Testing method and system for satellite-borne intelligent health management software

Abstract

The invention provides a test method for satellite-borne intelligent health management software, which comprises the following steps of: establishing a ground test platform to connect a satellite and ground equipment, and electrifying and initializing the satellite; setting a plurality of on-orbit working conditions through a remote control instruction, collecting telemetering data to train a software algorithm model, and uploading to a satellite after training is completed; starting software health, abnormity and efficiency evaluation functions in sequence; verifying health state output of subsystem components under a normal working condition; simulating a fault through fault injection equipment, and verifying abnormal label generation and recovery; verifying the matching between the real-time efficiency value and the standard value in satellite maneuver; and integrating all test results to judge whether the software is qualified or not. The invention further provides a testing system of the satellite-borne intelligent health management software. Therefore, the accuracy and the reliability of judgment when the satellite-borne intelligent health management software is applied can be improved, and the method has universal adaptability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of satellite software testing, in particular to a test method and system for on-board intelligent health management software. BACKGROUND

[0002] The failure detection, isolation and recovery (FDIR) technology of a spacecraft runs through each mission phase of the satellite from launch to normal on-orbit operation to the end of life. With the rapid development of satellite manufacturing technology, its functional performance is becoming more and more powerful, and the satellite system is becoming more and more complex. The various subsystems on the satellite are interrelated and work cooperatively, and a failure of one subsystem may affect other subsystems, and even cause the entire satellite system to be paralyzed. Such complexity also makes the FDIR technology of the satellite change rapidly.

[0003] At present, the problem of insufficient autonomous fault diagnosis and decision-making ability of the satellite leading to fault propagation and causing serious loss has appeared. An on-board intelligent health management software based on an artificial intelligence diagnosis model is introduced to solve the problem of insufficient information fusion degree existing in the existing fault diagnosis rule system. The software learns and trains an artificial intelligence model based on historical data to realize unknown fault diagnosis outside the rule knowledge base. Therefore, in order to fully test the on-board software on the ground, a test method for the on-board intelligent health management software is urgently needed.

[0004] At present, in the field of satellite FDIR technology testing, for example, patent No. CN202110908853.5 discloses a "verification method and system for whole-satellite FDIR scheduling strategy", which proposes a method for building a hardware and software environment for ground testing of the whole-satellite FDIR. However, this method does not involve the data model training and testing method for the on-board software based on the artificial intelligence model. SUMMARY

[0005] In view of the problem in the prior art that the current satellite ground testing cannot effectively verify the function of the new on-board intelligent health management software of the satellite, the purpose of the present application is to provide a test method and system for on-board intelligent health management software, which can fully test the on-board intelligent health management software on the ground, and can be adapted to the ground testing of most on-board software.

[0006] In order to achieve the above technical effects, on the one hand, the present application provides a test method for on-board intelligent health management software, comprising the steps of:

[0007] Building a ground test platform, connecting the satellite and the ground equipment, and initializing the satellite;

[0008] Multiple on-orbit operating conditions are set for the satellite using remote control commands. Telemetry data of the entire satellite under the corresponding operating conditions are collected and transmitted to the algorithm model of the onboard intelligent health management software for training. After training is completed, the onboard intelligent health management software is uploaded to the satellite.

[0009] Enable the health assessment function of the onboard intelligent health management software to verify whether the software's assessment of the health status of each subsystem component of the satellite is accurate;

[0010] Enable the anomaly assessment function of the onboard intelligent health management software, simulate the fault state of a single unit on the satellite by setting abnormal parameter values ​​through the fault injection device, and verify the generation and recovery of anomaly tags;

[0011] Enable the performance evaluation function of the onboard intelligent health management software and verify the matching between the real-time performance value and the standard value of the performance indicator through the satellite maneuver process;

[0012] The overall test results are evaluated to determine whether the onboard intelligent health management software is qualified.

[0013] Optionally, the satellite's on-orbit operating conditions include:

[0014] Power-on / off operations and telemetry parameter checks for individual satellite equipment;

[0015] Adjustment of battery pack charging and discharging current and cutoff voltage;

[0016] Switching between channel types for satellite-to-ground uplink and downlink communication links;

[0017] Switching between operating modes of the control subsystem.

[0018] Optionally, the training of the algorithm model specifically includes:

[0019] The collected whole-satellite telemetry data is used as a training data source to train the AI ​​diagnostic model of the onboard intelligent health management software. After training, the model is uploaded to the satellite software.

