Method and device for determining satellite field map, storage medium and electronic equipment

Abstract

This invention discloses a method, apparatus, storage medium, and electronic device for determining a satellite domain map. The method includes: determining fault parameters included in the telemetry parameters of a target satellite; determining single-parameter fault entities based on the fault parameters, wherein a single-parameter fault entity is a fault entity caused by a fault of a single parameter; determining system-level fault entities based on the single-parameter fault entities, wherein a system-level fault entity is a fault entity caused by a fault of multiple parameters; determining the correlation relationships between the parameters included in the fault parameters; and constructing a domain map of the target satellite based on the single-parameter fault entities, system-level fault entities, and correlation relationships. This application solves the problem of inaccurate domain map determination, thereby improving the accuracy of domain map determination.

Description

Technical Field

[0001] This application relates to the field of communications, and more specifically, to a method, apparatus, storage medium, and electronic device for determining a satellite domain map. Background Technology

[0002] In related technologies, general maps are usually created for satellite fault detection. However, these maps cannot fully take into account the special properties of satellite telemetry parameters and the unique requirements of fault detection scenarios. In other words, general maps cannot be correlated with satellite parameters and adapted to fault detection mechanisms, and cannot be directly applied to actual satellite fault detection.

[0003] This indicates that there is a technical problem in the related technologies where the defined domain map is inaccurate.

[0004] There is currently no effective solution to the aforementioned problems in the relevant technologies. Summary of the Invention

[0005] This application provides a method, apparatus, storage medium, and electronic device for determining satellite domain maps, in order to at least solve the technical problem of inaccurate domain map determination in related technologies.

[0006] According to one aspect of the embodiments of this application, a method for determining a satellite domain map is provided, comprising: determining fault parameters included in the telemetry parameters of a target satellite; determining single-parameter fault entities based on the fault parameters, wherein the single-parameter fault entities are fault entities caused by a single parameter fault; determining system-level fault entities based on the single-parameter fault entities, wherein the system-level fault entities are fault entities caused by multiple parameter faults; determining the correlation relationships between the parameters included in the fault parameters; and constructing a domain map of the target satellite based on the single-parameter fault entities, the system-level fault entities, and the correlation relationships.

[0007] In an exemplary embodiment, determining a system-level fault entity based on the single-parameter fault entity includes: determining an entity pair included in the single-parameter fault entity, wherein the entity pair includes a first single-parameter fault entity and a second single-parameter fault entity; determining the association strength between the first single-parameter fault entity and the second single-parameter fault entity; and determining the first single-parameter fault entity and the second single-parameter fault entity as system-level fault entities if the association strength is greater than or equal to a second preset threshold.

[0008] In an exemplary embodiment, determining the association strength between the first single-parameter fault entity and the second single-parameter fault entity includes: determining the co-occurrence strength between the first single-parameter fault entity and the second single-parameter fault entity; determining the conditional probability that the second single-parameter fault entity will fail within a first preset time window in which the first single-parameter fault entity fails; determining a target time difference based on a first time when the first single-parameter fault entity fails and a second time when the second single-parameter fault entity fails; determining an order parameter based on the first single-parameter fault entity and the second single-parameter fault entity; and determining the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter.

[0009] In an exemplary embodiment, determining the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter includes: determining a first product of the co-occurrence strength and a first parameter; determining a second product of the conditional probability and a second parameter; determining a third product of the target time difference and a third parameter; determining a fourth product of the order parameter and a fourth parameter; and determining the sum of the first product, the second product, the third product, and the fourth product as the association strength.

[0010] In an exemplary embodiment, determining the co-occurrence intensity between a first single-parameter fault entity and a second single-parameter fault entity includes: determining a first number of times the first single-parameter fault entity and the second single-parameter fault entity simultaneously fail within a second preset time window, and a second number of times the first single-parameter fault entity fails and a third number of times the second single-parameter fault entity fails within a first preset period; determining a first sum of the second number and the third number; determining a first difference between the first sum and the first number; and determining the ratio of the first number to the first difference as the co-occurrence intensity.

[0011] In an exemplary embodiment, determining a target time difference based on the first time when the first single-parameter faulty entity fails and the second time when the second single-parameter faulty entity fails includes: determining a pair of single-parameter faulty entities based on the first time and the second time, wherein the pair of single-parameter faulty entities includes a first sub-entity and a second sub-entity, the first sub-entity and the second sub-entity are entities that fail consecutively, and the time when the first sub-entity fails is earlier than the time when the second sub-entity fails; for each pair of single-parameter faulty entities, performing the following operations to determine the initial time difference corresponding to the pair of single-parameter faulty entities: determining the first start time of the first sub-entity failure included in the first time, and determining the second start time of the second sub-entity failure included in the second time, and determining the difference between the second start time and the first start time as the initial time difference; and determining the average of the initial time differences of all pairs of single-parameter faulty entities as the target time difference.

[0012] In an exemplary embodiment, determining an order parameter based on the first single-parameter fault entity and the second single-parameter fault entity includes: determining a fourth number of times within a second preset time window that the first single-parameter fault entity fails before the second single-parameter fault entity fails; determining a second sum of a fifth number of times the first single-parameter fault entity fails and a sixth number of times the second single-parameter fault entity fails; and determining a second ratio of the fourth number of times to the second sum as the order parameter.

[0013] In an exemplary embodiment, determining a single-parameter fault entity based on the fault parameters includes: determining a first parameter and a second parameter included in the fault parameters, wherein the difference between the generation time of the first parameter and the generation time of the second parameter is less than a preset difference, and the first parameter and the second parameter are the same parameter; merging the first parameter and the second parameter to obtain a first merged parameter; and determining the first merged parameter and other parameters included in the fault parameters other than the first parameter and the second parameter as the single-parameter fault entity.

[0014] In an exemplary embodiment, determining a single-parameter fault entity based on the fault parameters includes: determining a multi-dimensional feature vector for each parameter included in the fault parameters; determining the similarity between any two sub-feature vectors included in the multi-dimensional feature vectors; determining a target similarity greater than a preset similarity included in the similarity; determining a third parameter and a fourth parameter corresponding to the target similarity; merging the third parameter and the fourth parameter to obtain a second merged parameter; and determining the second merged parameter and other parameters included in the fault parameters other than the third parameter and the fourth parameter as the single-parameter fault entity.

[0015] In an exemplary embodiment, determining the correlation between parameters included in the fault parameters includes: determining a support count of an initial itemset based on the fault parameters, wherein the initial itemset includes N types of itemsets, the i-th type of itemset includes i parameters, i=1,2,3...N, where N is the number of fault parameters, and each type of itemset is different; determining frequent itemsets within each third preset time window based on the support count, wherein the frequent itemsets are the itemsets included in the initial itemset; determining a global frequent itemset based on the frequent itemsets; and determining that the correlation exists between the parameters included in the global frequent itemset.

[0016] In an exemplary embodiment, determining a global frequent itemset based on the frequent itemset includes: determining the weight of each target itemset included in the frequent itemset in each third preset time window; determining the target weight of the target itemset, wherein the target weight is the sum of the weights of the target itemset in each third preset time window; and determining the target itemset whose sum is greater than a preset weight threshold as the global frequent itemset.

[0017] In an exemplary embodiment, after constructing the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the association relationship, the method further includes: receiving a second telemetry parameter; if the domain map includes the second telemetry parameter, determining a first single-parameter fault entity corresponding to the second telemetry parameter; determining the fault type of the first single-parameter fault entity; if the fault type is an existing type, storing the second telemetry parameter; if the fault type is a newly occurring type, updating the single-parameter fault entity in the domain map based on the second telemetry parameter to obtain an updated single-parameter fault entity; and if the updated single-parameter fault entity exists in the domain map, updating the system-level fault entity based on the updated single-parameter fault entity to obtain an updated domain map.

