Satellite anomaly detection control method based on state twinning

Abstract

According to the satellite anomaly detection control method based on state twinning provided by the invention, the twinning system which is the same as the satellite is established, and the actual operation state of the satellite is compared with the state of the twinning node in real time, so that the performance of the satellite is monitored more comprehensively, integration of multivariate data sources is realized, and the satellite anomaly detection control method based on state twinning is realized. And the accuracy and the reliability of the data are effectively improved. And secondly, the state difference between the actual satellite and the twin node is monitored in real time, once the state difference exceeds a threshold value, the satellite system can immediately start an anomaly detection mechanism and rapidly identify and analyze the anomaly reason, the processing efficiency of potential risks is enhanced, and the risk in satellite operation is reduced. Besides, the instructions are copied to the twin system in real time before and after launching, it is ensured that execution of each instruction can obtain accurate simulation feedback, the accuracy of instruction execution is improved, reliable data support is provided for error recording, and follow-up analysis is more accurate.

Description

Technical Field

[0001] The present invention belongs to the technical field of satellite anomaly detection, and in particular relates to a satellite anomaly detection control method based on state twins. Background Art

[0002] Satellite anomaly detection is a key technology in aerospace work. By monitoring satellites in real time, early detection and early warning of anomalies can be achieved to ensure that possible faults can be discovered and handled as early as possible, thereby reducing the losses caused by anomalies. With the substantial increase in the number of satellites, the increased complexity of satellite systems, and the diversity of satellite missions, satellite anomaly detection has become one of the most critical tasks in aerospace engineering. Therefore, it is particularly important to develop more advanced detection technologies. Spacecraft performance and health monitoring are essential to ensure the reliability, safety, and availability of spacecraft and to ensure their normal operation. Since it is challenging to repair space systems on orbit, health monitoring and early anomaly detection methods are essential to the success of space missions.

[0003] Most current satellite status monitoring relies on a single data source, such as telemetry data from satellites or a small amount of external sensor data. These data usually include basic information such as the satellite's attitude, orbital parameters, battery power, temperature, etc. However, this single data source approach has many limitations. For example, satellites may encounter tiny meteoroid collisions, solar storms, or other space environment changes in orbit, and these events may not be directly reflected in the telemetry data. Therefore, this monitoring method often has difficulty detecting some potential, gradually accumulated abnormal conditions. For example, after some satellites have been exposed to a strong radiation environment for a long time, their electronic components may gradually degrade. Relying solely on telemetry data may not be able to detect this degradation in time, resulting in the failure of the satellite to provide early warning before a serious failure occurs.

[0004] In the prior art, the identification of satellite abnormal conditions usually relies on multiple consecutive data collection and analysis, which means that the identification of abnormal conditions has a certain lag. For example, if the attitude control system of a satellite fails, the existing method may require several comparisons and analyses of telemetry data to confirm the abnormality. However, during this lag period, the attitude control system may have caused the satellite to deviate from the predetermined orbit or orientation, leading to more serious consequences. Such lag is particularly fatal in critical missions. For example, in high-resolution earth observation missions, a slight deviation in attitude control may cause inaccurate observation data, directly affecting the success rate of the mission. In addition, the delayed identification of abnormal conditions may result in the problem being discovered only when it has accumulated to an irreversible level, making subsequent remedial measures more complicated and expensive.

[0005] When satellites execute commands sent from the ground, they often face a problem: these commands lack sufficient simulation and verification before being sent, which may cause unpredictable deviations during the execution of the commands. For example, ground control personnel may send an orbit adjustment command, but due to the lack of sufficient simulation testing, the command may produce greater thrust than expected during actual execution, causing the satellite to deviate from the target orbit. Worse, the lack of effective simulation verification means that such deviations are difficult to record or analyze, further increasing the risk accumulation of subsequent commands. To give a specific example, when a series of attitude adjustment commands are sent from the ground to ensure that the satellite is aligned with the ground target, if these commands have not been rigorously simulated and verified, the satellite's attitude adjustment range may exceed the expected range, affecting the success rate of the mission. This inaccuracy in error recording not only weakens the grasp of the satellite's status, but may also increase the risk of similar operations in the future. Summary of the invention

