A sub-area division method for digital twin platform of space computing system

Abstract

The present invention discloses a sub-area division method for a digital twin platform of a space computing system. Physical entities are classified according to their functions in the computing system. The capacity of physical entities and the capacity of digital twin platform servers are used as screening conditions during division to ensure that the digital twin platform can simulate the occupancy of servers when physical entities perform computing tasks. The twin sub-areas of satellites and ground stations are divided in accordance with the principles of "master satellite first, slave satellite later" and "distance from near to far" to form master satellite twin sub-areas, slave satellite twin sub-areas and ground station twin sub-areas. The method fully considers the laws of satellite operation and the capacity limitations of twin platform servers to meet the functional requirements of the digital twin platform of the space computing system.

Description

Technical Field

[0001] The present invention relates to the field of space computing, and in particular to a sub-area division method for a digital twin platform of a space computing system. Background Art

[0002] The space computing system is an open technology facility designed and built to address the current situation of severe shortage of public computing resources in space and the urgent needs of future advanced space activities. It is composed of thousands of low-cost satellites interconnected by intersatellite laser communications to form a space computing network, providing a universal, affordable and reliable space computing infrastructure for spacecraft and mission constellations. The large-scale system is accompanied by a complex network structure, which makes its daily management and update and upgrade process extremely complicated and time-consuming. In addition, the high stability requirements for network operations, high failure costs, high testing costs and difficulty in network adjustment make the implementation and maintenance of new technologies challenging. These factors together bring a series of challenges to the planning and demonstration, construction and implementation of the space computing system, as well as the subsequent maintenance and performance improvement.

[0003] The digital twin platform builds a digital model of the physical system, monitors the system status in real time and drives the dynamic update of the model to achieve more accurate simulation and prediction of system behavior. This method can perform online decision optimization and feedback adjustment, and is particularly suitable for systems or equipment that are of great value, have strict requirements on accuracy, and have complex structures. It can provide strong technical support and solutions for the design demonstration, construction deployment, and operation and maintenance optimization of space computing systems.

[0004] In order for the digital twin platform to support the optimization and scheduling tasks of the space computing system, the parallel storage and computing functions of thousands of satellites need to be synchronized in the digital twin platform. In other words, the digital twin platform should not only display the actual occupancy of satellites and ground station servers, but also perform twin design on the servers themselves in the satellite and ground station twins, so that the running process of computing tasks can be simulated on the digital twin platform. Therefore, how to divide the satellite twins and ground station twins into sub-areas and deploy a large number of satellite twin servers and ground station twin servers on a small number of digital twin platform servers has become a technical problem.

[0005] There are many technologies for the sub-area partitioning of distributed computing systems. These technologies regard the servers that generate data as vertices, the network connections between servers as edges, simplify the computing system network into a graph network form, and use the graph partitioning method to partition it. Currently, they can handle systems with millions of vertices. However, the satellites of the space computing system orbit the earth along different satellite orbits. The connections between satellites and the ground are always changing, and the graph network changes anytime and anywhere. At the same time, the traditional graph partitioning method only considers data calculation, and does not consider the capacity of the twin platform server, and cannot support the system scheduling tasks of the twin platform. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention proposes a sub-area division method for a digital twin platform of a space computing system, which fully considers the laws of satellite operation and the capacity limitations of the twin platform server, and meets the functional requirements of the digital twin platform of a space computing system.

[0007] The objective of the present invention is achieved through the following technical solutions:

[0008] A sub-area division method for a digital twin platform of a space computing system, the method comprising the following steps:

[0009] Step 1: According to the actual design of the space computing system, the ground station and space computing satellite of the space computing system are classified, with the ground station as a separate category and the master satellite and slave satellite of the space satellite each classified into one category; through the twin modeling technology, the corresponding ground station twin, master satellite twin and slave satellite twin are constructed in the digital twin platform;

[0010] Step 2: Sort the ground station twin, the master satellite twin, the slave satellite twin, and the server of the digital twin platform by capacity; the capacity includes three parameters: the number of cores of the central processing unit, the memory capacity, and the hard disk capacity;

[0011] Step 3: Divide the primary satellite twin into sub-areas based on the satellite twin’s functions and capacity;

[0012] Step 4: Divide the average number of CPU cores of a server of a single digital twin platform by the average number of CPU cores of a single slave satellite twin to obtain the estimated number of slave satellite twins that can be accommodated by the server of a single digital twin platform; divide the average number of CPU cores of a server of a single digital twin platform by the average number of CPU cores of a single ground station twin to obtain the estimated number of ground station twins that can be accommodated by the server of a single digital twin platform;

[0013] Step 5: Divide the satellite twin sub-areas according to the satellite twin capacity, digital twin platform capacity and the orbital similarity of the actual satellite;

[0014] Step 6: Divide the ground station twin sub-areas according to the ground station twin capacity, digital twin platform capacity and the Euclidean distance of the actual ground station.