[0018] In an exemplary embodiment, after constructing the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the association relationship, the method further includes: determining a second single-parameter fault entity corresponding to the target system-level fault entity included in the domain map; and determining that the target system-level fault entity has a fault if all the second single-parameter fault entities have faults.

[0019] According to another aspect of the embodiments of this application, a satellite domain map determination apparatus is also provided, comprising: a first determination module, configured to determine fault parameters included in the telemetry parameters of a target satellite; a second determination module, configured to determine single-parameter fault entities based on the fault parameters, wherein the single-parameter fault entity is a fault entity caused by a single parameter fault; a third determination module, configured to determine system-level fault entities based on the single-parameter fault entities, wherein the system-level fault entities are fault entities caused by multiple parameter faults; a fourth determination module, configured to determine the correlation relationships between the parameters included in the fault parameters; and a construction module, configured to construct a domain map of the target satellite based on the single-parameter fault entities, the system-level fault entities, and the correlation relationships.

[0020] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed by a processor.

[0021] According to another aspect of the embodiments of this application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform the steps in any of the method embodiments described above.

[0022] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to perform the steps of any of the above method embodiments through the computer program.

[0023] This application allows for the identification of fault parameters included in the telemetry parameters of a target satellite. These fault parameters can then be used to identify single-parameter fault entities. Furthermore, these single-parameter fault entities can be used to determine system-level fault entities. Single-parameter fault entities are those caused by a single-parameter fault, while system-level fault entities are those caused by multiple-parameter faults. Based on these single-parameter and system-level fault entities, the relationships between the parameters included in the fault parameters can also be determined. Finally, the domain map of the target satellite can be constructed using these relationships, single-parameter fault entities, and system-level fault entities. By employing a two-layer fault mode (single-parameter fault mode entities and system-level fault mode entities) and a three-level entity structure (telemetry parameters, single-parameter fault entities, and system-level fault entities), this approach can retain fine-grained information about lower-level parameter anomalies while integrating the commonalities and coupling characteristics of multi-parameter faults at a higher level. This results in better hierarchical interpretability of the reasoning results obtained using the domain map, clearly presenting the evolutionary logic from parameter anomalies to system-level faults. Therefore, the problem of inaccurate determination of the domain map can be solved, thereby improving the accuracy of the determination of the domain map. Attached Figure Description

[0024] Figure 1 This is a schematic diagram illustrating an application scenario of a method for determining a satellite domain map according to an embodiment of this application;

[0025] Figure 2 This is a flowchart illustrating an optional method for determining a satellite domain map according to an embodiment of this application;

[0026] Figure 3 This is a schematic diagram illustrating the construction of a domain map based on fault detection results according to an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of the domain map according to an embodiment of this application;

[0028] Figure 5 This is a flowchart of domain graph knowledge reasoning and application according to an embodiment of this application;

[0029] Figure 6 This is a flowchart illustrating the method for determining the satellite domain map in this optional example;

[0030] Figure 7 This is a diagram showing the relationship between various satellite parameters, instantiated according to a specific embodiment of this application;

[0031] Figure 8 This is a structural block diagram of an optional satellite domain map determination device according to an embodiment of this application;

[0032] Figure 9This is a computer system architecture block diagram of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0034] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0035] According to one aspect of the embodiments of this application, a method for determining a domain map is provided. Optionally, in this embodiment, the above-described method for determining a domain map may be applied, but is not limited to, to applications such as... Figure 1 The diagram shows the architecture of a satellite communication system. This satellite communication system may include a satellite 101, a terminal 102, and a gateway station 103.

[0036] In this disclosure, satellite 101 is an entity used for transmitting or receiving signals, and there can be multiple satellites. This disclosure does not limit the specific technology or equipment form used in the satellites.

[0037] In this disclosure, terminal 102 refers to a processing device within the satellite coverage beam range for communicating with a satellite. For example, the terminal can be a car, smart car, mobile phone, wearable device, tablet computer, etc., equipped with satellite communication capabilities. This disclosure does not limit the specific technology or device form used in the terminal. It should be noted that... Figure 1 The example uses two terminal devices 102.

[0038] In one embodiment of this disclosure, gateway station 103 is connected to satellite 101.

[0039] In this embodiment, the gateway station 103 is a ground-based node in a satellite communication system used for transmitting and receiving data. This embodiment does not limit the specific technology or equipment form employed by the gateway station.

[0040] It is understood that the satellite communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions provided in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in this disclosure are also applicable to similar technical problems.

[0041] Figure 2 This is a flowchart illustrating an optional method for determining a satellite domain map according to an embodiment of this application, as shown below. Figure 2 As shown, the process of this method may include the following steps:

[0042] Step S202: Determine the fault parameters included in the telemetry parameters of the target satellite;

[0043] The method for determining the satellite domain map in this embodiment can be applied to satellite health monitoring and fault diagnosis, data mining, and can be specifically applied to scenarios where fault diagnosis domain maps are constructed based on satellite fault diagnosis results, and the correlation between parameters and fault modes, and between parameters, is mined.

[0044] In related technologies, it is difficult to trace the causal chain of faults from diagnosed fault warnings and to delve into the causes of fault occurrence. This is due to the difficulty in obtaining accurate fault labels during satellite system operation in orbit, the complexity and diversity of satellite fault types, and the lack of understanding of the internal systems of satellites. In other words, current technical research mainly focuses on training relevant models based on labeled data to identify satellite fault modes, while research on satellite fault diagnosis using unlabeled data is relatively scarce. In addition, related technologies usually rely on known physical models for system modeling. When there are unknown factors within the satellite system, there will be certain limitations in application. Therefore, ordinary general graphs cannot correctly show the correlation and influence between parameters, and thus cannot accurately determine satellite faults.

[0045] To at least partially address the aforementioned technical problems, this application can complete satellite system fault diagnosis even in the absence of data annotation. By incorporating a set prohibited / allowed access list database, expert experience is fed back into the fault detection results, fully utilizing the knowledge of relevant personnel. Specifically, this application can deeply analyze the potential connections between satellite fault modes among various parameters. Through time-series analysis of the detected satellite fault mode data, combined with telemetry data of different parameters, multi-dimensional correlation analysis is performed from a time perspective to uncover deep-seated relationships between parameters, thereby revealing the causes that may lead to satellite faults. In this embodiment, telemetry data streams of various telemetry parameters can be received from the target satellite system. Then, multimodal unsupervised algorithms such as clustering algorithms and statistical methods can be used to automatically identify potential abnormal patterns in the unannotated telemetry data stream, providing semantic nodes for downstream map modeling. Telemetry parameters can include solar array current, power bus voltage, onboard computer temperature, etc. For multiple discrete detection algorithms, a majority voting method can be used to determine the fault detection results. That is, finally, the fault parameters and their corresponding telemetry data can be determined from the telemetry parameters included in the telemetry data stream.

[0046] Step S204: Determine a single-parameter fault entity based on the fault parameters, wherein the single-parameter fault entity is a fault entity caused by a single parameter fault;

[0047] In the above embodiments, after determining the fault parameters, the discrete anomaly fragments detected from the telemetry parameters using unsupervised algorithms can be transformed into fault event entities with clear engineering significance. Figure 3 This is a schematic diagram of the domain map construction based on fault detection results according to an embodiment of this application, such as... Figure 3 As shown, the original fault fragments of fault parameters can be subjected to time alignment, interpolation, and parameter normalization to eliminate data noise and ensure data quality. Furthermore, metadata such as the source device and sampling frequency for each fault parameter can be added, providing a foundation for subsequent data mining and fault pattern recognition. Subsequently, time proximity merging and multi-feature similarity merging algorithms can be used to transform the discrete abnormal fragments identified by the fault detection algorithm into single-parameter fault pattern entities with clear engineering significance, and automatically extract attribute information such as fault start time, end time, duration, and statistical features.