[0020] Optionally, setting abnormal parameter values ​​via the fault injection device specifically involves setting abnormal telemetry parameters of the power manager via the fault injection device, including:

[0021] Set the BCDR (Battery Charge Discharge Regulator) temperature value to the first abnormal value that exceeds the normal range of 15-30℃;

[0022] Set the BCDR charge / discharge current to the second abnormal value, which exceeds the normal threshold of 7.5A.

[0023] Optionally, the generation and recovery of the verification anomaly label includes:

[0024] After the fault injection device injects the corresponding abnormal telemetry parameters, it monitors and verifies whether the BCDR abnormal tag and the battery abnormal tag output abnormality, and whether the tag returns to normal synchronously after the parameters return to normal.

[0025] Optionally, the verification of the matching between the real-time performance value and the performance index standard value through the satellite maneuver process includes:

[0026] Real-time performance values ​​of the attitude control subsystem are collected before, during, and after the satellite maneuver.

[0027] The real-time performance value was verified to be in the range of 0.83 to 1 and to match the standard value of the performance index.

[0028] Optional, also includes:

[0029] The health assessment function, the anomaly assessment function, and the performance assessment function of the onboard intelligent health management software can be independently started and stopped via remote control commands.

[0030] On the other hand, the present invention also provides a spaceborne intelligent health management software testing system for implementing the method described above, comprising:

[0031] The operational condition simulation module is used to generate a database of telemetry values ​​for the satellite's on-orbit operational conditions.

[0032] The fault injection device connects to the satellite interface and is used to simulate satellite fault conditions without affecting the health of the onboard unit.

[0033] The data acquisition module is used to record telemetry data from the entire satellite and transmit it to the AI ​​training model;

[0034] The evaluation and verification module is used to perform functional tests, namely health evaluation, anomaly evaluation, and performance evaluation, in sequence.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] (1) This invention fills the gap in training methods for artificial intelligence models of satellite software. It designs and sets up various satellite on-orbit conditions, records the corresponding whole satellite telemetry data, and uses the data source to train the algorithm model, which greatly improves the accuracy and reliability of the judgment when the satellite intelligent health management software is applied.

[0037] (2) The present invention uses ground fault injection test equipment, which can simulate fault conditions without affecting the satellite itself, fully verify the complete workflow of the onboard intelligent health management software, and has universal adaptability. Most onboard software can achieve the effect on the ground through the test software of the present invention, and has broad application prospects in the future. Attached Figure Description

[0038] Figure 1 A flowchart illustrating the steps of a testing method for spaceborne intelligent health management software provided in an embodiment of the present invention;

[0039] Figure 2 This is a diagram showing the device connection relationship of a testing method for spaceborne intelligent health management software provided in an embodiment of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0041] It should be noted that references to "an embodiment," "embodiment," "example embodiment," etc., in this specification refer to the described embodiment including specific features, structures, or characteristics, but not every embodiment must include these specific features, structures, or characteristics. Furthermore, such expressions do not refer to the same embodiment. Moreover, when describing specific features, structures, or characteristics in conjunction with embodiments, whether or not explicitly described, it is indicated that incorporating such features, structures, or characteristics into other embodiments is within the knowledge of those skilled in the art.

[0042] Furthermore, certain terms are used in the specification and subsequent claims to refer to specific components or parts. Those skilled in the art will understand that manufacturers may use different names or terms to refer to the same component or part. This specification and subsequent claims do not distinguish components or parts by differences in name, but rather by differences in function. The terms "comprising" and "including" used throughout the specification and subsequent claims are open-ended and should be interpreted as "including but not limited to." Additionally, the term "connection" here includes any direct and indirect electrical connection means. Indirect electrical connection means include connections made through other means.

[0043] Before describing the embodiments of this application in detail, the technical concept of this application is briefly described first: The core of this invention is to build a ground test platform, connect ground equipment with the satellite, and use remote control commands to set various on-orbit operating conditions such as satellite single-unit health checks and battery charging and discharging, record telemetry data to train the AI ​​model of the onboard intelligent health management software; at the same time, set up fault injection equipment to simulate faults, test the software's health, anomaly and performance evaluation functions, and comprehensively evaluate the test results to verify the software's functions and reliability.