[0015] Furthermore, the sorting by capacity in step 2 specifically includes: the ground station twin, the master star twin, the slave star twin, and the server of the digital twin platform are first sorted from large to small according to the number of cores of the central processing unit; if the number of cores of the central processing unit is the same, they are sorted from large to small according to the memory capacity; if the memory capacity is the same, they are sorted from large to small according to the hard disk capacity; if the hard disk capacity is the same, they are randomly sorted.

[0016] Furthermore, the step three specifically includes the following sub-steps:

[0017] S3.1: Assign each master star twin to a server of a digital twin platform as the core of the master star twin sub-area; calculate the remaining available capacity of the server of the digital twin platform;

[0018] S3.2: Among the slave star twins that transmit information to each master star twin, select the slave star twins whose capacity is less than the remaining available capacity of the server of the digital twin platform where each master star twin is located, and the capacity requirement also meets the three capacity conditions of the number of cores of the central processor, memory capacity, and hard disk capacity;

[0019] S3.3: From the slave star twins selected in step S3.2, select the slave star twin with the largest number of cores of the central processor, and divide it into the corresponding master star twin sub-area; calculate the remaining available capacity of the server of the digital twin platform again;

[0020] S3.4: Repeat S3.2 and S3.3 until there is no slave twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the master twin is located.

[0021] Furthermore, the remaining available capacity of S3.1 refers to the designed available capacity of the digital twin platform server minus the designed capacity of the twin of the server that has been divided into the current digital twin platform.

[0022] Furthermore, the step five includes the following sub-steps:

[0023] S5.1: Divide the number of all unpartitioned slave star twins by the estimated number of slave star twins that can be accommodated by the server of a single digital twin platform obtained in step 4, and obtain the number of unpartitioned slave star twin partitions Num_fs; from all unpartitioned slave star twins, select Num_fs slave star twins with the highest number of cores of the central processor in descending order, and divide each of them into a server of a digital twin platform as the core of their respective slave star twin sub-areas;

[0024] S5.2: Calculate the remaining available capacity of the server of the digital twin platform, and take the minimum values ​​of the number of cores, memory capacity, and hard disk capacity of the central processing unit to obtain the minimum value of the remaining available capacity of the server of the digital twin platform;

[0025] S5.3: Filter by capacity, and select the first Num_fs slave star twins sorted by capacity from the slave star twins whose capacities are all less than the minimum value of the remaining available capacity;

[0026] S5.4: Calculate the orbital similarities between the first Num_fs slave star twins and the core of each slave star twin sub-region respectively, and form an orbital similarity reciprocal matrix with Num_fs rows and Num_fs columns by the reciprocals of the orbital similarities;

[0027] S5.5: Use the Hungarian algorithm to solve the similarity inverse matrix, so that each slave star twin sub-region is assigned to a slave star twin, and the sum of the similarity inverses of each slave star twin and the corresponding slave star twin sub-region core is the smallest;

[0028] S5.6: Repeat S5.2 to S5.5 until there is no slave twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the core of the slave twin sub-area is located;

[0029] S5.7: Determine whether there are any undivided slave star twins. If yes, return to S5.1; if not, end the loop; at this time, all slave star twins are divided into the slave star twin sub-area.

[0030] Furthermore, the track similarity is calculated using six track numbers.

