platform support capability and the load working status.

[0033] In this embodiment, in order to gain a deeper understanding of the relationship between the platform system and the payload system, it is necessary to calculate the dynamic correlation between the platform feature subset and the payload feature subset.

[0034] Step S1241: extracting time series data of power output parameters in the platform feature subset and load operation modes in the load feature subset.

[0035] Time series data for power supply output parameters is extracted from the platform feature subset, recording how these parameters change over time. Similarly, time series data for load operating patterns is extracted from the load feature subset, recording how these patterns change over the same time period. By analyzing this time series data, we can understand the dynamic relationship between power supply output and load operating patterns.

[0036] Step S1242: Calculate the matching degree between the power output parameters and the load operation mode at the same time point. The higher the matching degree, the more coordinated the platform power supply capacity is with the load working requirements.

[0037] At the same point in time, analyze the match between the power supply output parameters and the load operating mode. For example, when the load is in a high-power mode, a higher power output is required; whereas, when the load is in a low-power mode, the power output demand is relatively low. By comparing the correspondence between the power supply output parameters and the load operating mode at different points in time, the degree of their match can be calculated. A matching evaluation standard can be set to assign a score based on the degree of compliance between the power supply output parameters and the load operating mode, with a higher score indicating a greater degree of match.

[0038] Step S1243: extracting the change trend data of the thermal control adjustment parameters in the platform feature subset and the processing delay parameters in the load feature subset.

[0039] We extract trend data for thermal control parameters from the platform feature subset to observe their direction and magnitude of change over time. We also extract trend data for processing delay parameters from the load feature subset to understand how these parameters change over time. By analyzing these two sets of trend data, we can uncover potential relationships between thermal control and load processing delay.

[0040] Step S1244: Calculate the synchronization between the rate of change of the thermal control adjustment parameter and the rate of change of the processing delay parameter. The higher the synchronization, the more coordinated the thermal environment control and the load processing efficiency.

[0041] Calculate the rate of change of the thermal control parameter and the processing delay parameter, and compare their changes to assess their synchronization. For example, you can calculate the correlation coefficient between the two rates. The closer the correlation coefficient is to 1, the higher the synchronization. High synchronization means that thermal environment control can better match the load processing efficiency, reducing the impact of temperature changes on load processing.

[0042] Step S1245: combining the matching degree and synchronization to form a first correlation feature reflecting the platform support capability and the load working status.

[0043] The calculated matching degree between the power supply output parameters and the load operating mode, and the synchronization between the change rate of the thermal control parameter and the change rate of the processing delay parameter, are combined to form a first correlation feature. This first correlation feature can comprehensively reflect the coupling relationship between the platform system's support capability for the load system and the load system's operating state.

[0044] Step S125: Calculate the temporal matching degree between the load feature subset and the link feature subset, and generate a second correlation feature reflecting the load data output and the link transmission capability.

[0045] In this embodiment, in order to analyze the relationship between the payload system and the measurement and control link, it is necessary to calculate the temporal matching degree between the payload feature subset and the link feature subset.

[0046] Step S1251: extracting the timestamp alignment data of the data acquisition rate in the load feature subset and the downlink data packet loss rate in the link feature subset.

[0047] Extract the time series data for data acquisition rates from the load feature subset and record the timestamp corresponding to each data acquisition rate. Similarly, extract the time series data for downlink packet loss rates from the link feature subset and record the corresponding timestamps. Align the timestamps of these two sets of data to ensure that the data comparison is at the same time point.

[0048] Step S1252: Calculate the time offset between the peak value of the data acquisition rate and the peak value of the downlink data packet loss rate. The smaller the time offset, the more closely the load data output matches the link transmission capacity.

[0049] In the timestamp-aligned data, find the peak data acquisition rate and the peak downlink packet loss rate. Calculate the time offset between these two peaks, that is, the difference between the two peak times. A smaller time offset indicates that the peak value of the payload data output is well aligned with the changes in the link transmission capacity, and the link can process the payload data in a timely manner, reducing the possibility of data loss.

[0050] Step S1253: extracting the continuous sampling data of the processing delay parameter in the load feature subset and the uplink instruction response time in the link feature subset.

[0051] Continuously sampled data for processing delay parameters is extracted from the load feature subset, and multiple sampled values ​​of the processing delay parameters over a period of time are recorded. Simultaneously, continuously sampled data for uplink command response time is extracted from the link feature subset. By analyzing these two sets of continuously sampled data, the relationship between load processing time and link command response time can be understood.