[0006] In order to solve the above problems existing in the prior art, the present invention provides a satellite anomaly detection control method based on state twin. The technical problem to be solved by the present invention is achieved through the following technical solutions: A satellite anomaly detection and control method based on state twins includes: S100, before the satellite is launched, a twin node consistent with each satellite is established and formed into a twin system; a first instruction identical to that of the corresponding satellite is copied to the twin node so that each twin node executes the first instruction and obtains a first state of the twin system; S200, after the satellite is launched, synchronously sending the second instruction sent by the ground station to each satellite to the twin system, so that the twin system completes the operation according to the corresponding second instruction, and obtains the second state of the twin system; S300, comparing the first state of the twin system with the state of the satellite executing the first instruction before launch, and comparing the second state of the twin system with the state of the satellite executing the second instruction after launch, to obtain an evaluation result of each satellite; S400, if the evaluation result of the satellite is abnormal, triggering the automatic data backtracking and troubleshooting mechanism to trace the cause of the satellite abnormality and formulate the best emergency measures according to the cause of the satellite abnormality; S500, feeding back the evaluation results of each satellite and the optimal emergency measures to the twin system, and regularly maintaining and upgrading the twin system so that the twin system always maintains a consistent state with the satellite.

[0007] Beneficial effects: The present invention provides a satellite anomaly detection and control method based on state twin, which is aimed at the problem that the existing technology relies on a single data source and cannot fully reflect the real state of the satellite. The present invention establishes a twin system identical to the satellite, and the actual operating state of the satellite can be compared with the state of the twin node in real time, so as to more comprehensively monitor the performance of the satellite, realize the integration of multiple data sources, and effectively improve the accuracy and reliability of the data. Secondly, in view of the problem that the traditional method has a delayed response to the recognition of abnormal states and is difficult to deal with potential risks in a timely manner, the present invention monitors the state difference between the actual satellite and the twin node in real time. Once the state difference exceeds the threshold, the satellite system will immediately start the anomaly detection mechanism to quickly identify and analyze the cause of the anomaly. This rapid response capability greatly enhances the efficiency of handling potential risks and reduces the risks in satellite operation. In addition, in view of the problem that the existing technology lacks an effective instruction replication and simulation verification mechanism, resulting in inaccurate error records, the present invention has an effective instruction replication and simulation verification mechanism. The present invention ensures that the execution of each instruction can obtain accurate simulation feedback by copying instructions to the twin system in real time before and after launch. This mechanism not only improves the accuracy of instruction execution, but also provides reliable data support for error records, making subsequent analysis more accurate.

[0008] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a flow chart of a satellite anomaly detection control method based on state twinning provided by the present invention; Figure 2 It is a process schematic diagram of a satellite anomaly detection control method based on state twin provided by the present invention. DETAILED DESCRIPTION

[0010] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0011] like Figure 1 As shown, the present invention provides a satellite anomaly detection control method based on state twins, comprising: S100, before the satellite is launched, a twin node consistent with each satellite is established and formed into a twin system; a first instruction identical to that of the corresponding satellite is copied to the twin node so that each twin node executes the first instruction and obtains a first state of the twin system; As an optional implementation manner of the present invention, S100 includes: S110, before the satellite is launched, a twin node is established that is consistent with each satellite and equipped with the same hardware and software configuration as the corresponding satellite; S120, all twin nodes form a twin system; S130, before the satellite is launched, obtaining static instructions executed by each satellite; S140, for each twin node, copy the corresponding static instruction for the twin node to obtain the first instruction of the twin node; S150, perform simulation test on each twin node to make it execute the corresponding first instruction, and obtain the first state of each twin node in the twin system.