[0031] Furthermore, the step six includes the following sub-steps:

[0032] S6.1: Divide the number of ground station twins by the estimated number of ground station twins that can be accommodated by the server of a single digital twin platform to obtain the number of ground station twin partitions Num_gs; select the ground station twins with the highest number of cores of the Num_gs central processors in descending order, and divide each of them into a server of a digital twin platform as the core of the ground station twin sub-area where they are located;

[0033] S6.2: Calculate the remaining available capacity of the servers of the digital twin platform, and take the minimum value of the remaining available capacity of the servers of the digital twin platform;

[0034] S6.3: Filter by capacity, and select the first Num_gs ground station twins sorted by capacity from the ground station twins whose capacities are all less than the minimum value of the remaining available capacity;

[0035] S6.4: Calculate the Euclidean distances between the first Num_gs ground station twins and the core of each ground station twin sub-area, respectively, to form a Euclidean distance matrix with Num_gs rows and Num_gs columns;

[0036] S6.5: Solve the Euclidean distance matrix using the Hungarian algorithm so that each ground station twin sub-area is assigned to a ground station twin, and the sum of the Euclidean distances between each ground station twin and the core of the corresponding ground station twin sub-area is minimized;

[0037] S6.6: Repeat S6.2 to S6.5 until there is no ground station twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the core of the ground station twin sub-area is located;

[0038] S6.7: Determine whether there are any undivided ground station twins. If so, return to S6.1; otherwise, the loop ends, and all ground station twins are now divided into ground station twin sub-areas.

[0039] Furthermore, the Euclidean distance in S6.4 is calculated using the actual longitude and latitude of the two ground stations.

[0040] Furthermore, the slave satellites in the space satellites are computing satellites that receive information from satellites of other systems and perform primary calculations; the master satellites in the space satellites are computing satellites that receive information from slave satellites and perform centralized calculations; the information of one slave satellite is only transmitted to one master satellite, and one master satellite can receive information from multiple slave satellites.

[0041] A sub-area division device for a digital twin platform of a space computing system includes one or more processors for implementing a sub-area division method for a digital twin platform of a space computing system.

[0042] The beneficial effects of the present invention are as follows:

[0043] The sub-area division method of the digital twin platform of the space computing system of the present invention classifies physical entities according to their functions in the computing system, and uses the capacity of physical entities and the capacity of twin platform servers as screening conditions during division, thereby ensuring that the twin platform can simulate the occupancy of the server when the physical entity performs computing tasks, and follows the principles of "master satellite first, then slave satellite" and "distance from near to far" to divide the twin sub-areas of satellites and ground stations, forming master satellite twin sub-areas, slave satellite twin sub-areas and ground station twin sub-areas, which fully considers the laws of satellite operation and the capacity limitations of twin platform servers, and meets the functional requirements of the digital twin platform of the space computing system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1This is a schematic diagram of the relationship between the sub-areas of the space computing system digital twin platform and the physical entities of the space computing system according to one embodiment of the present invention.

[0045] Figure 2 This is a workflow diagram of a sub-area division method for a digital twin platform of a space computing system according to one embodiment of the present invention.

[0046] Figure 3 This is a sub-flow chart for dividing the primary star twin body into sub-areas according to one embodiment of the present invention.

[0047] Figure 4 This is a sub-flow chart for dividing the star twin sub-areas according to one embodiment of the present invention.

[0048] Figure 5 This is a sub-flow chart of ground station twin sub-area division according to one embodiment of the present invention. DETAILED DESCRIPTION

[0049] The present invention will be described in detail below based on the accompanying drawings and preferred embodiments, and the purpose and effects of the present invention will become more clear. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] like Figure 1 As shown in the figure, the space computing system is an open scientific and technological facility designed and built for the current situation of serious shortage of public computing resources in space and the urgent needs of future advanced space activities. It is composed of thousands of low-cost satellites interconnected by intersatellite laser communication to form a space computing network, providing a universal, inclusive and reliable space computing infrastructure for spacecraft and mission constellations. The space computing system consists of ground stations and space computing satellites, among which space computing satellites are divided into master satellites and slave satellites. Slave satellites are computing satellites that receive information from satellites of other systems (such as remote sensing satellites, navigation satellites, etc.) and perform primary calculations. The master satellite is a computing satellite that receives information from slave satellites and performs centralized calculations. The information of a slave satellite is only transmitted to one master satellite, and a master satellite can receive information from multiple slave satellites. Ground stations are ground facilities used to communicate and exchange data with satellites in space. They are an important part of the satellite communication system and are responsible for tracking, telemetry, command and control of satellites. Ground stations and space computing satellites are both physical entities.

[0051] In the digital twin platform of the space computing system, physical entities are transformed into digital twins through digital twin modeling technology, which are divided into three categories: master satellite twins, slave satellite twins, and ground station twins. The relationship between digital twins and physical entities is close and dynamic. As a digital copy of a physical entity, digital twins make it possible to conduct in-depth analysis and optimization of physical entities in a virtual environment. Specifically, digital twins should meet the following requirements:

[0052] 1. Mapping relationship: Digital twins are the precise mapping of physical entities in the digital world. This means that every detail of the physical entity, including its structure, function, state, and behavior, is reflected in the digital twin.

