Abnormality sign determination device and abnormality sign determination method

Abstract

To provide an abnormality sign determination device capable of detecting abnormality signs in an area wider than an artificial satellite itself.SOLUTION: Second sign determination means 74 holds a second model representing an output pattern and output timing of a satellite control system log when a satellite belonging to the satellite system, a terrestrial antenna, and a satellite control system are normal, and applies the newly obtained time series data of the satellite control system log to the second model, thereby determining whether or not there are any signs of abnormality for a part including the satellite, the terrestrial antenna, and the satellite control system.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present invention relates to an abnormality sign determination device, an abnormality sign determination method, and an abnormality sign determination program that determine whether or not there is a sign of abnormality in an object. [Background technology]

[0002] Patent Document 1 describes an operation control device that can analyze the effects of and predict the occurrence of failures in a satellite based on telemetry data. Patent Document 1 also describes a neurochip that incorporates a neural network into a chip. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 3-244009 Summary of the Invention [Problem to be solved by the invention]

[0004] When detecting signs of anomaly based on telemetry data output by a satellite, the detection range of the signs of anomaly is limited to the satellite itself. However, the satellite is operated as part of a satellite system together with a ground antenna, a satellite control system, and the like. Therefore, it is preferable that the detection range of the signs of anomaly is wider than the satellite itself. In other words, it is preferable that signs of anomaly can be detected within a wider range than the satellite itself.

[0005] Therefore, an object of the present invention is to provide an abnormality sign determination device, an abnormality sign determination method, and an abnormality sign determination program that can detect abnormality signs in an area wider than the artificial satellite itself. [Means for solving the problem]

[0006] The abnormality sign determination device according to the present invention is characterized in that it comprises a first sign determination means that holds as a first model a set of calculation formulas for individual observation values ​​obtained by expressing each observation value included in telemetry data output by a satellite belonging to a satellite system as a function having other observation values ​​as variables, calculates a calculated value of each observation value by substituting the observation value included in the newly obtained telemetry data for the variable of the calculation formula for each observation value, calculates a difference between an actual measurement value and a calculated value of each observation value included in the newly obtained telemetry data, and determines that there is a sign of abnormality in the satellite if the absolute value of the difference between the actual measurement value and the calculated value is equal to or greater than a threshold, and determines that there is a sign of abnormality in a portion related to the observation value corresponding to the variable of the calculation formula for the calculated value; and a second sign determination means that holds a second model that represents the output pattern and output timing of the log of the satellite control system when a satellite belonging to the satellite system, a terrestrial antenna, and a satellite control system are normal, and determines whether or not there is a sign of abnormality by applying the newly obtained time series data of the log of the satellite control system to the second model.

[0007] The method for determining signs of anomaly according to the present invention is characterized in that the computer retains as a first model a set of calculation formulas for individual observation values ​​obtained by expressing each observation value included in telemetry data output by a satellite belonging to a satellite system as a function with other observation values ​​as variables, calculates a calculated value for each observation value by substituting the observation value included in the newly obtained telemetry data for the variable of the calculation formula for each observation value, calculates the difference between the actual measurement value and the calculated value for each observation value included in the newly obtained telemetry data, and determines that there is a sign of anomaly in the satellite if the absolute value of the difference between the actual measurement value and the calculated value is equal to or greater than a threshold value, determines that there is a sign of anomaly in the part related to the observation value corresponding to the variable of the calculation formula for the calculated value, retains a second model representing the output pattern and output timing of the log of the satellite control system when the satellite belonging to the satellite system, the terrestrial antenna, and the satellite control system are normal, and determines whether there is a sign of anomaly for the part including the satellite, the terrestrial antenna, and the satellite control system by applying the newly obtained time series data of the log of the satellite control system to the second model.