[0048] Step S206: Determine system-level fault entities based on the single-parameter fault entities, wherein the system-level fault entities are fault entities caused by multiple parameter faults;

[0049] In the above embodiments, after identifying the single-parameter fault entity, the single-parameter fault entity can be further mined to identify system-level fault entities, which are used to characterize the collaborative fault phenomenon caused by the same reason and spanning multiple subsystems. That is, by analyzing the correlation between single-parameter fault entities, system-level fault modes can be identified, which are system faults caused by multiple parameter anomalies, and system-level fault entities can be constructed.

[0050] Step S208: Determine the correlation between the parameters included in the fault parameters;

[0051] In the above embodiments, the frequent itemset mining method improved by time-delay windowing can be used to mine the correlation relationships among various parameters within a satellite. The time-delay sliding window can be understood as employing an improved Apnioni algorithm, dynamically capturing short-term and long-term correlations between parameters through time-delay windowing technology, and identifying potential dependencies between parameters. Frequent itemset mining can be understood as mining frequently occurring parameter combinations within each window, extracting potential correlations between parameters through association rules, and supporting the mining of dynamically changing telemetry data features. The improved Apnioni algorithm can solve the problems of ignoring time order, ignoring dynamic changes, and weakening abnormal linkages between parameters in related technologies. Ignoring time order can be understood as relying solely on the "number of occurrences" to measure the frequency of itemsets, failing to capture sequential relationships or temporal clustering effects. Ignoring dynamic changes can be understood as traditional frequent itemset mining failing to discover situations where some itemsets are frequent in a certain time period but not frequent overall. Weakening abnormal linkages between parameters can be understood as several parameters frequently correlated in a short period, but the number of correlations globally is low; traditional frequent itemset mining may ignore such correlations. Therefore, a frequent itemset mining method based on time-based windowing can be introduced, which can support the collection of frequent information related to parameters through dynamic windowing.

[0052] Step S210: Construct the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the correlation relationship.

[0053] In the above embodiments, after determining the single-parameter fault entities, system-level fault entities, and the relationships between parameters, a domain graph can be constructed using these entities and relationships. A blank graph can be constructed first, and then entities and relationships can be added. Entities can include telemetry parameters, single-parameter fault entities, and system-level fault entities, while relationships encompass the connections between parameters and fault entities, as well as the relationships between different fault entities. Figure 4 This is a schematic diagram of the domain map according to an embodiment of this application, such as... Figure 4As shown, it can include parameter nodes, single-parameter fault entity nodes and system-level fault entity nodes. The relationship between parameters can be represented by the associated values, and the relationship between single-parameter fault entity nodes can be represented by co-occurrence values ​​and time-series values.

[0054] In the above embodiments, an industrial-grade, domain-level knowledge graph is created based on disordered and fragmented fault detection algorithm fragments. Inference can be completed without prior knowledge of the satellite's internal physical information, enabling the transformation from low-value, high-noise raw telemetry fragments into high-quality, reasonable, and computable domain knowledge assets. This embodiment can generate a knowledge graph structure with two-layer fault modes and a three-level entity structure. The graph schema and the final visualization presentation are explicit and standardized. Moreover, the two-layer fault modes created in the domain graph allow the reasoning process to be naturally layered, enabling rapid identification of the trigger source from the bottom layer and directly revealing the complete link from signal anomaly to single-parameter fault to system-level fault. Furthermore, through the full-process automation from data access, fault detection, graph construction to knowledge reasoning, the graph can be endowed with the dynamic evolution capability of continuous learning and self-updating.

[0055] This application allows for the identification of fault parameters included in the telemetry parameters of a target satellite. These fault parameters can then be used to identify single-parameter fault entities. Furthermore, these single-parameter fault entities can be used to determine system-level fault entities. Single-parameter fault entities are those caused by a single-parameter fault, while system-level fault entities are those caused by multiple-parameter faults. Based on these single-parameter and system-level fault entities, the relationships between the parameters included in the fault parameters can also be determined. Finally, the domain map of the target satellite can be constructed using these relationships, single-parameter fault entities, and system-level fault entities. By employing a two-layer fault mode (single-parameter fault mode entities and system-level fault mode entities) and a three-level entity structure (telemetry parameters, single-parameter fault entities, and system-level fault entities), this approach can retain fine-grained information about lower-level parameter anomalies while integrating the commonalities and coupling characteristics of multi-parameter faults at a higher level. This results in better hierarchical interpretability of the reasoning results obtained using the domain map, clearly presenting the evolutionary logic from parameter anomalies to system-level faults. Therefore, the problem of inaccurate determination of the domain map can be solved, thereby improving the accuracy of the determination of the domain map.

[0056] Optionally, the entity performing the above steps may be a terminal, a server, a client, or other devices with similar processing capabilities, but is not limited to these.

[0057] In an exemplary embodiment, determining a system-level fault entity based on the single-parameter fault entity includes: determining an entity pair included in the single-parameter fault entity, wherein the entity pair includes a first single-parameter fault entity and a second single-parameter fault entity; determining the association strength between the first single-parameter fault entity and the second single-parameter fault entity; and determining the first single-parameter fault entity and the second single-parameter fault entity as system-level fault entities if the association strength is greater than or equal to a second preset threshold.

[0058] In the above embodiments, system-level fault entity extraction can be accomplished by constructing a fault mode relationship feature matrix, that is, for any two single-parameter fault entities (i.e., the first single-parameter fault entity and the second single-parameter fault entity mentioned above) forming an entity pair ( , Each of these can be used to calculate a comprehensive relational feature vector. The correlation strength is fully quantified, and then a threshold K (i.e. the second preset threshold mentioned above) can be selected. The system-level fault mode (i.e. the system-level fault entity mentioned above) is constructed using the single-parameter fault mode group (i.e. the first single-parameter fault entity and the second single-parameter fault entity mentioned above) whose correlation strength is greater than the threshold K.

[0059] By identifying system-level fault entities, complex faults caused by the combined effects of multiple parameter failures can be more accurately identified, thereby improving the overall accuracy of fault diagnosis. Furthermore, timely identification of system-level faults helps to accelerate response times, reduce the impact of faults on satellite operations, and improve maintenance efficiency. In addition, the construction of system-level fault entities not only identifies the fault itself but also reveals its root cause—that is, which combinations of single-parameter failures might lead to more serious consequences. This provides a clear direction for tracing and repairing the root cause of the fault.

[0060] In an exemplary embodiment, determining the association strength between the first single-parameter fault entity and the second single-parameter fault entity includes: determining the co-occurrence strength between the first single-parameter fault entity and the second single-parameter fault entity; determining the conditional probability that the second single-parameter fault entity will fail within a first preset time window in which the first single-parameter fault entity fails; determining a target time difference based on a first time when the first single-parameter fault entity fails and a second time when the second single-parameter fault entity fails; determining an order parameter based on the first single-parameter fault entity and the second single-parameter fault entity; and determining the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter.

[0061] In the above embodiments, the quantified value of the association strength can be calculated by weighting the co-occurrence strength and temporal relationship features (target time difference, order parameter, and conditional probability). That is, the co-occurrence strength between the first single-parameter fault entity and the second single-parameter fault entity can be determined first. The first preset time window for the failure to occur Second single-parameter fault entity Conditional probability of failure ,in, That is, the first single-parameter fault entity. The second single-parameter fault entity can also determine the target time difference based on the first time of the first single-parameter fault entity and the second time of the second single-parameter fault entity, as well as the order parameter based on the first single-parameter fault entity and the second single-parameter fault entity. Finally, the co-occurrence strength, conditional probability, target time difference, and order parameter after weighted summation are determined as the association strength.

[0062] By calculating co-occurrence intensity, it is possible to identify which faulty entities tend to appear simultaneously or sequentially. Entity pairs with high co-occurrence intensity reveal the fault propagation path or fault cluster within the system to a great extent, providing clues for precise fault location. Moreover, the introduction of conditional probability improves fault prediction capability, that is, determining the probability that the first single-parameter faulty entity will trigger the second single-parameter faulty entity within a first preset time window helps predict the probability of fault occurrence under given conditions. In addition, the target time difference quantifies the speed of fault propagation, and the order parameter also ensures the correct ordering of the fault causal chain.