[0052] Step S1254: Calculate the difference between the mean value of the processing delay parameter and the mean value of the uplink command response time. The smaller the difference, the more coordinated the load processing time and the link command response time.

[0053] Calculate the mean of the processing delay parameter and the mean of the uplink command response time, then calculate the difference between these two means. The smaller the difference, the closer the time it takes for the payload to process data and the time it takes for the link to respond to ground commands, making the entire system more coordinated and improving satellite efficiency.

[0054] Step S1255: combining the time offset and the difference to form a second correlation feature reflecting the load data output and the link transmission capability.

[0055] The calculated time offset and difference are combined to form a second correlation feature, which can comprehensively reflect the coupling relationship between the data output of the payload system and the transmission capacity of the measurement and control link.

[0056] Step S126: Calculate the environmental adaptability correlation between the platform feature subset and the link feature subset, and generate a third correlation feature reflecting the platform operation status and link communication stability.

[0057] In this embodiment, in order to analyze the environmental adaptability relationship between the platform system and the measurement and control link, it is necessary to calculate the environmental adaptability correlation between the platform feature subset and the link feature subset.

[0058] Step S1261: extracting spatial coverage data of the attitude control parameters in the platform feature subset and the signal strength fluctuation parameters in the link feature subset.

[0059] Spatial coverage data for attitude control parameters is extracted from the platform feature subset, i.e., their distribution at different spatial locations. Simultaneously, spatial coverage data for signal strength fluctuation parameters is extracted from the link feature subset, recording how signal strength fluctuations vary at different spatial locations. By analyzing these two sets of spatial coverage data, we can understand the relationship between platform attitude and link signal strength.

[0060] Step S1262: Calculate the deviation between the satellite pointing angle corresponding to the attitude control parameter and the optimal communication angle corresponding to the signal strength fluctuation parameter. The smaller the deviation, the more consistent the platform pointing is with the link communication direction.

[0061] Determine the satellite pointing angle corresponding to the attitude control parameters and the optimal communication angle corresponding to the signal strength fluctuation parameters. Calculate the deviation between these two angles. A smaller deviation indicates that the satellite's pointing direction is closer to the optimal communication direction, which can improve the quality and stability of the link communication.

[0062] Step S1263: extracting the power output parameter in the platform feature subset and the energy consumption data of the uplink command response time in the link feature subset.

[0063] Energy consumption data for power supply output parameters is extracted from the platform feature subset, specifically the power supply's energy output over different time periods. Simultaneously, energy consumption data for uplink command response time is extracted from the link feature subset, recording the energy consumption during uplink command execution. By analyzing these two sets of energy consumption data, we can understand the relationship between platform power supply and link communication energy consumption.

[0064] Step S1264: Calculate the ratio of the redundant power in the power output parameter to the energy consumption corresponding to the uplink command response time. The larger the ratio, the more compatible the platform power redundancy is with the link communication energy consumption requirement.

[0065] Calculate the redundant power within the power output parameters—the remaining power after subtracting the power required for normal operation of the satellite's various systems from the power output. Calculate the energy consumption corresponding to the uplink command response time—the power consumed during uplink command execution. Then, calculate the ratio of redundant power to energy consumption. A larger ratio indicates that the platform's power redundancy can better meet the energy requirements of link communications, ensuring link stability.

[0066] Step S1265: combining the deviation value and the ratio to form a third correlation feature reflecting the platform operation status and link communication stability.

[0067] The calculated deviation value and ratio are combined to form a third correlation feature. The third correlation feature can comprehensively reflect the coupling relationship between the operating status of the platform system and the communication stability of the measurement and control link.

[0068] Step S127: combining the first association feature, the second association feature, and the third association feature to form an association feature set reflecting the coupling relationship between systems.

[0069] In this embodiment, the first correlation feature, the second correlation feature, and the third correlation feature are combined to form a correlation feature set that can fully reflect the coupling relationship between the platform system, the payload system, and the measurement and control link.

[0070] Step S130: inputting the associated feature set into a pre-trained satellite fault diagnosis model to perform fault mode analysis processing to obtain a fault feature description including the fault triggering node and the affected link.

[0071] In this embodiment, after generating the associated feature set, it is input into a pre-trained satellite fault diagnosis model for fault mode analysis. The satellite fault diagnosis model is trained with a large amount of historical data and can analyze and process the associated feature set to identify the hidden fault information.