[0008] The abnormality sign determination program according to the present invention stores in a computer, as a first model, a set of calculation formulas for each observed value obtained by expressing each observed value included in telemetry data output by a satellite belonging to a satellite system as a function having other observed values ​​as variables, calculates a calculated value of each observed value by substituting the observed value included in newly obtained telemetry data for the variable of the calculation formula for each observed value, calculates a difference between an actual measured value and a calculated value of the observed value included in the newly obtained telemetry data for each observed value, and when an absolute value of the difference between the actual measured value and the calculated value becomes equal to or greater than a threshold value, determines that the satellite is abnormal. A first precursor determination process is executed to determine that there is a precursor to an abnormality in the star and to determine that there is a precursor to an abnormality in a part related to an observation value that corresponds to a variable in a calculation formula for a calculated value, and a second precursor determination process is executed to determine whether there is a precursor to an abnormality in a part including the satellite, the terrestrial antenna, and the satellite control system by retaining a second model that represents the output pattern and output timing of the satellite control system log when the satellite belonging to the satellite system, the terrestrial antenna, and the satellite control system are normal, and applying the newly obtained time series data of the satellite control system log to the second model. Effect of the Invention

[0009] According to the present invention, it is possible to detect signs of abnormality in a wider area than the artificial satellite itself. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a block diagram showing an example of a satellite system in which a sign of an abnormality is detected. [Diagram 2] 1 is a block diagram showing an example of the configuration of an abnormality sign determination device according to an embodiment of the present invention; [Diagram 3] 11 is a flowchart showing an example of a process flow when a first sign determination unit determines the presence or absence of a sign of abnormality. [Figure 4] 13 is a flowchart showing an example of a process flow when a second sign determination unit determines the presence or absence of a sign of abnormality. [Diagram 5]1 is a schematic block diagram showing an example of the configuration of a computer related to an abnormality sign determination device of the present invention; [Figure 6] 1 is a block diagram showing an overview of an abnormality sign determination device according to the present invention; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0012] First, a satellite system in which a sign of an abnormality is detected will be described. Fig. 1 is a block diagram showing an example of a satellite system in which a sign of an abnormality is detected. The satellite system 50 includes an artificial satellite 51, a terrestrial antenna 52, and a satellite control system 53. The satellite system 50 may further include other elements.

[0013] The artificial satellite 51 transmits telemetry data to a ground antenna 52, and the ground antenna 52 sends the received telemetry data to a satellite control system 53. The satellite control system 53 holds the time-series data of the telemetry data as a database.

[0014] The telemetry data includes observation values ​​indicating the status of devices mounted on the artificial satellite 51. In the following, an example will be described in which the telemetry data includes at least an observation value of the satellite power, an observation value of the heater temperature, and an observation value of the internal pressure of the satellite.

[0015] The time-series data of the telemetry data output by the satellite 51 when the satellite 51 is normal is used as learning data when a first model, which will be described later, is generated by machine learning.

[0016] The satellite control system 53 outputs a log. The log includes, for example, an operation record of data transmission and reception, the start and end of processing, etc. However, the information included in the log is not limited to these. The satellite control system 53 holds the time series data of the output log as a database.

[0017] The time series data of the log output by the satellite control system 53 when the artificial satellite 51, the ground antenna 52, and the satellite control system 53 are normal is used as learning data when generating the second model described below by machine learning.

[0018] The abnormality sign determination device of this embodiment (not shown in Figure 1) is connected to a satellite control system 53, and can acquire newly obtained telemetry data from a telemetry data database, and can acquire newly obtained time series data of logs of the satellite control system 53 from a log database.

[0019] 2 is a block diagram showing an example of the configuration of an abnormality sign determination device according to an embodiment of the present invention. The abnormality sign determination device 10 includes a display control unit 11, a display device 12, a first sign determination unit 13, and a second sign determination unit 14.

[0020] The display control unit 11 displays the determination result by the first sign determination unit 13 and the determination result by the second sign determination unit 14 on the display device 12.

[0021] The first sign determining unit 13 holds a first model.

[0022] The first model is a set of calculation formulas for each observation value obtained by expressing each observation value included in the telemetry data output by the artificial satellite 51 as a function with other observation values ​​included in the telemetry data as variables.

[0023] For example, the telemetry data includes an observation value of satellite power, an observation value of heater temperature, an observation value of internal spacecraft pressure, etc. For example, the observation value of satellite power is expressed by a function that uses the observation value of heater temperature as a variable. The observation value of heater temperature is expressed by a function that uses the observation value of internal spacecraft pressure as a variable. Similarly, the observation value of internal spacecraft pressure is expressed by a function that uses other observation values ​​as variables. These functions are, respectively, a formula for calculating the observation value of satellite power, a formula for calculating the observation value of heater temperature, and a formula for calculating the observation value of internal spacecraft pressure. A collection of such calculation formulas is a first model.