[0063] By comprehensively utilizing the co-occurrence intensity and temporal relationship characteristics among single-parameter fault modes, this embodiment can determine the possible system-level fault mode to which they belong, thereby achieving accurate modeling of complex multi-parameter coupling relationships and accurate identification of system-level fault modes.

[0064] In an exemplary embodiment, determining the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter includes: determining a first product of the co-occurrence strength and a first parameter; determining a second product of the conditional probability and a second parameter; determining a third product of the target time difference and a third parameter; determining a fourth product of the order parameter and a fourth parameter; and determining the sum of the first product, the second product, the third product, and the fourth product as the association strength.

[0065] In the above embodiments, the correlation strength F can be calculated using the following formula: ,in, That is, the first parameter mentioned above. That is, the second parameter mentioned above. That is, the third parameter mentioned above. This refers to the fourth parameter mentioned above. That is, the first product mentioned above. That is, the second product mentioned above. That is, the third product mentioned above. This is the fourth product mentioned above, and the correlation strength is the sum of the first, second, third, and fourth products.

[0066] In this embodiment, by multiplying co-occurrence strength, conditional probability, target time difference, and order parameters by their respective weighting factors and then summing the results, a comprehensive score, namely the correlation strength, can be obtained. This score accurately quantifies the degree of correlation between two faulty entities, providing a more solid data foundation for fault mode identification and analysis. Furthermore, based on the analysis of correlation strength, common fault modes and potential fault propagation paths can be identified, providing key data points for building fault prediction models. By monitoring parameter entities with high correlation strength, potential system-level faults can be predicted in advance, enabling preventative maintenance to avoid fault occurrence or taking measures at the initial stage of a fault to reduce losses.

[0067] In an exemplary embodiment, determining the co-occurrence intensity between a first single-parameter fault entity and a second single-parameter fault entity includes: determining a first number of times the first single-parameter fault entity and the second single-parameter fault entity simultaneously fail within a second preset time window, and a second number of times the first single-parameter fault entity fails and a third number of times the second single-parameter fault entity fails within a first preset period; determining a first sum of the second number and the third number; determining a first difference between the first sum and the first number; and determining the ratio of the first number to the first difference as the co-occurrence intensity.

[0068] In the above embodiments, the Jaccard coefficient can be used to evaluate the co-occurrence intensity, and the formula for the co-occurrence intensity is as follows: Here, the "nearest occurrence count" can be understood as the number of times the first single-parameter fault entity and the second single-parameter fault entity simultaneously fail within the second preset time window (i.e., the first count mentioned above). The "total count" can be understood as the sum of the number of times the first single-parameter fault entity fails within the second preset time window (i.e., the second count mentioned above) and the number of times the second single-parameter fault entity fails within the second preset time window (i.e., the third count mentioned above) (i.e., the first sum value mentioned above). "Nearest" can be defined as within t minutes of each other (i.e., the second preset time window mentioned above). By calculating the relative ratio of the frequency of two fault entities simultaneously occurring within the same time window (the first count) to their independent occurrence frequencies (the second and third counts), fault combinations that tend to co-occur in actual operation can be identified.

[0069] In an exemplary embodiment, determining a target time difference based on the first time when the first single-parameter faulty entity fails and the second time when the second single-parameter faulty entity fails includes: determining a pair of single-parameter faulty entities based on the first time and the second time, wherein the pair of single-parameter faulty entities includes a first sub-entity and a second sub-entity, the first sub-entity and the second sub-entity are entities that fail consecutively, and the time when the first sub-entity fails is earlier than the time when the second sub-entity fails; for each pair of single-parameter faulty entities, performing the following operations to determine the initial time difference corresponding to the pair of single-parameter faulty entities: determining the first start time of the first sub-entity failure included in the first time, and determining the second start time of the second sub-entity failure included in the second time, and determining the difference between the second start time and the first start time as the initial time difference; and determining the average of the initial time differences of all pairs of single-parameter faulty entities as the target time difference.

[0070] In the above embodiments, regarding the multiple first times when the first single-parameter faulty entity fails and the multiple second times when the second single-parameter faulty entity fails, since there will be multiple... Prior to Given the current situation, we can first identify a pair of consecutively failing single-parameter entities by using the first and second time points. Each single-parameter failure entity pair includes a first sub-entity and a second sub-entity, where the first sub-entity fails earlier than the second sub-entity. For each single-parameter failure entity pair, we can determine the average difference between their start times (i.e., the first and second start times mentioned above), thus determining the initial time difference for each pair. The average of these initial time differences is then used as the target time difference.

[0071] Through this embodiment, the calculation of the target time difference enables the maintenance team to accurately grasp the average propagation time of a fault from one parameter entity to another. This is crucial for real-time monitoring of system health, predicting the scope of fault impact, and planning emergency response measures. In other words, by quantifying the speed of fault propagation, actions can be taken more promptly and effectively to mitigate the negative impact of faults on the overall operation of the system.

[0072] In an exemplary embodiment, determining an order parameter based on the first single-parameter fault entity and the second single-parameter fault entity includes: determining a fourth number of times within a second preset time window that the first single-parameter fault entity fails before the second single-parameter fault entity fails; determining a second sum of a fifth number of times the first single-parameter fault entity fails and a sixth number of times the second single-parameter fault entity fails; and determining a second ratio of the fourth number of times to the second sum as the order parameter.

[0073] In the above embodiments, order consistency (i.e., the aforementioned order parameters) can be understood as... Prior to The proportion of the number of occurrences to the total number of occurrences can be determined by first determining the number of times the first single-parameter fault entity fails before the second single-parameter fault entity fails within the second preset time window (i.e., the fourth occurrence mentioned above), and then determining the total number of occurrences, which is the sum of the number of times the first single-parameter fault entity fails (i.e., the fifth occurrence mentioned above) and the number of times the second single-parameter fault entity fails (i.e., the sixth occurrence mentioned above) (i.e., the second sum mentioned above). The ratio of the fourth occurrence to the second sum is the order parameter.

[0074] In this embodiment, the determination of the order parameter can be used as one of the input features of the fault prediction model. It can reflect the temporal sequence tendency of fault events, which helps the model learn and predict the propagation path of faults. Especially in complex scenarios where fault events frequently alternate, it can accurately predict the next possible location of the fault.

[0075] In an exemplary embodiment, determining a single-parameter fault entity based on the fault parameters includes: determining a first parameter and a second parameter included in the fault parameters, wherein the difference between the generation time of the first parameter and the generation time of the second parameter is less than a preset difference, and the first parameter and the second parameter are the same parameter; merging the first parameter and the second parameter to obtain a first merged parameter; and determining the first merged parameter and other parameters included in the fault parameters other than the first parameter and the second parameter as the single-parameter fault entity.

[0076] In the above embodiments, for fault scenarios such as instantaneous spikes and long-term slow drift, an aggregation process can be used to extract single-parameter fault entities. That is, strategy one can be time proximity merging, which can be mainly used to merge the same fault event that is broken due to noise interference. That is, if there are two segments of telemetry data in the fault parameters that are the same parameter (i.e., the first parameter and the second parameter mentioned above), and the time interval between the two segments (i.e., the difference between the generation time of the first parameter and the generation time of the second parameter mentioned above) is less than T (i.e., the preset difference mentioned above), then the two segments can be merged into a candidate event (i.e., the first merging parameter mentioned above) and determined as a single-parameter fault entity. After that, other parameters in the fault parameters other than the first parameter and the second parameter can also be determined as single-parameter fault entities.