[0044] This invention aims to fill the gap in training methods for artificial intelligence models in spaceborne software. It designs and sets up various satellite on-orbit operating conditions, records corresponding whole-satellite telemetry data, and uses this data source to train the algorithm model, thereby greatly improving the accuracy and reliability of judgments when spaceborne intelligent health management software is applied.

[0045] The specific principles of the testing method for the spaceborne intelligent health management software of this application will be described below with reference to specific embodiments.

[0046] Figure 1 This invention illustrates a testing method for spaceborne intelligent health management software provided in an embodiment of the present invention, comprising the following steps:

[0047] S101: Set up a ground test platform, connect the satellite to ground equipment, and power on and initialize the satellite; after connecting the ground equipment to the satellite in each location, turn on the ground equipment and set its status, then power on the satellite via remote control commands. Figure 2 As shown, the ground equipment includes power supply and telemetry / remote control ground equipment, control system ground equipment, on-board payload system ground equipment, and central control equipment, etc.; the equipment on the satellite includes an integrated business management unit, power controller, fusion heterogeneous management unit, payload unit equipment, control unit equipment, etc.

[0048] S102: Set various on-orbit operating conditions for the satellite through remote control commands, collect the whole satellite telemetry data of the corresponding operating conditions and transmit it to the algorithm model of the on-board intelligent health management software for training. After the training is completed, upload the on-board intelligent health management software to the satellite.

[0049] In an optional implementation, the satellite's on-orbit operating conditions include, but are not limited to: power-on / off operations of individual satellite equipment and telemetry parameter checks; adjustment of the charging / discharging current and cutoff voltage of the battery pack; switching of the channel type of the satellite-to-ground uplink and downlink communication links; and switching of the operating mode of the control subsystem.

[0050] The training of the algorithm model specifically includes: using the collected whole-satellite telemetry data as the training data source to train the AI ​​diagnostic model of the onboard intelligent health management software, and then uploading the model to the satellite software after training is completed.

[0051] Using remote control commands to set various operational states for the satellite, specifically including:

[0052] (1) Health check of each unit on the satellite: Use remote control commands to power on and off the satellite-borne unit, and check the relevant remote control commands and telemetry parameters of the unit.

[0053] (2) Battery pack charging / discharging conditions: The satellite battery pack status is simulated using remote control commands, including battery charging and discharging, adjusting different charging / discharging currents and cutoff voltages, and recording telemetry parameters during the condition setting period.

[0054] (3) Operating conditions of different types of communication links between satellite and ground: Use remote control commands to set up satellite and ground uplink links, such as wired channel, spread spectrum channel, spread jump channel, etc.

[0055] (4) Different operating conditions of the control subsystem: Use ground remote control commands to set the operating conditions of the control subsystem, such as normal mode, position holding mode, etc.

[0056] The telemetry data of the entire satellite under different operating conditions is recorded and transmitted to the algorithm model of the onboard intelligent health management software for training. After the training is completed, the software is uploaded to the satellite for model training, and then the software is uploaded to the satellite.

[0057] This embodiment also simulates the on-board unit required for your test by setting up a fault injection device.

[0058] Furthermore, the test status settings of each subsystem on the satellite were configured remotely. Specific tests are as follows:

[0059] S103: Enable the health assessment function of the onboard intelligent health management software to verify the accuracy of the software's assessment of the health status of various subsystem components of the satellite.

[0060] In practice, the onboard intelligent health management software, as well as the health assessment function switches for the attitude control subsystem, power supply and distribution subsystem, and telemetry and control subsystem, and the health assessment function switches for each component, are activated via remote control commands. The software begins executing the health assessment function, and the health assessment results for each component are checked over a period of time. For example, the health assessment results for momentum wheels 1-6 (output labels D1-D6), BCDR1-4 (output labels B1-B4), battery (output label X1), and fixed amplifier 1-4 (output labels G1-G4) are all "normal," meeting actual expectations. Afterward, the ground sends remote control commands to the satellite to disable the software's health assessment function.

[0061] S104: Enable the anomaly assessment function of the onboard intelligent health management software, simulate the fault state of a single unit on the satellite by setting abnormal parameter values ​​through the fault injection device, and verify the generation and recovery of anomaly tags.

[0062] In an optional implementation, setting abnormal parameter values ​​via the fault injection device specifically involves setting abnormal telemetry parameters of the power manager via the fault injection device, including:

[0063] Set the BCDR temperature value to the first abnormal value, which is outside the normal range of 15-30℃; set the BCDR charging and discharging current to the second abnormal value, which is outside the normal threshold of 7.5A.