[0024] The first model is generated in advance by machine learning and is stored in the first sign determination unit 13. As described above, in this machine learning, time-series data of the telemetry data output by the satellite 51 when the satellite 51 is normal is used as learning data.

[0025] When new telemetry data of the artificial satellite 51 is added to the telemetry data database in the satellite control system 53, the first sign determining unit 13 acquires the new telemetry data.

[0026] Then, the first sign determination unit 13 calculates the calculation value of each observation value by substituting the observation value included in the new telemetry data into the variable of the calculation formula for each observation value. It is assumed that the calculation formula in the above example is included in the first model. In this case, the first sign determination unit 13 calculates the calculation value of the observation value of the satellite power by substituting the observation value of the heater temperature into the variable of the calculation formula for the observation value of the satellite power. The first sign determination unit 13 similarly calculates the calculation value of the observation value of the heater temperature and the calculation value of the observation value of the spacecraft internal pressure.

[0027] In addition, when comparing with calculated values, the observation values ​​themselves contained in the telemetry data are referred to as actual values.

[0028] The first sign determining unit 13 calculates, for each observed value, the difference between the calculated value and the actual measured value included in the new telemetry data.

[0029] Then, when the absolute value of the difference between the calculated value and the actual measured value is equal to or greater than the threshold value, the first sign determination unit 13 determines that there is a sign of an abnormality in the artificial satellite 51. In this case, the first sign determination unit 13 further determines that there is a sign of an abnormality in a portion related to an observed value corresponding to a variable of the calculation formula of the calculated value. Note that the above threshold value may be different for each type of observed value.

[0030] For example, suppose that the absolute value of the difference between the calculated value and the measured value of the satellite power is equal to or greater than the threshold value. In this case, the first sign determination unit 13 determines that there is a sign of an abnormality in the satellite 51. In the above example, the variable in the formula for the calculated value of the satellite power is the observed value of the heater temperature. Therefore, the first sign determination unit 13 determines that there is a sign of an abnormality in the part related to the observed value of the heater temperature (i.e., the heater in the satellite 51). In this case, the display control unit 11 displays information such as "There is a sign of an abnormality in the satellite 51" or "There is a sign of an abnormality in the heater in the satellite 51" on the display device 12.

[0031] Also, for example, suppose that the absolute value of the difference between the calculated value and the actual measured value of the heater temperature is equal to or greater than the threshold value. In this case, the first sign determination unit 13 determines that there is a sign of an abnormality in the artificial satellite 51. In the above example, the variable in the formula for the calculated value of the heater temperature is the observed value of the pressure inside the aircraft. Therefore, the first sign determination unit 13 determines that there is a sign of an abnormality in a portion related to the observed value of the pressure inside the aircraft. In this case, the display control unit 11 displays information such as "There is a sign of an abnormality in the artificial satellite 51" or "There is a sign of an abnormality in a portion related to the observed value of the pressure inside the aircraft" on the display device 12.

[0032] Furthermore, if the absolute value of the difference between the calculated value and the actual measured value is less than the threshold for any of the observed values, the first sign determination unit 13 determines that there is no sign of an abnormality in the artificial satellite 51. In this case, the display control unit 11 displays on the display device 12 the information that "There is no sign of an abnormality in the artificial satellite 51."

[0033] The second sign determining unit 14 holds a second model.

[0034] The second model is a model that represents the output pattern and output timing of the log of the satellite control system 53 when the artificial satellite 51, the ground antenna 52, and the satellite control system 53 are normal.

[0035] The second model is generated in advance by machine learning and is stored in the second sign determination unit 14. As described above, in this machine learning, time-series data of the log output by the satellite control system 53 when the artificial satellite 51, the ground antenna 52, and the satellite control system 53 are normal is used as learning data.

[0036] When time series data of a new log output by the satellite control system 53 is added to the log database in the satellite control system 53, the second sign determining unit 14 acquires the time series data of the new log.

[0037] Then, the second precursor judgment unit 14 applies the time series data of the new log output by the satellite control system 53 to the second model, thereby determining whether or not there are precursors of an abnormality for the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53.