[0012] This implementation method constructs a simulation node that is exactly the same as the satellite, which is a twin node. All twin nodes form a twin system, and the twin system is a real-time state dictionary. The twin system should have the same hardware and software configuration as the satellite to ensure that the behavior of the satellite can be accurately simulated. At the same time, it is necessary to complete the command replication and ground simulation test between the twin system and the satellite, copy all commands sent to the satellite, execute these commands in the twin system, and save the state corresponding to the command. In this way, the impact of the command on the satellite performance can be grasped to obtain the state of the satellite, and the state data of the twin node and the satellite can be recorded. These data will be used as a reference for subsequent detection and correction.

[0013] Among them, the instructions executed by the satellite include position, attitude control and power adjustment.

[0014] S200, after the satellite is launched, synchronously sending the second instruction sent by the ground station to each satellite to the twin system, so that the twin system completes the operation according to the corresponding second instruction, and obtains the second state of the twin system; As an optional implementation manner of the present invention, S200 includes: S210, after the satellite is launched, obtaining a dynamic instruction sent by the ground station to the satellite; S220, for each twin node, copy the corresponding dynamic instruction for the twin node to obtain a second instruction of the twin node; S230, performing simulation test on each twin node so that it executes the corresponding second instruction to obtain the second state of each twin node in the twin system.

[0015] After the satellite is launched, all commands sent by the ground station to the satellite are also sent to the twin system at the same time. This ensures that the twin system can reflect the working status of the satellite in real time and achieve the purpose of real-time command convergence. While the commands are synchronized to the satellite and the twin system, synchronization status monitoring is also required. After receiving the command, the twin system monitors the second state achieved by each twin node executing the command in real time, and compares it with the actual state of the satellite to determine whether the error is within the normal threshold range.

[0016] S300, comparing the first state of the twin system with the state of the satellite executing the first instruction before launch, and comparing the second state of the twin system with the state of the satellite executing the second instruction after launch, to obtain an evaluation result of each satellite; As an optional implementation manner of the present invention, S300 includes: S310, for each satellite, comparing the state of the satellite executing the first instruction before launch with the first state of the corresponding twin node, and if the difference between the two exceeds a predetermined first threshold, marking the evaluation result of the satellite as abnormal; S320: For each satellite, compare the state of the satellite executing the second instruction after launch with the second state of the corresponding twin node. If the difference between the two exceeds a predetermined second threshold, mark the evaluation result of the satellite as abnormal.

[0017] The first threshold and the second threshold of the present invention can be set in advance according to actual conditions.

[0018] This implementation method regularly compares the actual state of the satellite before launch with the state of the twin node, and also compares it in time after sending the command, that is, compares the actual state of the satellite after launch with the state of the twin node, focusing on the differences in key parameters such as temperature, voltage, attitude, etc. If it is found that the state difference between the actual satellite and the twin node exceeds the set threshold, the system will mark the satellite's evaluation result as abnormal.

[0019] S400, if the evaluation result of the satellite is abnormal, triggering the automatic data backtracking and troubleshooting mechanism to trace the cause of the satellite abnormality and formulate emergency measures according to the cause of the satellite abnormality; As an optional implementation manner of the present invention, S400 includes: S410: If the evaluation result of the satellite is abnormal, data backtracking is performed to obtain historical data of the abnormal satellite and the corresponding twin node; S420, analyzing historical data of the abnormal satellite and the corresponding twin node to track the abnormality that occurred during the execution of the instruction and to trace the root cause of the satellite abnormality; S430, formulate the best emergency response measures based on the root cause of the satellite anomaly.

[0020] If an abnormality occurs, the current state of the satellite is considered abnormal based on the historical command impact and error records, and the satellite is checked for abnormalities. The check requires the immediate sending of telemetry commands to return all the current states of the satellite, so as to evaluate the satellite, discover and sort out the commands that cause state errors, and make decisions and judgments by ground personnel. By analyzing the historical data of satellite status and twin nodes, the abnormalities that occur during the execution of commands can be tracked and the root causes of the abnormalities can be found.

[0021] S500, feeding back the evaluation results of each satellite and the optimal emergency measures to the twin system, and regularly maintaining and upgrading the twin system so that the twin system always maintains a consistent state with the satellite.