[0064] Furthermore, the generation and recovery of the verification anomaly tag includes: after the fault injection device injects the corresponding abnormal telemetry parameters, monitoring and verifying whether the BCDR anomaly tag and the battery anomaly tag output as abnormal, and whether the tag synchronously recovers to normal after the parameters return to normal.

[0065] In practice, the anomaly assessment function of the onboard intelligent health management software is activated via remote control. Ground fault injection equipment is configured, and simulations of the required onboard units, such as the power manager, are initiated. Certain telemetry parameters are set to abnormal values. For example, the temperature telemetry values ​​for BCDR1-4 (Tem1-Tem4) are set to 49℃, while the normal range is 15-30℃; the charging / discharging current for BCDR1-4 is set to 49A, which normally does not exceed 7.5A. Changes in the telemetry labels for the corresponding components are recorded, such as BCDR anomaly labels (B5-B10) and battery anomaly labels (X2-X6). When anomalies are set, the label is "abnormal." After the anomaly value returns to normal, the corresponding component parameter anomaly labels also return to "normal," consistent with expectations. Then, a remote control command is sent to the satellite to disable the software anomaly assessment function.

[0066] S105: Enable the performance evaluation function of the onboard intelligent health management software, and verify the matching between the real-time performance value and the standard value of the performance indicator through the satellite maneuver process;

[0067] In an optional implementation, verifying the matching between the real-time performance value and the performance indicator standard value through satellite maneuvering includes:

[0068] Before, during, and after the satellite maneuver, the real-time performance values ​​of the attitude control subsystem were collected; the real-time performance values ​​were verified to be in the range of 0.83 to 1 and to match the performance index standard values.

[0069] In practice, the performance evaluation function of the onboard intelligent health management software is activated via remote control commands. The remote control commands are then used to set the maneuvering status of the ground control equipment and the onboard control system. Before and after the maneuver, the real-time performance evaluation values ​​and standard performance index values ​​are viewed for a period of time. Throughout the entire maneuvering process, from the start to the end, the real-time performance evaluation values ​​and standard performance index values ​​are viewed again. During non-maneuvering periods, the real-time performance value of the corresponding telemetry word ACS (ACS1) and the standard values ​​of performance indices 1-8 (ACS2-ACS9) are recorded. The results range from a minimum of 0.83 to a maximum of 1, which meets the expected performance. Afterwards, remote control commands are sent to the satellite to deactivate the software performance evaluation function.

[0070] In this embodiment, the health assessment function, anomaly assessment function, and performance assessment function of the onboard intelligent health management software are independently started and stopped via remote control commands.

[0071] S106: A comprehensive evaluation of all test results is conducted to determine whether the onboard intelligent health management software is qualified. By comprehensively evaluating the results of each stage of the entire testing process, if all tests are completed, it indicates that the onboard intelligent health management software has undergone comprehensive functional testing and the results are normal; otherwise, it is determined to be abnormal.

[0072] This embodiment improves the accuracy of unknown fault diagnosis through multi-condition telemetry data training; it uses fault injection equipment to safely simulate anomalies, avoiding satellite damage and achieving non-destructive full-process verification; and it ensures on-orbit reliability by covering the core functions of the onboard software through health / anomaly / performance testing.

[0073] Another embodiment of the present invention provides a spaceborne intelligent health management software testing system for implementing the method described in the above embodiments, comprising a working condition simulation module, a fault injection device, a data acquisition module, and an evaluation and verification module, wherein:

[0074] The operational condition simulation module is used to generate a database of telemetry values ​​for the satellite's on-orbit operational conditions; the fault injection device is connected to the satellite interface and is used to simulate satellite fault conditions without affecting the health of individual onboard units; the data acquisition module is used to record the entire satellite's telemetry data and transmit it to the AI ​​training model; the evaluation and verification module is used to perform functional tests, including health evaluation, anomaly evaluation, and performance evaluation, in sequence.

[0075] In an optional implementation, the functions of the operating condition simulation module and the evaluation and verification module can be implemented by the main control computer. For example, both the fault injection device and the data acquisition module are connected to the satellite interface, and the data acquisition module is communicatively connected to the main control computer. The main control computer generates a telemetry database of the satellite's on-orbit operating conditions based on the data obtained from the data acquisition module, and sequentially performs functional tests such as health assessment, anomaly assessment, and performance assessment.