[0038] For example, the second sign determination unit 14 compares the output pattern and output timing of a log identified by the time series data of a new log with the output pattern and output timing of a log represented by the second model, and determines whether there is a discrepancy between the two.

[0039] If there is a discrepancy between the output pattern and output timing of the log specified by the time series data of the new log and the output pattern and output timing of the log represented by the second model, the second sign determination unit 14 determines that there is a sign of an abnormality somewhere in the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53. In this case, the display control unit 11 displays on the display device 12 the information that "There is a sign of an abnormality somewhere in the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53."

[0040] The second sign determination unit 14 cannot determine the specific contents of the sign, such as "there is a sign of a communication abnormality between the artificial satellite 51 and the terrestrial antenna 52." However, the second sign determination unit 14 can determine that there is a sign of an abnormality somewhere in the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53.

[0041] If there is no deviation between the output pattern and output timing of the log specified by the time series data of the new log and the output pattern and output timing of the log represented by the second model, the second sign determination unit 14 determines that there are no signs of abnormality in the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53. In this case, the display control unit 11 displays on the display device 12 the information that "There are no signs of abnormality in the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53."

[0042] The display control unit 11, the first sign determination unit 13, and the second sign determination unit 14 are realized, for example, by a CPU (Central Processing Unit) of a computer that operates according to an abnormality sign determination program. In this case, the CPU reads the abnormality sign determination program from a program recording medium such as a program storage device of the computer, and operates as the display control unit 11, the first sign determination unit 13, and the second sign determination unit 14 according to the abnormality sign determination program.

[0043] Next, the process will be described. Fig. 3 is a flow chart showing an example of the process when the first sign determination unit 13 determines the presence or absence of a sign of abnormality. Detailed description of items that have already been described will be omitted.

[0044] First, the first sign determining unit 13 acquires new telemetry data (step S1).

[0045] Then, the first sign determining unit 13 calculates the calculation value of each of the observation values ​​by substituting the observation value included in the new telemetry data into the variable of the calculation formula of each of the observation values ​​(step S2).

[0046] Next, the first sign determining unit 13 calculates the difference between the calculated value and the actual measured value for each observed value (step S3).

[0047] Next, the first sign determining unit 13 determines whether or not there is an observed value whose absolute value of the difference is equal to or greater than a threshold value (step S4).

[0048] If there is an observed value whose absolute value of the difference is equal to or greater than the threshold value (Yes in step S4), the first sign determination unit 13 determines that there is a sign of an abnormality in the artificial satellite 51, and further determines which part of the artificial satellite 51 has the sign of an abnormality (step S5). After step S5, the process proceeds to step S7.

[0049] If there is no observed value whose absolute value of the difference is equal to or greater than the threshold value (No in step S4), the first sign determiner 13 determines that there is no sign of an abnormality in the artificial satellite 51 (step S6). After step S6, the process proceeds to step S7.

[0050] In step S7, the display control unit 11 displays the determination result of step S5 or step S6 on the display device 12. More specifically, when the process moves from step S5 to step S7, the display control unit 11 displays the determination result of step S5 on the display device 12. In addition, when the process moves from step S6 to step S7, the display control unit 11 displays the determination result of step S6 on the display device 12.

[0051] 4 is a flowchart showing an example of the process flow when the second sign determination unit 14 determines the presence or absence of a sign of abnormality. Detailed description of items that have already been described will be omitted.

[0052] First, the second sign determining unit 14 acquires time-series data of a new log of the satellite control system 53 (step S11).

[0053] Then, the second precursor judgment unit 14 applies the time series data of the new log to the second model to judge whether there are any precursors of an abnormality for the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53 (step S12).

[0054] Then, the display control unit 11 displays the result of the determination in step S12 on the display device 12 (step S13).

[0055] In this embodiment, the first sign determination unit 13 determines whether there is a sign of an abnormality in the artificial satellite 51, and if there is a sign of an abnormality in the artificial satellite 51, determines which part of the artificial satellite has the sign of an abnormality. The second sign determination unit 14 cannot determine the specific content of the sign, such as "there is a sign of an abnormality in communication between the artificial satellite 51 and the terrestrial antenna 52." However, the second sign determination unit 14 can determine that there is a sign of an abnormality somewhere in the part including the artificial satellite 51, the terrestrial antenna 52, and the satellite control system 53. Since the abnormality sign determination device 10 includes such a second sign determination unit 14, it can detect an abnormality sign in a wider range than the artificial satellite 51 itself.