[0053] 2. Data-driven: Digital twins rely on real-time data from physical entities. Data collected by sensors, such as temperature, pressure, vibration, etc., are transmitted to the digital twin to ensure that the digital model can reflect the current state of the physical entity.

[0054] 3. Two-way interaction: Digital twins can not only receive data from physical entities, but also send instructions to physical entities. For example, in an automated production line, digital twins may adjust the operating parameters of the machine based on simulation results.

[0055] 4. Simulation and prediction: Digital twins are used to simulate the behavior of physical entities under different conditions to predict possible problems or performance changes in future operations. This helps with maintenance, optimization, and planning.

[0056] 5. Continuous Updates: As the physical entity operates and the environment changes, the digital twin will be continuously updated to keep pace with the physical entity. This continuous update ensures that the digital twin is always an accurate representation of the physical entity.

[0057] 6. Optimization and improvement: By performing simulation and analysis on the digital twin, different operations and design changes can be tested without directly affecting the physical entity. This helps optimize the performance and efficiency of the physical entity.

[0058] 7. Independence vs. Dependence: Although the digital twin is logically independent and can be operated and analyzed independently of the physical entity, it is highly dependent on the data provided by the physical entity to maintain its accuracy and usefulness.

[0059] In order for the digital twin platform to support the optimization and scheduling tasks of the space computing system, the parallel storage and computing functions of thousands of satellites need to be synchronized in the digital twin platform. That is to say, the digital twin platform should not only display the actual occupancy of satellites and ground station servers, but also perform twin design on the servers themselves in the twins of satellites and ground stations. Only in this way can the operation process of computing tasks be simulated on the digital twin platform.

[0060] A digital twin platform consists of multiple servers. The digital twin platform is divided into sub-areas. A large number of satellite twins and ground station twins are deployed on a small number of digital twin platform servers. The set of all twins deployed on the servers of each digital twin platform is called a twin sub-area. That is to say, only one twin sub-area is deployed in the server of a digital twin platform.

[0061] According to the different types of twins contained in the twin sub-area, the twin sub-area is divided into three categories: the master star twin sub-area, the slave star twin sub-area and the ground station twin sub-area. Among them, the master star twin sub-area is a sub-area that contains the master star twin, which may also have slave star twins associated with the master star. There is only one master star in a master star twin sub-area; the slave star twin sub-area is a sub-area that only contains slave star twins; the ground station twin sub-area is a sub-area that only contains ground station twins. Satellite twins and ground station twins will not be divided into the same twin sub-area.

[0062] The digital twin platform also has other servers that perform other tasks.

[0063] like Figure 2 As shown, the sub-area division method of the digital twin platform of the space computing system according to an embodiment of the present invention includes the following steps:

[0064] Step S1: According to the actual design of the space computing system, the ground station and the space computing satellite of the space computing system are classified, the ground station is a separate category, and the master satellite and the slave satellite of the space satellite are each classified into one category; through the twin modeling technology, the corresponding ground station twin, master satellite twin and slave satellite twin are constructed in the digital twin platform;

[0065] Step S2: sort the ground station twins, master satellite twins, slave satellite twins and servers of the digital twin platform by capacity (including the number of cores of the central processing unit, memory capacity and hard disk capacity);

[0066] Among them, the ground station twins, the master satellite twins and the slave satellite twins are sorted according to the design capacity of their corresponding physical entities; the twin platform servers are sorted according to the design available capacity. The design capacity is the capacity set for the physical entity and the twin platform server when the system is designed. The capacity of the twin is the same as the capacity of the physical entity; the design available capacity refers to the capacity of the twin platform server minus the redundant capacity set during the design. The redundant capacity is the spare capacity set to ensure the stability and continuity of the system, and to ensure that the server can automatically switch to the redundant module when a module fails. The redundant capacity can be a fixed value or a percentage.

[0067] Capacity sorting specifically includes:

[0068] The twin bodies and digital twin platform servers are sorted from large to small according to the number of cores of the central processor; if the number of cores of the central processor is the same, the twin bodies and the twin platform servers are sorted from large to small according to the memory capacity; if the memory capacity is the same, the twin bodies and the twin platform servers are sorted from large to small according to the hard disk capacity; if the hard disk capacity is the same, they are sorted randomly.