[0072] Step S131: receiving the associated feature set through the feature coding layer, performing temporal feature extraction processing on the first associated feature, the second associated feature, and the third associated feature, respectively, to generate an encoded feature vector having a time dimension representation.

[0073] The satellite fault diagnosis model consists of a feature encoding layer, a coupling analysis layer, and a fault resolution layer. The feature encoding layer first receives a set of correlation features and performs time series feature extraction on the first, second, and third correlation features. Time series feature extraction involves extracting features from the time series data of these correlation features that reflect their changing patterns. For example, by analyzing the numerical changes and trends of the correlation features at different time points, an encoded feature vector representing the time dimension is generated. This encoded feature vector more effectively represents the temporal changes of the correlation features.

[0074] Step S132: performing inter-system dependency modeling processing on the coded feature vectors through the coupling analysis layer to identify key coupling nodes between the platform system, the payload system and the measurement and control link.

[0075] In this embodiment, after receiving the coded feature vector, the coupling analysis layer performs inter-system dependency modeling on it.

[0076] Step S1321: Perform interactive impact analysis on the platform feature coding, load feature coding, and link feature coding in the coding feature vector through the coupling analysis layer, and calculate the impact weight of the platform feature change on the load feature, the impact weight of the load feature change on the link feature, and the impact weight of the link feature change on the platform feature.

[0077] The coded feature vector includes platform, load, and link feature codes. The coupled analysis layer performs interactive impact analysis on these codes. By analyzing the correlations and changes between different feature codes, it calculates the impact weights of platform feature changes on load features, load feature changes on link features, and link feature changes on platform features. For example, statistical analysis can be used to calculate the correlation coefficients between different feature codes and determine the impact weights based on the correlation coefficients.

[0078] Step S1322: constructing an inter-system dependency graph based on the impact weights, wherein nodes in the dependency graph represent system modules and edges represent impact relationships between modules.

[0079] Based on the calculated influence weights, an inter-system dependency graph is constructed. In this inter-system dependency graph, nodes represent modules in the platform system, payload system, and measurement and control link, while edges represent the influence relationships between modules. Edge weights are determined by the influence weights. This inter-system dependency graph provides a visual representation of the interdependencies between different system modules.

[0080] Step S1323: Calculate the centrality index of each node in the dependency graph. A higher centrality index indicates that the node is more critical in the system interaction.

[0081] In the constructed inter-system dependency graph, calculate the centrality index of each node. Centrality indexes measure the importance of a node in the overall system interaction. Common centrality indexes include degree centrality and betweenness centrality. Degree centrality refers to the number of edges connecting a node, while betweenness centrality refers to the frequency with which a node appears on the shortest path between other nodes. The higher the centrality index, the more critical the node plays in the system interaction.

[0082] Step S1324: Nodes whose centrality index exceeds a preset threshold are selected as key coupling nodes between the platform system, the payload system, and the measurement and control link.

[0083] A preset threshold is set and nodes whose centrality index exceeds the threshold are screened out as key coupling nodes. The state changes of the above key coupling nodes may have a greater impact on other system modules.

[0084] Step S133: performing abnormal feature detection processing based on the key coupling node through the fault analysis layer, and locating the original monitoring data source node corresponding to the abnormal feature as the fault triggering node.

[0085] In this embodiment, after receiving the key coupling node information, the fault analysis layer performs abnormal feature detection processing.

[0086] Step S1331: extracting the characteristic value corresponding to each key coupling node in the coded characteristic vector through the fault analysis layer, and calculating the degree of deviation between the characteristic value and the characteristic mean value under the historical normal operating state.

[0087] Extract the eigenvalues ​​corresponding to each key coupling node from the encoded feature vector and obtain the historical mean eigenvalues ​​of these key coupling nodes under normal operating conditions. Calculate the degree of deviation between the current eigenvalue and the historical mean eigenvalue, either by calculating the difference or ratio between the two. The greater the deviation, the more abnormal the state of the key coupling node.

[0088] Step S1332: Filter out key coupling nodes whose deviation exceeds a preset threshold as abnormal coupling nodes.

[0089] A preset threshold is set, and key coupling nodes whose deviation exceeds the threshold are screened out as abnormal coupling nodes. The above abnormal coupling nodes may be potential locations where faults occur.