[0022] See also Figure 2 The satellite anomaly detection control method based on state twin provided by the present invention is divided into four processes, namely, establishing a twin system, data synchronization, state comparison and correction, anomaly detection and response. S100 involves the process of establishing a twin system, S200 involves the process of establishing a twin system and data synchronization, involving copying uplink instructions and downlink data to the twin node, S300 is the state comparison process, and S400 is the anomaly detection and response mechanism.

[0023] As an optional implementation manner of the present invention, S410 includes: S411, if the satellite evaluation result is abnormal, define a length of time window; S412, from the current moment arrive The telemetry data of the satellite, the simulation data of the corresponding twin node, the static instructions and dynamic instructions sent to the satellite in this time period are extracted as historical data.

[0024] Define a length of The time window, starting from the current time Back to .generally Adjustments are made based on the satellite’s mission type and system response time. Then the satellite telemetry data for that time period is extracted. , simulation data of twin nodes , satellite command records , including key operations such as position, attitude control, and power adjustment.

[0025] The static instructions and dynamic instructions sent to the satellite include instructions for position, attitude control and power adjustment.

[0026] As an optional implementation manner of the present invention, S420 includes: S421, analyzing the execution of the static instructions and the dynamic instructions, and checking whether any instructions exceed the expected effect or show inconsistent behavior with the satellite in the twin node simulation; S422, if any command exceeds the expected effect or exhibits inconsistent behavior with the satellite in the twin node simulation, the command is sent to the ground station; S423, for the abnormal satellite, comparing the telemetry data of the abnormal satellite with the simulation data of the corresponding twin node to determine the abnormal time period causing the abnormality; This implementation analyzes the execution of these instructions to see if any instructions have exceeded the expected effect or have exhibited inconsistent behavior with the satellite in the twin node simulation. Then, by comparing the data of the satellite and the twin node, the time period that caused the state difference is found. Further narrow the time period of the anomaly.

[0027] S424, extracting log data of the abnormal satellite and the corresponding twin node within the abnormal time period, and dividing the log data into operation records of different modules; The operation records include: attitude control, orbit adjustment and communication status; the fault causes include a certain instruction repeatedly causing attitude control deviations and similar attitude deviations occurring multiple times when working at a predetermined temperature.

[0028] S425, analyzing the log data using a decision tree algorithm to automatically generate possible fault causes that may cause the abnormality; S426, sorting all possible fault causes that may cause the abnormality according to the probability, and obtaining a fault cause sorting result.

[0029] This implementation method extracts the system logs of satellites and twin nodes during abnormal periods. and Including instruction execution logs, system status logs, etc., the log data is then classified into operation records of different modules, such as attitude control, orbit adjustment, communication status, etc. Then the decision tree algorithm is used to analyze the data in the log and automatically generate possible paths that lead to abnormalities. For example, if a certain instruction repeatedly causes deviations in attitude control, the decision tree can help determine whether it is caused by system errors or environmental factors. The abnormal data is then matched with historical failure modes to identify anomalies similar to known failure modes. For example, similar attitude deviations occur multiple times when the system is working at a predetermined temperature, which may indicate that temperature is the main cause of the failure. Based on the results of the decision tree analysis, possible causes of failure are output and sorted by probability. These results are provided to the emergency response mechanism.

[0030] By comparing the log data of the actual satellite and the twin node, possible causes of abnormalities can be identified, such as command errors, system failures, or external environmental influences. In troubleshooting, decision trees can be used to analyze the associations between different system parameters and match them according to known failure modes to identify the most likely cause of the failure.

[0031] As an optional implementation manner of the present invention, S430 includes: According to the fault cause ranking results, countermeasures are formulated, and the effects of the countermeasures are optimized using a dynamic programming algorithm to determine the best emergency measures.

[0032] For the identified abnormal causes, corresponding countermeasures are taken immediately, such as adjusting instructions, restarting the system, or conducting further fault inspections to prevent potential harm. In order to find the best emergency measures, dynamic programming algorithms can be used to optimize the effects of different emergency measures. In addition, twin nodes are used to simulate these emergency measures to ensure their effectiveness in actual execution.