[0077] By employing time correlation analysis and parameter merging, the misidentification of the same fault event as multiple independent events can be avoided, thereby reducing false alarms and false negatives and improving the accuracy of fault detection. Furthermore, the constructed single-parameter fault entity not only includes fault parameters but also organizes the time and parameter information of the fault event through a merging strategy, aiding in understanding the overall picture of the fault event and providing a clearer context for fault analysis and diagnosis. In addition, parameter merging reduces the complexity of subsequent fault mode analysis and system-level fault entity construction, enabling the algorithm to process data more efficiently and accelerate fault mode identification and system-level fault determination.

[0078] In an exemplary embodiment, determining a single-parameter fault entity based on the fault parameters includes: determining a multi-dimensional feature vector for each parameter included in the fault parameters; determining the similarity between any two sub-feature vectors included in the multi-dimensional feature vectors; determining a target similarity greater than a preset similarity included in the similarity; determining a third parameter and a fourth parameter corresponding to the target similarity; merging the third parameter and the fourth parameter to obtain a second merged parameter; and determining the second merged parameter and other parameters included in the fault parameters other than the third parameter and the fourth parameter as the single-parameter fault entity.

[0079] In the above embodiments, strategy two can be multi-feature similarity merging, which can discover the same fault that recurs periodically or intermittently. That is, for segments that may not be temporally adjacent, a multi-dimensional feature vector of each parameter in the fault parameters can be determined. The multi-dimensional feature vector can include time-domain features, frequency-domain features, morphological features, etc. Morphological features can be understood as the image features of the parameters, such as median, maximum, minimum, etc. The cosine similarity of every two sub-feature vectors in the multi-dimensional feature vector is calculated. If there is a target similarity greater than a preset similarity, it can be determined that they are the same type of fault. Then, the third and fourth parameters corresponding to the target similarity can be merged to obtain the second merged parameter, which is determined as a single-parameter fault entity. After that, the other parameters in the fault parameters other than the third and fourth parameters can also be determined as single-parameter fault entities.

[0080] Through this embodiment, the two-layer structure of single-parameter fault mode entity and system-level fault mode entity can integrate the commonalities and coupling characteristics of multi-parameter faults at a higher level while retaining the fine-grained information of parameter anomalies at the lower level. The two-level model can make the reasoning results have better hierarchical interpretability and clearly present the evolution logic from parameter anomalies to system-level faults.

[0081] In an exemplary embodiment, determining the correlation between parameters included in the fault parameters includes: determining a support count of an initial itemset based on the fault parameters, wherein the initial itemset includes N types of itemsets, the i-th type of itemset includes i parameters, i=1,2,3...N, where N is the number of fault parameters, and each type of itemset is different; determining frequent itemsets within each third preset time window based on the support count, wherein the frequent itemsets are the itemsets included in the initial itemset; determining a global frequent itemset based on the frequent itemsets; and determining that the correlation exists between the parameters included in the global frequent itemset.

[0082] In the above embodiments, a frequent itemset mining method based on time-delay windowing can be used to mine the correlation between various parameters within the satellite: a time window can be divided according to a specified length. A sliding window with a time delay factor of m is used to obtain the window. ... (i.e., the third preset window mentioned above), for each third preset window, a transaction set can be constructed separately, that is, the signals within a fixed interval t can be regarded as a transaction, where , If the i-th fault parameter indicates a fault mode, then the first window... The transactions include: ,in, Representative window Within each time interval, a transaction database is constructed based on this. : ].

[0083] In the above embodiments, a transaction database can be utilized. First, determine L1 as a set of frequent 1-itemsets. The frequent 1-itemsets and their support counts are shown in Table 1. Itemset and itemsets The frequent 1-itemset set consists of support counts of 4, 2, and 2 respectively; the frequent 21-itemset set and its support count are shown in Table 2. , Itemset , and itemsets , This forms a frequent 2-itemset set with support counts of 1, 2, and 1 respectively. We can then continue to evaluate frequent n-itemset sets until we reach the entire set, which will form the initial itemset.

[0084] Table 1

[0085]

[0086] Table 2

[0087]

[0088] In the above embodiments, the support count corresponding to each itemset included in the initial itemset within each third preset window can be determined to be equal to a preset threshold m. Itemsets with support counts greater than or equal to the preset threshold m can be identified as frequent itemsets. Finally, global frequent itemsets can be determined based on these frequent itemsets, and the relationships between parameters included in the global frequent itemsets can be identified. For each window... After processing is complete, slide the window until all windows have been processed. Example: If the preset threshold m is 3, then the itemset corresponding to a support count of 4... It can be identified as a frequent itemset. If the preset threshold m is 2, then the itemset corresponding to a support count of 2 is... Itemset Itemset , and the itemset corresponding to support count 4 Frequent itemsets can be identified.

[0089] This embodiment demonstrates a frequent itemset mining method based on time-delay windowing to identify potential relationships between parameters. Addressing the strong temporal sequence of satellite data, the windowing mechanism allows for refined analysis of data within each window, uncovering local features, dynamically expanding the window, and capturing short-term and long-term correlations within the data. Compared to traditional frequent itemset mining algorithms, the improved method is more adaptable to dynamic changes and complexity in data, exhibiting high accuracy and processing power.

[0090] In an exemplary embodiment, determining a global frequent itemset based on the frequent itemset includes: determining the weight of each target itemset included in the frequent itemset in each third preset time window; determining the target weight of the target itemset, wherein the target weight is the sum of the weights of the target itemset in each third preset time window; and determining the target itemset whose sum is greater than a preset weight threshold as the global frequent itemset.

[0091] In the above embodiments, by mining each target itemset included in the frequent itemsets within each third preset time window, the weights of the frequent itemsets within each third preset time window can be obtained, as shown in Table 3. Then, the sum of the weights of the target itemsets in each third preset time window (i.e., the target weights mentioned above) can be determined, i.e., the frequent patterns in each window are summarized according to their weights. The summation results (i.e., the target weights mentioned above) that are higher than the threshold n (i.e., the preset weight thresholds mentioned above) are regarded as global frequent itemsets.

[0092] Table 3

[0093]

[0094] By analyzing existing satellite fault case data, the correlations between telemetry parameters can be deeply explored. First, fault data and telemetry data from different satellite systems are collected and organized. Using a time-delay sliding window mining method, potential dependencies between short-term satellite telemetry parameters are identified. Then, weighted fusion is used to infer long-term dependencies between satellite telemetry parameters. By flexibly dividing the window, both short-term and long-term correlations can be captured simultaneously, effectively adapting to the strong temporal sequence and rapid dynamic changes characteristic of satellite internet data.

[0095] In an exemplary embodiment, after constructing the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the association relationship, the method further includes: receiving a second telemetry parameter; if the domain map includes the second telemetry parameter, determining a first single-parameter fault entity corresponding to the second telemetry parameter; determining the fault type of the first single-parameter fault entity; if the fault type is an existing type, storing the second telemetry parameter; if the fault type is a newly occurring type, updating the single-parameter fault entity in the domain map based on the second telemetry parameter to obtain an updated single-parameter fault entity; and if the updated single-parameter fault entity exists in the domain map, updating the system-level fault entity based on the updated single-parameter fault entity to obtain an updated domain map.

[0096] In the above embodiments, after the domain graph is constructed, deep reasoning can be performed on the graph using graph theory algorithms or knowledge reasoning engines. Based on knowledge reasoning, the causes of the fault and the causal chain can be further discovered. Figure 5 This is a flowchart of domain graph knowledge reasoning and application according to an embodiment of this application, such as... Figure 5 As shown, the process includes the following steps:

[0097] Step S502, Begin;

[0098] Step S504: Activate the single-parameter fault node;

[0099] Step S506: Determine whether it belongs to an existing fault mode. If yes, proceed to step S508; otherwise, proceed to step S510.

[0100] Step S508: Add to the current single-parameter fault mode;

[0101] Step S510: Add a single-parameter fault mode;

[0102] Step S512: Add a system-level fault mode;

[0103] Step S514, dynamic weight adjustment;

[0104] Step S516, reverse reasoning;

[0105] Step S518, output the result;

[0106] Step S520, End.