[0090] Step S1333: Tracking the original data mapping relationship of the abnormal coupling node during the feature encoding process, and determining the original monitoring data type and specific parameters corresponding to the abnormal coupling node.

[0091] By tracking the raw data mapping relationship of abnormal coupling nodes during the feature encoding process, the corresponding raw monitoring data type and specific parameters can be found. For example, an abnormal coupling node may correspond to an attitude control parameter in the platform system status data. Through further analysis, the specific attitude control parameter with an abnormality can be determined.

[0092] Step S1334: Based on the original monitoring data type and specific parameters, locate the satellite hardware module or software function module that generates the data as a fault trigger node.

[0093] In this embodiment, based on the determined raw monitoring data type and specific parameters, the satellite hardware module or software functional module generating the data can be located. For example, if the abnormal raw monitoring data is a parameter related to satellite attitude control, the satellite's hardware architecture and software logic can be used to track the hardware modules responsible for attitude control, such as the gyroscope and accelerometer, or the software functional modules corresponding to the attitude control algorithm, and identify these modules as the fault triggering nodes. In this way, through in-depth analysis of the abnormal coupling node, the source of the fault can be accurately identified.

[0094] Step S134: According to the connection path of the fault triggering node in the coupling relationship, the system module sequence of the abnormal feature propagation is tracked as the influencing link.

[0095] In this embodiment, after determining the fault trigger node, it is necessary to further analyze the propagation path of the fault. Due to the coupling relationship between systems, the abnormal characteristics of the fault trigger node will propagate to other system modules along a certain path. By viewing the dependency graph between systems, the connection edges between the fault trigger node and other modules are found, and according to the influence relationship represented by these connection edges, the system module sequence where the abnormal characteristics are propagated is tracked. For example, if the fault trigger node is a hardware module in the platform system, it is connected to a module of the load system through an edge in the dependency graph, and the edge indicates that the abnormality of the platform module will affect the load module, then it can be determined that the load module is the next node where the abnormal characteristics are propagated. According to the above method, by tracing in sequence, the system module sequence where the abnormal characteristics are propagated can be obtained, and this system module sequence is the influence link. The influence link can clearly show how the fault propagates between different system modules starting from the trigger node.

[0096] Step S135: combining the fault triggering node and the affected link to form a fault feature description including the fault triggering node and the affected link.

[0097] In this embodiment, the determined fault triggering nodes and impacting links are combined to form a fault signature. This is a comprehensive description of the satellite fault condition, which not only identifies the source of the fault (fault triggering node) but also illustrates the fault propagation path within the system (impacting links).

[0098] Step S140: Based on the fault feature description, a preset fault knowledge base is matched to generate a fault diagnosis result including a fault type identifier and a propagation path.

[0099] In this embodiment, after obtaining the fault feature description, it needs to be matched with the preset fault knowledge base to generate the fault diagnosis result. The fault knowledge base is a database that stores a large number of historical fault cases. Each case contains information such as the fault type, fault triggering node, and affected link.

[0100] Step S141: extracting the module identifier of the fault triggering node and the module sequence of the affected link from the fault feature description.

[0101] The module identifier of the fault-triggering node is extracted from the fault signature. This module identifier uniquely identifies the specific module where the fault occurred. The module sequence affecting the link is also extracted, recording the order in which the fault propagates through the system. This information is crucial for fault matching.

[0102] Step S142: Query the fault knowledge base for historical fault cases that match the module identifier and module sequence.

[0103] Using the extracted module identifiers of the fault-triggering node and the module sequence of the affected link, a query is performed in the fault knowledge base. This information is compared with corresponding information in historical fault cases to identify matching historical fault cases. This matching process can employ similarity calculations, such as calculating the consistency of module identifiers and the similarity of module sequences. Cases with higher similarities are more likely to be matches.

[0104] Step S143: extracting the fault type label in the historical fault case as the fault type identifier of the current fault.

[0105] After finding a matching historical fault case, the fault type label is extracted from the historical fault case and used as the fault type identifier of the current fault. The fault type identifier can clearly indicate the type of the current fault, such as power failure, communication failure, attitude control failure, etc.

[0106] Step S144: extracting the abnormal propagation path in the historical fault case as the propagation path of the current fault.

[0107] Similarly, the abnormal propagation path is extracted from the matched historical fault cases and used as the propagation path of the current fault. The abnormal propagation path can further verify the accuracy of the affected link.