[0056] Furthermore, in this embodiment, the first sign determination unit 13 and the second sign determination unit 14 determine the presence or absence of an abnormality sign, so the determination result of the presence or absence of an abnormality sign does not depend on the knowledge or experience of the user of the abnormality sign determination device 10. For example, even if a newcomer who is not familiar with the operation of the satellite system 50 uses the abnormality sign determination device 10, an appropriate determination result can be obtained.

[0057] 5 is a schematic block diagram showing an example of the configuration of a computer relating to the abnormality sign determination device 10 of the present invention. The computer 2000 includes, for example, a CPU 2001, a main memory device 2002, an auxiliary memory device 2003, an interface 2004, and a display device 2005.

[0058] An abnormality sign determination device 10 according to an embodiment of the present invention is realized, for example, by a computer 2000. The operation of the abnormality sign determination device 10 is stored in the form of a program (abnormality sign determination program) in an auxiliary storage device 2003. A CPU 2001 reads out the program from the auxiliary storage device 2003, loads the program into a main storage device 2002, and executes the processing described in the above embodiment according to the program.

[0059] The auxiliary storage device 2003 is an example of a non-transient tangible medium. Other examples of non-transient tangible media include a magnetic disk, a magneto-optical disk, a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), a semiconductor memory, and the like, which are connected via an interface 2004.

[0060] Next, an overview of the present invention will be described. Fig. 6 is a block diagram showing an overview of an abnormality sign determination device of the present invention. The abnormality sign determination device includes a first sign determination means 73 and a second sign determination means 74.

[0061] The first sign determination means 73 (e.g., first sign determination unit 13) holds as a first model a set of calculation formulas for individual observation values ​​obtained by expressing each observation value included in telemetry data output by a satellite (e.g., satellite 51) belonging to a satellite system (e.g., satellite system 50) as a function with other observation values ​​as variables, calculates a calculated value of each observation value by substituting the observation value included in the newly obtained telemetry data for the variable of the calculation formula for each observation value, calculates the difference between the actual measurement value and the calculated value of the observation value included in the newly obtained telemetry data for each observation value, and determines that there is a sign of abnormality in the satellite if the absolute value of the difference between the actual measurement value and the calculated value is equal to or greater than a threshold value, and determines that there is a sign of abnormality in a part related to the observation value corresponding to the variable of the calculation formula for the calculated value.

[0062] The second sign determination means 74 (e.g., the second sign determination unit 14) holds a second model representing the output pattern and output timing of the satellite control system log when the satellite belonging to the satellite system, the terrestrial antenna (e.g., the terrestrial antenna 52), and the satellite control system (e.g., the satellite control system 53) are normal, and applies the newly obtained time series data of the satellite control system log to the second model to determine whether there are any signs of abnormality for the part including the satellite, the terrestrial antenna, and the satellite control system.

[0063] Such a configuration makes it possible to detect signs of abnormalities over a wider area than the satellite itself.

[0064] The second model may be a model obtained by machine learning using time-series data of logs output by the satellite control system when the satellite, the ground antenna, and the satellite control system are normal as learning data.

[0065] The first model may be a model obtained by machine learning using time-series data of telemetry data output by the satellite when the satellite is normal as learning data.

[0066] The telemetry data output by the satellite may include an observation of the satellite power, an observation of the heater temperature, and an observation of the internal pressure of the satellite.