[0107] In the above embodiments, when the received second telemetry parameter is an existing (i.e., a single-parameter node in the domain graph) single-parameter node... Then each node's The state at a certain time t can be represented as In this context, fault activation can be understood as a failure occurring when the observed / received current parameter (i.e., the second telemetry parameter is fault activated). The corresponding parameter (the second telemetry parameter) is marked as the fault starting point, and... Assign a value of 1 as the starting point of the inference chain. Then, the first single-parameter fault entity corresponding to the second telemetry parameter can be determined, and the current node... (Second telemetry parameters) Stored fault modes { The fault type of the first single-parameter fault entity is determined. If the fault type already exists, the second telemetry parameter can be stored directly. If a new fault mode occurs for the current node (i.e., the fault type is the first occurrence of the type), then the {...} parameter needs to be updated. This process involves obtaining updated single-parameter fault entities, updating system-level fault entities based on these entities, accumulating fault node cases, and obtaining an updated domain graph. Furthermore, after the entities are dynamically updated, the relationships and weights also need to be adjusted and updated accordingly. The updates to the system-level fault mode-parameter fault mode relationship and the parameter-parameter relationship are consistent with the method used to construct the domain graph, and will not be elaborated here. Finally, the generated inference chain, along with information such as each node and relationship strength, can be output to provide a complete causal path, facilitating fault analysis and strategy formulation.

[0108] In this embodiment, graph-driven multi-level reasoning can be completed, automatically discovering system-level fault chains, breaking through the limitations of single-level relationships, and finally outputting an interpretable and traceable complete causal chain; it can also support dynamic adaptive reasoning, with the reasoning chain adaptively optimizing with new data and remaining effective in the long term.

[0109] In an exemplary embodiment, after constructing the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the association relationship, the method further includes: determining a second single-parameter fault entity corresponding to the target system-level fault entity included in the domain map; and determining that the target system-level fault entity has a fault if all the second single-parameter fault entities have faults.

[0110] In the above embodiments, each system-level failure mode (That is, the target system-level fault entities mentioned above) all contain several single-parameter fault nodes (that is, the second single-parameter fault entities mentioned above). ,in, This can indicate that parameter 1 is in a system-level fault mode. In the middle, failure mode The trigger is determined by the decision function. control: = ,in, =1 can be understood as triggering a system-level failure mode. ,only All values ​​are 1, indicating that the target system-level faulty entity is confirmed to be faulty; otherwise, they are 0.

[0111] In this embodiment, the graph reasoning application updates the graph information in real time to adapt to real-time changes during system operation, ensuring the timeliness and effectiveness of fault chain detection. Simultaneously, the existence of two-level fault modes ensures that the reasoning process gradually derives from underlying parameter anomalies to system-level faults, achieving root cause tracing and explanation.

[0112] The determination of the satellite domain map in the embodiments of this application will be explained below with reference to optional examples.

[0113] Figure 6 This is a flowchart illustrating the method for determining the satellite neighborhood map in this optional example, such as... Figure 6 As shown, the process of determining the satellite domain map may include the following steps:

[0114] Step S602, Begin;

[0115] Step S604, satellite parameter telemetry data;

[0116] Step S606, Multimodal fault detection;

[0117] Step S608: Constructing a domain map based on fault detection results;

[0118] Step S610: Extract information.

[0119] Step S612, End.

[0120] In the above embodiments, data can be received first, namely the telemetry data streams of various telemetry parameters of the access satellite; then, fault detection can be performed, applying existing multimodal unsupervised learning algorithms such as density clustering and statistical learning to automatically identify parameter-level faults based on the telemetry signals of the parameters; then, a graph can be constructed, identifying and recording single-parameter fault entities based on the fault detection results, and innovatively mining system-level fault entities based on single-parameter fault entities to construct a three-level entity structure of parameter-single-parameter fault entity-system-level fault entity; finally, knowledge reasoning can be achieved, mining information such as fault root cause chains based on the created domain graph of the target satellite.

[0121] In the above embodiment, taking the fault data of 100 telemetry parameters of a satellite as an example, the steps are as follows: Step 1: Multimodal fault detection, for the input time series signal, a multimodal algorithm is used to detect its fault status; Step 2: Domain knowledge graph construction, based on specific parameters and corresponding fault cases, for... Figure 4 Instantiate the graph constructed in the middle, Figure 7 This involves instantiating a satellite parameter relationship diagram based on a specific embodiment of this application, including fault parameters, single-parameter fault entities, and system-level fault entities. Step three: Knowledge reasoning and fault chain discovery. Based on the abnormal information of "slowly increasing onboard computer temperature," the onboard computer temperature parameter can be located. From the correlation information in the graph, the onboard computer temperature parameter is associated with the power bus voltage and solar array current parameters, and the abnormal pattern of these two parameters is associated with a higher-order system-level anomaly—power subsystem instability. Verification shows that the power bus voltage and solar array current parameters are simultaneously abnormal, and the root cause of this chain lies in the instability of the power subsystem.

[0122] This optional example demonstrates how a two-level fault entity modeling approach—comprising single-parameter fault entities and system-level fault entities—can preserve fine-grained information about parameter anomalies at the lower level while integrating the commonalities and coupling characteristics of multi-parameter faults at a higher level. The two-level entity model enhances the hierarchical interpretability of the inference results, clearly presenting the evolutionary logic from parameter anomalies to system-level faults. Ultimately, a domain graph is constructed, connecting parameters, single-parameter fault entities, and system-level fault entities. Relationship patterns between parameters, single-parameter fault entities, and system-level fault entities are then mined. A graph theory model is used to organize the relational structure of telemetry data, visualizing the dependencies and interactions between different parameters.

[0123] In the aforementioned embodiments, related technologies often discover the correlations between parameters through actual physical models inside the satellite. However, obtaining these models is costly and difficult, and without them, the satellite system remains a "black box." This application, through a correlation mining mechanism based on fault data, can deeply uncover the implicit relationships between satellite telemetry parameters. Many telemetry parameters exhibit complex, non-explicit correlations. In the absence of labeled fault accumulation, multi-layered fault modes can be constructed based on unsupervised algorithm fault detection results, revealing deep correlations between single-parameter fault entities and system-level fault entities, as well as between parameters. This mechanism, through association rule mining and time-series data analysis, can reveal hidden causal relationships, dependencies, and potential fault modes between parameters based on fragmented, unlabeled telemetry signal segments. Furthermore, many related technologies create general graphs without proper adaptation to satellite parameter correlation and fault detection mechanisms, making them unsuitable for practical satellite fault detection. The domain graph proposed in this application supports the visualization of correlations between various telemetry parameters in the satellite system, providing intuitive and easily understood relationship diagrams. This graph integrates three levels of entities: satellite telemetry parameters, single-parameter level fault entities, and multi-parameter level fault entities, as well as two levels of relationships: single-parameter fault entity-system-level fault entities and parameter-parameter relationships. When parameter anomalies are detected, it supports deep reasoning based on the domain graph to discover fault propagation chains. Furthermore, this application introduces graph reasoning and insight extraction techniques into the satellite fault detection process, making the fault diagnosis process more interpretable. In traditional machine learning or fault detection methods, many models are often "black boxes," making it difficult to explain their inference processes and results. However, through a parameter relationship graph based on graph reasoning, fault detection can not only yield results but also provide specific, traceable parameter paths and correlation descriptions.

[0124] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0125] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / random access memory (RAM), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0126] According to another aspect of the embodiments of this application, a satellite territory map determining apparatus is also provided. This satellite territory map determining apparatus can be used to implement the satellite territory map determining method provided in the above embodiments, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0127] Figure 8 This is a structural block diagram of an optional satellite territory map determining device according to an embodiment of this application, such as... Figure 8 As shown, the apparatus for determining the satellite's geographic area map includes:

[0128] The first determining module 802 is used to determine the fault parameters included in the telemetry parameters of the target satellite;

[0129] The second determining module 804 is used to determine a single-parameter fault entity based on the fault parameters, wherein the single-parameter fault entity is a fault entity caused by a single parameter fault.