[0108] Step S145: combining the fault type identifier and the propagation path to form a fault diagnosis result including the fault type identifier and the propagation path.

[0109] The extracted fault type identifier and propagation path are combined to form a fault diagnosis result, which can fully describe the type and propagation of the current fault.

[0110] Step S150: generating a satellite fault handling solution including positioning coordinates and handling strategies according to the fault diagnosis result.

[0111] In this embodiment, after obtaining the fault diagnosis result, it is necessary to generate a satellite fault processing solution based on the result to effectively process the fault.

[0112] Step S151: extracting the module identifier of the fault triggering node from the fault diagnosis result, querying the satellite cabin layout corresponding to the module identifier, and obtaining the three-dimensional coordinates of the module in the satellite cabin as positioning coordinates.

[0113] The module identifier of the fault-triggering node is extracted from the fault diagnosis results, and then the satellite cabin layout diagram is consulted. The satellite cabin layout diagram details the location of each module within the satellite cabin. Using the module identifier, the corresponding module can be found in the layout diagram and its three-dimensional coordinates within the satellite cabin are obtained. These three-dimensional coordinates are the location coordinates of the fault-triggering node, which can help ground control personnel accurately determine the location of the fault.

[0114] Step S152: querying a preset handling strategy library according to the fault type identifier, and extracting emergency operation instructions and repair priorities that match the fault type identifier.

[0115] Based on the fault type identifier in the fault diagnosis results, the pre-set handling strategy library is queried. This handling strategy library stores emergency operation instructions and repair priorities for different fault types. The fault type identifier is used to match the handling strategy library, extracting the emergency operation instructions and repair priorities corresponding to the current fault type. The emergency operation instructions guide ground control personnel to take appropriate measures to mitigate the impact of the fault, such as adjusting the satellite's operating mode or restarting related modules. Repair priorities help determine the order in which fault repairs should be performed, ensuring that limited resources are prioritized for addressing critical faults.

[0116] Step S153: Predicting the extension direction of the fault within a preset time window according to the propagation path, and generating an operation constraint condition including an extension direction restriction.

[0117] Based on the propagation path from the fault diagnosis results, the fault's propagation direction within a preset time window is predicted. Because fault propagation exhibits certain regularities, analyzing the propagation path allows a rough prediction of the next system module the fault may affect. Based on the predicted propagation direction, operational constraints are generated, including restrictions on the propagation direction. For example, if a fault is predicted to propagate to a specific system module, troubleshooting operations must be performed to avoid further impact on that module. Appropriate operational constraints are established to ensure the safety and effectiveness of the fault handling process.

[0118] Step S154: combining the positioning coordinates, emergency operation instructions, repair priorities, and operation constraints to form a satellite fault handling solution including positioning coordinates and handling strategies.

[0119] The acquired positioning coordinates, extracted emergency operation instructions, repair priorities, and generated operational constraints are combined to form a satellite fault handling plan. This satellite fault handling plan comprehensively considers factors such as the fault location, handling measures, repair priorities, and operational constraints. It can provide ground control personnel with a comprehensive and effective fault handling guide, helping them to quickly and accurately handle satellite faults and ensure the normal operation of the satellite.

[0120] In this embodiment, the satellite fault diagnosis model training process enables the model to learn the fault modes and patterns between systems, thereby improving the accuracy of fault diagnosis. The satellite fault diagnosis model includes a feature encoding layer, a coupling analysis layer, and a fault resolution layer.

[0121] First, a large amount of historical satellite operational data is collected, including multi-source monitoring data sets and corresponding fault labels. The multi-source monitoring data sets include platform system status data, payload system operating data, and measurement and control link interaction data. The fault labels clearly identify the fault type and fault triggering node for each data sample. This data is preprocessed, including data cleaning and normalization, to ensure data quality and consistency. Data cleaning removes noise and outliers, while normalization converts data of different dimensions to the same scale, facilitating model processing.

[0122] Then, the parameters of the feature encoding layer, coupling analysis layer, and fault resolution layer of the satellite fault diagnosis model are initialized. Random initialization can be used to set initial weights and biases for the neurons in each layer. These initial parameters will be continuously adjusted during the training process to ensure that the model better fits the training data.