[0033] As an optional implementation manner of the present invention, S500 includes: S510, feeding back the evaluation result of each satellite and the optimal emergency measure to the corresponding satellite and the twin system, so that the corresponding satellite uses the optimal emergency measure and the corresponding historical data as training data to update and train its own integrated machine learning module, and the satellite itself performs anomaly detection and autonomous decision-making; S520: Based on the evaluation results and the optimal emergency measures, regularly maintain and upgrade the twin system so that the twin system always maintains a consistent state with the satellite.

[0034] The satellite of the present invention is integrated with a machine learning module, which uses historical data and ground decisions as training data for training, so that the satellite can automatically alarm the next time an abnormality occurs, and adopt the decision made by the satellite system when the accuracy rate is higher than a certain threshold. And the evaluation results are fed back to the twin system to continuously optimize the simulation model and monitoring mechanism. This can improve the accuracy and real-time performance of future command execution. At the same time, during the operation of the twin system, the twin system is regularly maintained and upgraded to ensure that it is always in a state consistent with the satellite to adapt to the ever-changing mission requirements.

[0035] The present invention provides a satellite anomaly detection and control method based on state twin, which is aimed at the problem that the existing technology relies on a single data source and cannot fully reflect the real state of the satellite. The present invention establishes a twin system identical to the satellite, and the actual operating state of the satellite can be compared with the state of the twin node in real time, so as to more comprehensively monitor the performance of the satellite, realize the integration of multiple data sources, and effectively improve the accuracy and reliability of the data. Secondly, in view of the problem that the traditional method has a delayed response to the recognition of abnormal states and is difficult to deal with potential risks in a timely manner, the present invention monitors the state difference between the actual satellite and the twin node in real time. Once the state difference exceeds the threshold, the satellite system will immediately start the anomaly detection mechanism to quickly identify and analyze the cause of the anomaly. This rapid response capability greatly enhances the efficiency of handling potential risks and reduces the risks in satellite operation. In addition, in view of the problem that the existing technology lacks an effective instruction replication and simulation verification mechanism, resulting in inaccurate error records, the present invention has an effective instruction replication and simulation verification mechanism. The present invention ensures that the execution of each instruction can obtain accurate simulation feedback by copying instructions to the twin system in real time before and after launch. This mechanism not only improves the accuracy of instruction execution, but also provides reliable data support for error records, making subsequent analysis more accurate.

[0036] It is worth noting that the terms "first" and "second" in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0037] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other variations of the disclosed embodiments by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality of components or steps.

[0038] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A satellite anomaly detection and control method based on state twins, characterized in that: include: S100, before the satellite is launched, establish twin nodes consistent with each satellite and form them into a twin system; Copying a first instruction identical to that of a corresponding satellite for the twin node so that each twin node executes the first instruction to obtain a first state of the twin system; S200, after the satellite is launched, synchronously sending the second instruction sent by the ground station to each satellite to the twin system, so that the twin system completes the operation according to the corresponding second instruction, and obtains the second state of the twin system; S300, comparing the first state of the twin system with the state of the satellite executing the first instruction before launch, and comparing the second state of the twin system with the state of the satellite executing the second instruction after launch, to obtain an evaluation result of each satellite; S400, if the evaluation result of the satellite is abnormal, triggering the automatic data backtracking and troubleshooting mechanism to trace the cause of the satellite abnormality and formulate the best emergency measures according to the cause of the satellite abnormality; S500, feeding back the evaluation results of each satellite and the optimal emergency measures to the twin system, and regularly maintaining and upgrading the twin system so that the twin system always maintains a consistent state with the satellite.

2. The satellite anomaly detection and control method based on state twin according to claim 1 is characterized in that: S100 includes: S110, before the satellite is launched, a twin node is established that is consistent with each satellite and equipped with the same hardware and software configuration as the corresponding satellite; S120, all twin nodes form a twin system; S130, before the satellite is launched, obtaining static instructions executed by each satellite; S140, for each twin node, copy the corresponding static instruction for the twin node to obtain the first instruction of the twin node; S150, perform simulation test on each twin node to make it execute the corresponding first instruction, and obtain the first state of each twin node in the twin system.