[0130] The third determining module 806 is used to determine a system-level fault entity based on the single-parameter fault entity, wherein the system-level fault entity is a fault entity caused by multiple parameter faults.

[0131] The fourth determining module 808 determines the correlation between the parameters included in the fault parameters;

[0132] The construction module 820 is used to construct the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the correlation relationship.

[0133] In an exemplary embodiment, the third determining module 806 can determine a system-level fault entity based on the single-parameter fault entity in the following manner: determining an entity pair included in the single-parameter fault entity, wherein the entity pair includes a first single-parameter fault entity and a second single-parameter fault entity; determining the correlation strength between the first single-parameter fault entity and the second single-parameter fault entity; and determining the first single-parameter fault entity and the second single-parameter fault entity as system-level fault entities if the correlation strength is greater than or equal to a second preset threshold.

[0134] In an exemplary embodiment, the third determining module 806 can determine the association strength between the first single-parameter fault entity and the second single-parameter fault entity in the following manner: determining the co-occurrence strength between the first single-parameter fault entity and the second single-parameter fault entity; determining the conditional probability of the second single-parameter fault entity failing within a first preset time window in which the first single-parameter fault entity fails; determining a target time difference based on the first time when the first single-parameter fault entity fails and the second time when the second single-parameter fault entity fails; determining an order parameter based on the first single-parameter fault entity and the second single-parameter fault entity; and determining the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter.

[0135] In an exemplary embodiment, the third determining module 806 can determine the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter in the following manner: determining a first product of the co-occurrence strength and a first parameter; determining a second product of the conditional probability and a second parameter; determining a third product of the target time difference and a third parameter; determining a fourth product of the order parameter and a fourth parameter; the sum of the first product, the second product, the third product, and the fourth product is determined as the association strength.

[0136] In an exemplary embodiment, the third determining module 806 can determine the co-occurrence intensity between the first single-parameter fault entity and the second single-parameter fault entity in the following manner: determining the first number of times the first single-parameter fault entity and the second single-parameter fault entity simultaneously fail within a second preset time window, and the second number of times the first single-parameter fault entity fails and the third number of times the second single-parameter fault entity fails within the first preset period; determining the first sum of the second number and the third number; determining the first difference between the first sum and the first number; and determining the ratio of the first number to the first difference as the co-occurrence intensity.

[0137] In an exemplary embodiment, the third determining module 806 can determine a target time difference based on the first time when the first single-parameter faulty entity fails and the second time when the second single-parameter faulty entity fails, in the following manner: determining a pair of single-parameter faulty entities based on the first time and the second time, wherein the pair of single-parameter faulty entities includes a first sub-entity and a second sub-entity, the first sub-entity and the second sub-entity are entities that fail consecutively, and the time when the first sub-entity fails is earlier than the time when the second sub-entity fails; for each pair of single-parameter faulty entities, performing the following operations to determine the initial time difference corresponding to the pair of single-parameter faulty entities: determining the first start time of the first sub-entity failure included in the first time, and determining the second start time of the second sub-entity failure included in the second time, and determining the difference between the second start time and the first start time as the initial time difference; and determining the average of the initial time differences of all pairs of single-parameter faulty entities as the target time difference.

[0138] In an exemplary embodiment, the third determining module 806 can determine the order parameter based on the first single-parameter fault entity and the second single-parameter fault entity in the following manner: determining a fourth number of times the first single-parameter fault entity fails earlier than the second single-parameter fault entity within a second preset time window; determining a second sum of a fifth number of times the first single-parameter fault entity fails and a sixth number of times the second single-parameter fault entity fails; and determining a second ratio of the fourth number to the second sum as the order parameter.

[0139] In an exemplary embodiment, the second determining module 804 can determine a single-parameter fault entity based on the fault parameters in the following manner: determining a first parameter and a second parameter included in the fault parameters, wherein the difference between the generation time of the first parameter and the generation time of the second parameter is less than a preset difference, and the first parameter and the second parameter are the same parameter; merging the first parameter and the second parameter to obtain a first merged parameter; and determining the first merged parameter and other parameters included in the fault parameters other than the first parameter and the second parameter as the single-parameter fault entities respectively.

[0140] In an exemplary embodiment, the second determining module 804 can determine a single-parameter fault entity based on the fault parameters in the following manner: determining a multi-dimensional feature vector for each parameter included in the fault parameters; determining the similarity between any two sub-feature vectors included in the multi-dimensional feature vectors; determining a target similarity greater than a preset similarity included in the similarity; determining a third parameter and a fourth parameter corresponding to the target similarity; merging the third parameter and the fourth parameter to obtain a second merged parameter; and determining the second merged parameter and other parameters included in the fault parameters other than the third parameter and the fourth parameter as the single-parameter fault entity.

[0141] In an exemplary embodiment, the fourth determining module 808 can determine the correlation between the parameters included in the fault parameters in the following manner: determining the support count of an initial itemset based on the fault parameters, wherein the initial itemset includes N types of itemsets, the i-th type of itemset includes i parameters, i=1,2,3...N, where N is the number of fault parameters, and each type of itemset is different; determining frequent itemsets within each third preset time window based on the support count, wherein the frequent itemsets are the itemsets included in the initial itemset; determining global frequent itemsets based on the frequent itemsets; and determining that the correlation exists between the parameters included in the global frequent itemsets.

[0142] In an exemplary embodiment, the fourth determining module 808 may determine a global frequent itemset based on the frequent itemset in the following manner: determining the weight of each target itemset included in the frequent itemset in each third preset time window; determining the target weight of the target itemset, wherein the target weight is the sum of the weights of the target itemset in each third preset time window; and determining the target itemset whose sum is greater than a preset weight threshold as the global frequent itemset.

[0143] In an exemplary embodiment, the apparatus may be configured to: after constructing a domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the association relationship, receive a second telemetry parameter; if the domain map includes the second telemetry parameter, determine a first single-parameter fault entity corresponding to the second telemetry parameter; determine the fault type of the first single-parameter fault entity; if the fault type is an existing type, store the second telemetry parameter; if the fault type is a first-time occurrence type, update the single-parameter fault entity in the domain map based on the second telemetry parameter to obtain an updated single-parameter fault entity; if the updated single-parameter fault entity exists in the domain map, update the system-level fault entity based on the updated single-parameter fault entity to obtain an updated domain map.

[0144] In an exemplary embodiment, the apparatus can also be used to: after constructing a domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the association relationship, determine a second single-parameter fault entity corresponding to the target system-level fault entity included in the domain map; and determine that the target system-level fault entity has a fault if all of the second single-parameter fault entities have faults.

[0145] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0146] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein the program executes the steps in any of the above method embodiments when it is run.

[0147] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0148] According to another aspect of the embodiments of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is configured to perform the steps of any of the method embodiments described above via the computer program. In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0149] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0150] According to another aspect of the embodiments of this application, a computer program product is also provided, comprising a computer program / instructions containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication section 909, and / or installed from a removable medium 911. When the computer program is executed by a central processing unit 901, it performs various functions provided in the embodiments of this application. The sequence numbers of the embodiments of this application above are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0151] Figure 9 A schematic block diagram of a computer system architecture for implementing embodiments of the present application is shown. Figure 9 As shown, the computer system 900 includes a Central Processing Unit (CPU) 901, which performs various appropriate actions and processes based on programs stored in ROM 902 or loaded into RAM 903 from storage section 908. Random Access Memory 903 also stores various programs and data required for system operation. The CPU 901, ROM 902, and RAM 903 are interconnected via bus 904. Input / Output (I / O) interface 905 is also connected to bus 904.

[0152] The following components are connected to I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), and speakers, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card, such as a local area network card or modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to I / O interface 905 as needed. Removable media 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 910 as needed so that computer programs read from them can be installed into storage section 908 as needed.