[0123] Next, the preprocessed training data is fed into the satellite fault diagnosis model for forward propagation. At the feature encoding layer, time series features are extracted from the input multi-source monitoring data to generate encoded feature vectors. At the coupling analysis layer, inter-system dependency relationships are modeled on the encoded feature vectors to identify key coupling nodes. At the fault analysis layer, abnormal feature detection is performed based on key coupling nodes, and fault prediction results are output.

[0124] Compare the model's fault predictions with the actual fault labels and calculate the loss function. The loss function measures the degree of discrepancy between the model's predictions and the true values. Common loss functions include the cross-entropy loss function, which can help the model's predictions more closely approximate the true values.

[0125] Backpropagation is performed based on the calculated loss function. The backpropagation process calculates the gradient of the loss function with respect to each model parameter through a chain derivation method. The gradient represents the degree of influence of each parameter on the loss function. Based on the size and direction of the gradient, the model parameters can be adjusted to gradually reduce the loss function.

[0126] Based on the gradient calculated by backpropagation, an optimization algorithm (such as stochastic gradient descent) is used to update the model parameters. The optimization algorithm adjusts the model's weights and biases at a certain learning rate based on the gradient information, making the model perform better on the training data.

[0127] Repeat the above steps until the training termination criteria are met. This can be when a preset number of training rounds is reached or when the loss function converges to a smaller value. Through continuous training, the model learns the failure modes and patterns between systems, improving the accuracy of fault diagnosis.

[0128] Finally, the trained satellite fault diagnosis model is evaluated using test data. Test data is a separate dataset from the training data. Model performance is evaluated by calculating metrics such as precision and recall on the test data. If the model performs as expected, it can be used in actual satellite fault diagnosis. If not, training parameters must be adjusted or retraining is necessary.

[0129] During data collection, data desensitization techniques are employed for potentially privacy-sensitive data. For example, sensitive data, such as satellite operator identity information, is replaced or masked to remove personally identifiable information without compromising fault diagnosis and analysis. During data transmission, data is encrypted using encryption algorithms, such as symmetric encryption algorithms. Only authorized parties with the decryption key can decrypt the data. Data is stored in a secure database with strict access control, ensuring only authorized personnel have access. Furthermore, data is regularly backed up to prevent loss or corruption. These technical measures effectively protect privacy-sensitive data and prevent data leaks.

[0130] Figure 2The satellite fault diagnosis system 100 based on multi-source information fusion shown includes: a processor 1001 and a memory 1003. The processor 1001 and the memory 1003 are connected, for example, via a bus 1002. Optionally, the satellite fault diagnosis system 100 based on multi-source information fusion may further include a transceiver 1004, which may be used for data exchange between the server and other servers, such as data transmission and / or data reception. It should be noted that in actual scheduling, the number of transceivers 1004 is not limited to one, and the structure of the satellite fault diagnosis system 100 based on multi-source information fusion does not constitute a limitation on the embodiments of the present application.

[0131] The processor 1001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 1001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0132] The bus 1002 may include a path for transmitting information between the above components. The bus 1002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 1002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0133] The memory 1003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store program code and can be read by a computer, without limitation here.

[0134] The memory 1003 is used to store program codes for executing the embodiments of the present application, and the execution is controlled by the processor 1001. The processor 1001 is used to execute the program codes stored in the memory 1003 to implement the steps shown in the above method embodiments.

[0135] An embodiment of the present application provides a computer-readable storage medium having program code stored thereon. When the program code is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0136] It should be understood that, although each operation step is characterized by directed lines in the flow chart of the embodiment of the present application, the execution order of these steps is not limited to the order covered by the directed lines. Unless otherwise clearly stated herein, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flow chart can be performed in other orders based on demand. In addition, some or all of the steps in each flow chart can include multiple sub-steps or multiple stages according to actual implementation scenario, and some or all of these sub-steps or stages can be executed in the same stage, and each sub-step or stage in these sub-steps or stages can also be executed in different stages respectively. Under different scenarios in the execution stage, the execution order of these sub-steps or stages can be flexibly configured based on demand, and the embodiment of the present application does not limit this.

[0137] The above is only an optional implementation method for some implementation scenarios of this application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the solution of this application, other similar implementation methods based on the technical ideas of this application also fall within the protection scope of the embodiments of this application.