[0069] Step S3: Divide the primary satellite twin into sub-areas according to the satellite twin's functions and capacities.

[0070] like Figure 3 As shown, step S3 specifically includes the following sub-steps:

[0071] S3.1: Assign each master star twin to a server of a digital twin platform as the core of each master star twin sub-area; calculate the remaining available capacity of the digital twin platform server. The remaining available capacity here refers to the design available capacity of the digital twin platform server minus the design capacity of the twin of the server that has been assigned to the current twin platform. The number of cores of the central processor, memory capacity, and hard disk capacity are subtracted accordingly.

[0072] S3.2: Among the slave star twins that transmit information to each master star twin, screen by capacity and select the slave star twins whose capacity is less than the remaining available capacity of the server of the twin platform where the master star twin is located, and the capacity requirement must meet the three capacity conditions of the number of cores of the central processing unit, memory capacity, and hard disk capacity at the same time.

[0073] S3.3: From the slave star twins selected in step S3.2, select the slave star twin with the largest number of cores of the central processor, and divide it into the corresponding master star twin sub-area; calculate the remaining available capacity of the server of the digital twin platform again;

[0074] S3.4: Repeat S3.2 and S3.3 until there is no slave twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the master twin is located.

[0075] Step S4: Divide the average number of CPU cores designed for the server of a single digital twin platform by the average number of CPU cores of a single slave satellite twin to obtain the estimated number of slave satellite twins that the server of a single digital twin platform can accommodate; divide the average number of CPU cores designed for the server of a single digital twin platform by the average number of CPU cores of a single ground station twin to obtain the estimated number of ground station twins that the server of a single digital twin platform can accommodate.

[0076] Step S5: Divide the satellite twin sub-areas according to the satellite twin capacity, the digital twin platform capacity and the orbital similarity of the actual satellite.

[0077] like Figure 4 As shown, step S5 specifically includes the following sub-steps:

[0078] S5.1: Divide the number of all unpartitioned slave star twins by the estimated number of slave star twins that can be accommodated by the server of a single digital twin platform obtained in step S4, and obtain the number of unpartitioned slave star twin partitions Num_fs; from all unpartitioned slave star twins, select Num_fs slave star twins with the highest number of central processor cores in descending order, and divide each of them into a server of a digital twin platform as the core of their respective slave star twin sub-areas;

[0079] S5.2: Calculate the remaining available capacity of the server of the digital twin platform, and take the minimum values ​​of the number of cores, memory capacity, and hard disk capacity of the central processing unit to obtain the minimum value of the remaining available capacity of the server of the digital twin platform;

[0080] S5.3: Filter by capacity, and select the first Num_fs slave star twins sorted by capacity from the slave star twins whose capacities are all less than the minimum value of the remaining available capacity;

[0081] S5.4: Calculate the orbital similarities between the first Num_fs slave star twins and the core of each slave star twin sub-region respectively, and form an orbital similarity reciprocal matrix with Num_fs rows and Num_fs columns by the reciprocals of the orbital similarities;

[0082] The orbit similarity is calculated using the six orbit elements, where the six orbit elements are six parameters describing the shape and position of the orbit, including the semi-major axis a, eccentricity e, inclination i, right ascension Ω, argument of perigee ω and true anomaly θ. If the six elements of two orbits are (a1, e1, i1, Ω1, ω1, θ1) and (a2, e2, i2, Ω2, ω2, θ2) respectively, then the similarity S between them can be defined as:

[0083] .

[0084] S5.5: Use the Hungarian algorithm to solve the similarity inverse matrix, so that each slave twin sub-area is assigned to a slave twin, and at the same time, the sum of the similarity inverses of each slave twin and the corresponding slave twin sub-area core is the smallest, thus ensuring that this allocation is the allocation with the highest similarity between the slave twin and the slave twin sub-area core. The Hungarian algorithm refers to a mature method for finding the maximum allocation problem, which is suitable for the optimization problem of allocating m objects to m areas, assigning one object to each area, and minimizing the total goal.

[0085] S5.6: Repeat S5.2 to S5.5 until there is no slave twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the core of the slave twin sub-area is located;

[0086] S5.7: Determine whether there are any undivided slave star twins. If yes, return to S5.1; if not, end the loop; at this time, all slave star twins are divided into the slave star twin sub-area.