[0076] In summary, this invention establishes a testing platform for the novel onboard intelligent health management software during the satellite ground testing phase. It designs and sets various on-orbit operating conditions for the satellite, providing sufficient data samples for the software's model training. Furthermore, it offers a complete testing approach to test various software functions, fully verifying the software's practical effectiveness in satellite systems. This invention improves the accuracy of unknown fault diagnosis through multi-condition telemetry data training; it safely simulates anomalies using fault injection equipment, avoiding satellite damage and achieving non-destructive full-process verification; and it ensures on-orbit reliability by covering the core functions of the onboard software through health / anomaly / performance testing.

[0077] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of the present invention is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0078] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A testing method for spaceborne intelligent health management software, characterized in that, Including the following steps: A ground test platform was set up, the satellite was connected to the ground equipment, and the satellite was powered on and initialized. Multiple on-orbit operating conditions are set for the satellite using remote control commands. Telemetry data of the entire satellite under the corresponding operating conditions are collected and transmitted to the algorithm model of the onboard intelligent health management software for training. After training is completed, the onboard intelligent health management software is uploaded to the satellite. Enable the health assessment function of the onboard intelligent health management software to verify whether the software's assessment of the health status of each subsystem component of the satellite is accurate; Enable the anomaly assessment function of the onboard intelligent health management software, simulate the fault state of a single unit on the satellite by setting abnormal parameter values ​​through the fault injection device, and verify the generation and recovery of anomaly tags; Enable the performance evaluation function of the onboard intelligent health management software and verify the matching between the real-time performance value and the standard value of the performance indicator through the satellite maneuver process; The overall test results are evaluated to determine whether the onboard intelligent health management software is qualified.

2. The testing method for the spaceborne intelligent health management software according to claim 1, characterized in that, The satellite's on-orbit operating conditions include: Power-on / off operations and telemetry parameter checks for individual satellite equipment; Adjustment of battery pack charging and discharging current and cutoff voltage; Switching between channel types for satellite-to-ground uplink and downlink communication links; Switching between operating modes of the control subsystem.

3. The testing method for the spaceborne intelligent health management software according to claim 1, characterized in that, The training of the algorithm model specifically includes: The collected whole-satellite telemetry data is used as a training data source to train the AI ​​diagnostic model of the onboard intelligent health management software. After training, the model is uploaded to the satellite software.

4. The testing method for the spaceborne intelligent health management software according to claim 1, characterized in that, Setting abnormal parameter values ​​via fault injection specifically involves setting abnormal telemetry parameters for the power manager via fault injection, including: Set the BCDR temperature value to the first abnormal value that exceeds the normal range of 15-30℃. Set the BCDR charge / discharge current to the second abnormal value, which exceeds the normal threshold of 7.5A.

5. The testing method for the spaceborne intelligent health management software according to claim 4, characterized in that, The generation and recovery of the verification anomaly label includes: After the fault injection device injects the corresponding abnormal telemetry parameters, it monitors and verifies whether the BCDR abnormal tag and the battery abnormal tag output abnormality, and whether the tag returns to normal synchronously after the parameters return to normal.

6. The testing method for the spaceborne intelligent health management software according to claim 1, characterized in that, The verification of the match between real-time performance values ​​and performance indicator standard values ​​through satellite maneuvering processes includes: Real-time performance values ​​of the attitude control subsystem are collected before, during, and after the satellite maneuver. The real-time performance value was verified to be in the range of 0.83 to 1 and to match the standard value of the performance index.

7. The testing method for the spaceborne intelligent health management software according to claim 1, characterized in that, Also includes: The health assessment function, the anomaly assessment function, and the performance assessment function of the onboard intelligent health management software can be independently started and stopped via remote control commands.

8. A spaceborne intelligent health management software testing system for implementing the method as described in any one of claims 1 to 7, characterized in that, include: The operational condition simulation module is used to generate a database of telemetry values ​​for the satellite's on-orbit operational conditions. The fault injection device connects to the satellite interface and is used to simulate satellite fault conditions without affecting the health of the onboard unit. The data acquisition module is used to record telemetry data from the entire satellite and transmit it to the AI ​​training model; The evaluation and verification module is used to perform functional tests, namely health evaluation, anomaly evaluation, and performance evaluation, in sequence.

Images