Claims

1. A satellite fault diagnosis method based on multi-source information fusion, characterized in that: The method comprises: A multi-source monitoring data set generated during the satellite testing process based on the emergency quick response satellite test system, wherein the multi-source monitoring data set includes platform system status data, payload system operating data, and measurement and control link interaction data; Performing multi-source information correlation extraction processing on the multi-source monitoring data set to generate a correlation feature set reflecting the coupling relationship between systems; Inputting the associated feature set into a pre-trained satellite fault diagnosis model for fault mode analysis to obtain a fault feature description including the fault triggering node and the affected link; Based on the fault feature description, a preset fault knowledge base is matched to generate a fault diagnosis result including a fault type identifier and a propagation path; A satellite fault handling strategy is generated according to the fault diagnosis result.

2. The satellite fault diagnosis method based on multi-source information fusion according to claim 1 is characterized in that: The performing multi-source information correlation extraction processing on the multi-source monitoring data set to generate a correlation feature set reflecting the coupling relationship between systems includes: Extracting attitude control parameters, power output parameters, and thermal control adjustment parameters from the platform system state data as a platform feature subset; Extracting load operation mode, data acquisition rate and processing delay parameters from the load system working data as a load feature subset; Extracting uplink command response time, downlink data packet loss rate and signal strength fluctuation parameters from the measurement and control link interaction data as a link feature subset; Calculating a dynamic correlation between the platform feature subset and the payload feature subset to generate a first correlation feature reflecting the platform support capability and the payload working status; Calculating a temporal matching degree between the load feature subset and the link feature subset to generate a second correlation feature reflecting the load data output and the link transmission capacity; Calculating the environmental adaptability correlation between the platform feature subset and the link feature subset to generate a third correlation feature reflecting the platform operation status and link communication stability; The first association feature, the second association feature, and the third association feature are combined to form an association feature set reflecting the coupling relationship between systems.

3. The satellite fault diagnosis method based on multi-source information fusion according to claim 2 is characterized in that: The calculating the dynamic correlation between the platform feature subset and the payload feature subset to generate a first correlation feature reflecting the platform support capability and the payload working status includes: extracting time series data of power supply output parameters in the platform feature subset and load operation modes in the load feature subset; Calculate the degree of matching between the power supply output parameters and the load operating mode at the same time point. The higher the matching degree, the more coordinated the platform power supply capacity is with the load working requirements. Extracting change trend data of thermal control adjustment parameters in the platform feature subset and processing delay parameters in the load feature subset; Calculate the synchronization between the rate of change of thermal control parameters and the rate of change of processing delay parameters. The higher the synchronization, the more coordinated the thermal environment control and load processing efficiency. The matching degree and synchronization are combined to form a first correlation feature reflecting the platform support capability and the load working status.

4. The satellite fault diagnosis method based on multi-source information fusion according to claim 2 is characterized in that: The calculating the timing matching degree between the load feature subset and the link feature subset to generate a second correlation feature reflecting the load data output and the link transmission capability includes: Extracting timestamp alignment data of a data acquisition rate in the load feature subset and a downlink data packet loss rate in the link feature subset; Calculate the time offset between the peak data acquisition rate and the peak downlink data packet loss rate. A smaller time offset indicates a closer match between the payload data output and the link transmission capacity. Extracting continuous sampling data of a processing delay parameter in the load feature subset and an uplink instruction response time in the link feature subset; Calculate the difference between the mean processing delay parameter and the mean uplink command response time. The smaller the difference, the more coordinated the load processing time and the link command response time. The time offset and the difference are combined to form a second correlation feature reflecting the load data output and the link transmission capacity.

5. The satellite fault diagnosis method based on multi-source information fusion according to claim 2 is characterized in that: The calculating the environmental adaptability correlation between the platform feature subset and the link feature subset to generate a third correlation feature reflecting the platform operation status and link communication stability includes: Extracting spatial coverage data of attitude control parameters in the platform feature subset and signal strength fluctuation parameters in the link feature subset; Calculate the deviation between the satellite pointing angle corresponding to the attitude control parameter and the optimal communication angle corresponding to the signal strength fluctuation parameter. The smaller the deviation, the more consistent the platform pointing is with the link communication direction. extracting energy consumption data of power output parameters in the platform feature subset and uplink command response time in the link feature subset; Calculate the ratio of the redundant power in the power output parameters to the energy consumption corresponding to the uplink command response time. A larger ratio indicates a better fit between the platform's power redundancy and the link communication energy consumption requirements. The deviation value and the ratio are combined to form a third correlation feature reflecting the platform operation status and link communication stability.