[0067] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. [Industrial Applicability]

[0068] The present invention is suitably applied to an abnormality sign determination device that determines whether or not there is a sign of abnormality in an object. [Explanation of symbols]

[0069] 10 Abnormal sign determination device 11 Display control unit 12 Display device 13 First Sign Judgment Section 14 Second Predictor

Claims

1. a first sign determination means for retaining as a first model a set of calculation formulas for each observation value obtained by expressing each observation value included in telemetry data output by a satellite belonging to a satellite system as a function having other observation values ​​as variables, calculating a calculated value of each observation value by substituting the observation value included in the newly obtained telemetry data for a variable of the calculation formula for the individual observation value, calculating a difference between an actual measurement value and a calculated value of the observation value included in the newly obtained telemetry data for each observation value, determining that there is a sign of an abnormality in the satellite when an absolute value of the difference between the actual measurement value and the calculated value is equal to or greater than a threshold value, and determining that there is a sign of an abnormality in a portion related to the observation value corresponding to the variable of the calculation formula for the calculated value; and a second sign determination means for storing a second model representing an output pattern and output timing of the log of the satellite control system when the artificial satellite belonging to the satellite system, the terrestrial antenna, and the satellite control system are normal, and for determining whether or not there is a sign of an abnormality for a portion including the artificial satellite, the terrestrial antenna, and the satellite control system by applying newly obtained time-series data of the log of the satellite control system to the second model. An abnormality sign determination device comprising:

2. The second model is a model obtained by machine learning using time-series data of a log output by the satellite control system when the satellite, the ground antenna, and the satellite control system are normal as learning data. The abnormality sign determination device according to claim 1 .

3. The first model is a model obtained by machine learning using time-series data of telemetry data output by the satellite when the satellite is normal as learning data. The abnormality sign determination device according to claim 1 .

4. The telemetry data output by the satellite includes an observation value of satellite power, an observation value of heater temperature, and an observation value of internal pressure of the satellite. The abnormality sign determination device according to any one of claims 1 to 3.

5. The computer a set of calculation formulas for each observation value obtained by expressing each observation value included in telemetry data output by a satellite belonging to a satellite system as a function having other observation values ​​as variables is held as a first model, a calculated value of each observation value is calculated by substituting the observation value included in the newly obtained telemetry data for the variable of the calculation formula for the individual observation value, a difference between an actual measurement value and a calculated value of the observation value included in the newly obtained telemetry data is calculated for each observation value, and when an absolute value of the difference between the actual measurement value and the calculated value is equal to or greater than a threshold, it is determined that there is a sign of an abnormality in the satellite, and that there is a sign of an abnormality in a part related to the observation value corresponding to the variable of the calculation formula for the calculated value, A second model is held that represents an output pattern and output timing of the log of the satellite control system when the artificial satellite, the terrestrial antenna, and the satellite control system belonging to the satellite system are normal, and newly obtained time-series data of the log of the satellite control system is applied to the second model, thereby determining whether or not there is a sign of an abnormality for a portion including the artificial satellite, the terrestrial antenna, and the satellite control system. A method for determining an abnormality sign, comprising:

6. The second model is a model obtained by machine learning using time-series data of a log output by the satellite control system when the satellite, the ground antenna, and the satellite control system are normal as learning data. The method for determining an abnormality sign according to claim 5 .

7. The first model is a model obtained by machine learning using time-series data of telemetry data output by the satellite when the satellite is normal as learning data. The method for determining an abnormality sign according to claim 5 .

8. The telemetry data output by the satellite includes an observation value of satellite power, an observation value of heater temperature, and an observation value of internal pressure of the satellite. The method for determining an abnormality sign according to any one of claims 5 to 7.

9. On the computer, a first sign determination process for retaining as a first model a set of calculation formulas for each observation value obtained by expressing each observation value included in telemetry data output by a satellite belonging to a satellite system as a function having other observation values ​​as variables, calculating a calculated value of each observation value by substituting the observation value included in the newly obtained telemetry data for a variable of the calculation formula for the individual observation value, calculating a difference between an actual measurement value and a calculated value of the observation value included in the newly obtained telemetry data for each observation value, determining that there is a sign of an abnormality in the satellite when an absolute value of the difference between the actual measurement value and the calculated value is equal to or greater than a threshold value, and determining that there is a sign of an abnormality in a portion related to the observation value corresponding to the variable of the calculation formula for the calculated value; and A second sign determination process that holds a second model that represents an output pattern and output timing of the log of the satellite control system when the artificial satellite belonging to the satellite system, the terrestrial antenna, and the satellite control system are normal, and applies newly obtained time-series data of the log of the satellite control system to the second model, thereby determining whether or not there is a sign of an abnormality for a part including the artificial satellite, the terrestrial antenna, and the satellite control system. An abnormality prediction judgment program for executing the above.

Images