3. The satellite anomaly detection and control method based on state twin according to claim 1 is characterized in that: S200 includes: S210, after the satellite is launched, obtaining a dynamic instruction sent by the ground station to the satellite; S220, for each twin node, copy the corresponding dynamic instruction for the twin node to obtain a second instruction of the twin node; S230, performing simulation test on each twin node so that it executes the corresponding second instruction to obtain the second state of each twin node in the twin system.

4. The satellite anomaly detection and control method based on state twin according to claim 1 is characterized in that: S300 includes: S310, for each satellite, comparing the state of the satellite executing the first instruction before launch with the first state of the corresponding twin node, and if the difference between the two exceeds a predetermined first threshold, marking the evaluation result of the satellite as abnormal; S320: For each satellite, the state of the satellite executing the second instruction after launch is compared with the second state of the corresponding twin node. If the difference between the two exceeds a predetermined second threshold, the evaluation result of the satellite is marked as abnormal.

5. The satellite anomaly detection and control method based on state twin according to claim 1 is characterized in that: S400 includes: S410: If the evaluation result of the satellite is abnormal, data backtracking is performed to obtain historical data of the abnormal satellite and the corresponding twin node; S420, analyzing historical data of the abnormal satellite and the corresponding twin node to track the abnormality that occurred during the execution of the instruction and to trace the root cause of the satellite abnormality; S430, formulate the best emergency response measures based on the root cause of the satellite anomaly.

6. The satellite anomaly detection and control method based on state twin according to claim 5 is characterized in that: S410 includes: S411, if the satellite evaluation result is abnormal, define a length of time window; S412, from the current moment arrive The telemetry data of the satellite, the simulation data of the corresponding twin node, the static instructions and dynamic instructions sent to the satellite in this time period are extracted as historical data.

7. The satellite anomaly detection and control method based on state twin according to claim 6 is characterized in that: S420 includes: S421, analyzing the execution of the static instructions and the dynamic instructions, and checking whether any instructions exceed the expected effect or show inconsistent behavior with the satellite in the twin node simulation; S422, if any command exceeds the expected effect or exhibits inconsistent behavior with the satellite in the twin node simulation, the command is sent to the ground station; S423, for the abnormal satellite, comparing the telemetry data of the abnormal satellite with the simulation data of the corresponding twin node to determine the abnormal time period causing the abnormality; S424, extracting log data of the abnormal satellite and the corresponding twin node within the abnormal time period, and dividing the log data into operation records of different modules; S425, analyzing the log data using a decision tree algorithm to automatically generate possible fault causes that may cause the abnormality; S426, sorting all possible fault causes that may cause the abnormality according to the probability, and obtaining a fault cause sorting result.

8. The satellite anomaly detection and control method based on state twin according to claim 7 is characterized in that: The operation records include: attitude control, orbit adjustment and communication status; the fault causes include a certain instruction repeatedly causing attitude control deviations and similar attitude deviations occurring multiple times when working at a predetermined temperature.

9. The satellite anomaly detection and control method based on state twin according to claim 7 is characterized in that: S430 includes: According to the fault cause ranking results, countermeasures are formulated, and the effects of the countermeasures are optimized using a dynamic programming algorithm to determine the best emergency measures.

10. The satellite anomaly detection and control method based on state twin according to claim 1 is characterized in that: S500 includes: S510, feeding back the evaluation result of each satellite and the optimal emergency measure to the corresponding satellite and the twin system, so that the corresponding satellite uses the optimal emergency measure and the corresponding historical data as training data to update and train its own integrated machine learning module, and the satellite itself performs anomaly detection and autonomous decision-making; S520, based on the evaluation results and the optimal emergency measures, regularly maintain and upgrade the twin system so that the twin system always maintains a consistent state with the satellite.

Images