[0153] Specifically, according to embodiments of this application, the processes described in the various method flowcharts can be implemented as computer software programs. For example, embodiments of this application 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 909, and / or installed from removable medium 911. When the computer program is executed by central processing unit 901, it performs various functions defined in the system of this application.

[0154] It should be noted that, Figure 9 The computer system 900 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0155] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0156] The above are merely preferred embodiments of this application and are not intended to limit 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 principles of this application should be included within the protection scope of this application.

Claims

1. A method for determining a satellite domain map, characterized in that, include: Determine the fault parameters included in the telemetry parameters of the target satellite; A single-parameter fault entity is determined based on the fault parameters, wherein the single-parameter fault entity is a fault entity caused by a single parameter fault. System-level fault entities are determined based on the single-parameter fault entities, wherein the system-level fault entities are fault entities caused by faults based on multiple parameters. Determine the correlation between the parameters included in the fault parameters; The domain map of the target satellite is constructed based on the single-parameter fault entity, the system-level fault entity, and the correlation.

2. The method according to claim 1, characterized in that, Based on the single-parameter fault entity, system-level fault entities are determined, including: The entity pair included in the single-parameter fault entity is determined, wherein the entity pair includes a first single-parameter fault entity and a second single-parameter fault entity; Determine the association strength between the first single-parameter fault entity and the second single-parameter fault entity; If the correlation strength is greater than or equal to the second preset threshold, the first single-parameter fault entity and the second single-parameter fault entity are identified as system-level fault entities.

3. The method according to claim 2, characterized in that, Determining the association strength between the first single-parameter fault entity and the second single-parameter fault entity includes: Determine the co-occurrence strength between the first single-parameter fault entity and the second single-parameter fault entity; Determine the conditional probability of the second single-parameter fault entity failing within a first preset time window when the first single-parameter fault entity fails. The target time difference is determined based on the first time when the first single-parameter faulty entity fails and the second time when the second single-parameter faulty entity fails. The order parameter is determined based on the first single-parameter fault entity and the second single-parameter fault entity. The association strength is determined based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter.

4. The method according to claim 3, characterized in that, Determining the association strength based on the co-occurrence strength, the conditional probability, the target time difference, and the order parameter includes: Determine the first product of the co-occurrence intensity and the first parameter; Determine the second product of the conditional probability and the second parameter; Determine the third product of the target time difference and the third parameter; Determine the fourth product of the order parameter and the fourth parameter; The sum of the first product, the second product, the third product, and the fourth product is determined as the correlation strength.

5. The method according to claim 3, characterized in that, Determining the co-occurrence strength between the first single-parameter fault entity and the second single-parameter fault entity includes: The number of times the first single-parameter fault entity and the second single-parameter fault entity simultaneously fail within the second preset time window is determined, as well as the number of times the first single-parameter fault entity fails and the number of times the second single-parameter fault entity fails within the first preset period. Determine the first sum of the second number and the third number; Determine the first difference between the first sum and the first number; The ratio of the first occurrence to the first difference is determined as the co-occurrence intensity.

6. The method according to claim 3, characterized in that, Determining the target time difference based on the first time the first single-parameter faulty entity fails and the second time the second single-parameter faulty entity fails includes: A single-parameter fault entity pair is determined based on the first time and the second time, wherein the single-parameter fault entity pair includes a first sub-entity and a second sub-entity, the first sub-entity and the second sub-entity are entities that fail consecutively, and the time when the first sub-entity fails is earlier than the time when the second sub-entity fails. For each single-parameter fault entity pair, the following operations are performed to determine the initial time difference corresponding to the single-parameter fault entity pair: determining the first start time of the first sub-entity fault included in the first time and determining the second start time of the second sub-entity fault included in the second time, and determining the difference between the second start time and the first start time as the initial time difference; The average of the initial time differences of all the single-parameter fault entity pairs is determined as the target time difference.

7. The method according to claim 3, characterized in that, Determining the order parameter based on the first single-parameter fault entity and the second single-parameter fault entity includes: The failure of the first single-parameter fault entity occurs earlier than the fourth failure of the second single-parameter fault entity within the second preset time window. Determine the second sum of the fifth number of times the first single-parameter fault entity fails and the sixth number of times the second single-parameter fault entity fails; The second ratio of the fourth number to the second sum is determined as the order parameter.

8. The method according to claim 1, characterized in that, Determining a single-parameter fault entity based on the fault parameters includes: The fault parameters include a first parameter and a second parameter, wherein the difference between the generation time of the first parameter and the generation time of the second parameter is less than a preset difference, and the first parameter and the second parameter are the same parameter. The first parameter and the second parameter are combined to obtain the first combined parameter; The first merging parameter and the other parameters included in the fault parameters, excluding the first and second parameters, are respectively determined as the single-parameter fault entities.

9. The method according to claim 1, characterized in that, Determining a single-parameter fault entity based on the fault parameters includes: Determine the multi-dimensional feature vector of each parameter included in the fault parameters; Determine the similarity between any two sub-feature vectors included in the multi-dimensional feature vector; Determine the target similarity that is greater than the preset similarity in the similarity; Determine the third and fourth parameters corresponding to the target similarity; By combining the third parameter and the fourth parameter, we obtain the second combined parameter: The second merging parameter and the other parameters included in the fault parameters, excluding the third and fourth parameters, are respectively determined as the single-parameter fault entities.

10. The method according to claim 1, characterized in that, Determining the correlation between the parameters included in the fault parameters includes: The support count of the initial itemset is determined based on the fault parameters. The initial itemset includes N types of itemsets, and the i-th type of itemset includes i parameters, i=1,2,3...N, where N is the number of fault parameters, and each type of itemset is different. Based on the support count, a frequent itemset is determined within each third preset time window, wherein the frequent itemset is the itemset included in the initial itemset; Determine the global frequent itemset based on the aforementioned frequent itemset; The correlation relationship is determined to exist among the parameters included in the global frequent itemset.

11. The method according to claim 10, characterized in that, Determining a global frequent itemset based on the aforementioned frequent itemsets includes: Determine the weight of each target itemset included in the frequent itemset in each of the third preset time windows; Determine the target weight of the target itemset, wherein the target weight is the sum of the weights of the target itemset in each of the third preset time windows; The target itemset whose sum is greater than a preset weight threshold is determined as the global frequent itemset.

12. The method according to claim 1, characterized in that, After constructing the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the correlation, the method further includes: Receive the second telemetry parameters; If the second telemetry parameter is included in the domain map, the first single-parameter fault entity corresponding to the second telemetry parameter is determined; Determine the fault type of the first single-parameter fault entity; If the fault type is an existing type, store the second telemetry parameter; When the fault type is the first occurrence type, the single-parameter fault entity in the domain map is updated based on the second telemetry parameter to obtain the updated single-parameter fault entity; If the updated single-parameter fault entity exists in the domain graph, the system-level fault entity is updated based on the updated single-parameter fault entity to obtain the updated domain graph.

13. The method according to claim 1, characterized in that, After constructing the domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the correlation, the method further includes: Determine the second single-parameter fault entity corresponding to the target system-level fault entity included in the domain map; If all of the second single-parameter fault entities are faulty, it is determined that the target system-level fault entity is faulty.

14. A device for determining a satellite domain map, characterized in that, include: The first determining module is used to determine the fault parameters included in the telemetry parameters of the target satellite; The second determining module is used to determine a single-parameter fault entity based on the fault parameters, wherein the single-parameter fault entity is a fault entity caused by a single parameter fault. The third determining module is used to determine the system-level fault entity based on the single-parameter fault entity, wherein the system-level fault entity is a fault entity caused by multiple parameter faults. The fourth determining module determines the correlation between the parameters included in the fault parameters; A construction module is used to construct a domain map of the target satellite based on the single-parameter fault entity, the system-level fault entity, and the correlation relationship.

15. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 13.

17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 13.

Images