6. The satellite fault diagnosis method based on multi-source information fusion according to claim 2 is characterized in that: The satellite fault diagnosis model includes a feature coding layer, a coupling analysis layer and a fault analysis layer; The step of inputting the associated feature set into a pre-trained satellite fault diagnosis model for fault mode analysis to obtain a fault feature description including a fault triggering node and an affected link includes: Receiving the associated feature set through the feature coding layer, performing time series feature extraction processing on the first associated feature, the second associated feature, and the third associated feature, respectively, to generate an encoded feature vector having a time dimension representation; The coupling analysis layer performs inter-system dependency modeling on the coded feature vectors to identify key coupling nodes between the platform system, the payload system, and the measurement and control link; Perform abnormal feature detection processing based on the key coupling node through the fault analysis layer, and locate the original monitoring data source node corresponding to the abnormal feature as the fault trigger node; According to the connection path of the fault triggering node in the coupling relationship, the system module sequence of the abnormal feature propagation is tracked as the impact link; The fault triggering node and the affected link are combined to form a fault feature description including the fault triggering node and the affected link.

7. The satellite fault diagnosis method based on multi-source information fusion according to claim 6 is characterized in that: The method of performing inter-system dependency modeling on the coded feature vectors through the coupling analysis layer to identify key coupling nodes between the platform system, the payload system, and the measurement and control link includes: Performing interactive impact analysis on the platform feature code, load feature code, and link feature code in the code feature vector through a coupling analysis layer, and calculating the impact weight of the platform feature change on the load feature, the impact weight of the load feature change on the link feature, and the impact weight of the link feature change on the platform feature; Constructing an inter-system dependency graph based on the impact weights, wherein nodes in the dependency graph represent system modules and edges represent impact relationships between modules; Calculate the centrality index of each node in the dependency graph, where a higher centrality index indicates that the node is more critical in the system interaction; Nodes whose centrality index exceeds the preset threshold are selected as key coupling nodes between the platform system, payload system and measurement and control link.

8. The satellite fault diagnosis method based on multi-source information fusion according to claim 6 is characterized in that: The abnormal feature detection process is performed based on the key coupling node through the fault analysis layer, and the original monitoring data source node corresponding to the abnormal feature is located as the fault trigger node, including: Extracting the eigenvalues ​​corresponding to the key coupling nodes in the coded feature vector through the fault analysis layer, and calculating the degree of deviation between the eigenvalues ​​and the mean values ​​of the features under the historical normal operating state; Screen out key coupling nodes whose deviation exceeds a preset threshold as abnormal coupling nodes; Tracking the original data mapping relationship of the abnormal coupling node during the feature encoding process, and determining the original monitoring data type and specific parameters corresponding to the abnormal coupling node; According to the original monitoring data type and specific parameters, the satellite hardware module or software function module that generates the data is located as a fault trigger node.

9. The satellite fault diagnosis method based on multi-source information fusion according to claim 1, characterized in that: The method of matching a preset fault knowledge base based on the fault feature description to generate a fault diagnosis result including a fault type identifier and a propagation path includes: Extracting the module identifier of the fault triggering node and the module sequence of the affected link from the fault feature description; Searching the fault knowledge base for historical fault cases that match the module identifier and module sequence; Extracting the fault type label in the historical fault case as the fault type identifier of the current fault; Extracting the abnormal propagation path in the historical fault case as the propagation path of the current fault; Combining the fault type identifier and the propagation path to form a fault diagnosis result including the fault type identifier and the propagation path; Furthermore, generating a satellite fault handling solution including positioning coordinates and handling strategies based on the fault diagnosis results includes: Extracting the module identifier of the fault triggering node from the fault diagnosis result, querying the satellite cabin layout corresponding to the module identifier, and obtaining the three-dimensional coordinates of the module identifier in the satellite cabin as positioning coordinates; Querying a preset handling strategy library according to the fault type identifier, and extracting emergency operation instructions and repair priorities that match the fault type identifier; Predicting the extension direction of the fault within a preset time window based on the propagation path, and generating an operation constraint condition including an extension direction restriction; The positioning coordinates, emergency operation instructions, repair priorities and operation constraints are combined to form a satellite fault handling solution including positioning coordinates and handling strategies.

10. A satellite fault diagnosis system based on multi-source information fusion, characterized in that: The invention comprises a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by the processor, the satellite fault diagnosis method based on multi-source information fusion according to any one of claims 1 to 9 is implemented.

Images