[0087] Step S6: Divide the ground station twin sub-areas according to the ground station twin capacity, the digital twin platform capacity and the Euclidean distance of the actual ground station.

[0088] like Figure 5 As shown, step S6 specifically includes the following sub-steps:

[0089] S6.1: Divide the number of ground station twins by the estimated number of ground station twins that can be accommodated by the server of a single digital twin platform to obtain the number of ground station twin partitions Num_gs; select the ground station twins with the highest number of CPU cores in descending order, and divide each of them into a server of a digital twin platform as the core of the ground station twin sub-area where they are located;

[0090] S6.2: Calculate the remaining available capacity of the servers of the digital twin platform, and take the minimum value of the remaining available capacity of the servers of the digital twin platform;

[0091] S6.3: Filter by capacity, and select the first Num_gs ground station twins sorted by capacity from the ground station twins whose capacities are all less than the minimum value of the remaining available capacity;

[0092] S6.4: Calculate the Euclidean distances between the first Num_gs ground station twins and the core of each ground station twin sub-area respectively, and form a Euclidean distance matrix with Num_gs rows and Num_gs columns.

[0093] The Euclidean distance is calculated using the actual longitude and latitude of the two ground stations. If the longitude and latitude coordinates of the two ground stations are (lat1, lon1) and (lat2, lon2), the Euclidean distance D between them can be defined as:

[0094] .

[0095] S6.5: Use the Hungarian algorithm to solve the Euclidean distance matrix so that each ground station twin sub-area is assigned to a ground station twin, and at the same time, the sum of the Euclidean distances between each ground station twin and the corresponding ground station twin sub-area core is minimized, thus ensuring that this allocation is the allocation with the minimum Euclidean distance between the ground station twin and the ground station twin sub-area core;

[0096] S6.6: Repeat S6.2 to S6.5 until there is no ground station twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the core of the ground station twin sub-area is located;

[0097] S6.7: Determine whether there are any undivided ground station twins. If so, return to S6.1; otherwise, the loop ends, and all ground station twins are now divided into ground station twin sub-areas.

[0098] Corresponding to the embodiment of the sub-area division method of the digital twin platform of the space computing system mentioned above, the present invention also provides a sub-area division device of the digital twin platform of the space computing system, which includes one or more processors for implementing the sub-area division method of the digital twin platform of the space computing system in the above embodiment.

[0099] The sub-area division device of the digital twin platform of the space computing system of the present invention can be applied to any device with data processing capabilities, and the any device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capabilities in which it is located to read the corresponding computer program instructions in the non-volatile memory into the memory for execution. From a hardware perspective, in addition to the processor, memory, network interface, and non-volatile memory, any device with data processing capabilities in which the device in the embodiment is located may also include other hardware according to the actual functions of the device with data processing capabilities, which will not be described in detail.

[0100] Those skilled in the art can understand that the above are only preferred examples of the invention and are not intended to limit the invention. Although the invention is described in detail with reference to the above examples, those skilled in the art can still modify the technical solutions recorded in the above examples or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, etc. made within the spirit and principle of the invention shall be included in the protection scope of the invention.

Claims

1. A sub-area division method for a digital twin platform of a space computing system, characterized in that: The method comprises the following steps: Step 1: According to the actual design of the space computing system, the ground station and space computing satellite of the space computing system are classified, with the ground station as a separate category and the master satellite and slave satellite of the space satellite each classified into one category; through the twin modeling technology, the corresponding ground station twin, master satellite twin and slave satellite twin are constructed in the digital twin platform; Step 2: Sort the ground station twin, the master satellite twin, the slave satellite twin, and the server of the digital twin platform by capacity; the capacity includes three parameters: the number of cores of the central processing unit, the memory capacity, and the hard disk capacity; Step 3: Divide the twin sub-areas of the main satellite according to the functions and capacities of the satellite twin, including: S3.1: Each master star twin is divided into a server of a digital twin platform as the core of its respective master star twin sub-area; Calculate the remaining available capacity of the server of the digital twin platform; S3.2: Among the slave star twins that transmit information to each master star twin, select the slave star twins whose capacity is less than the remaining available capacity of the server of the digital twin platform where each master star twin is located, and the capacity requirement also meets the three capacity conditions of the number of cores of the central processor, memory capacity, and hard disk capacity; S3.3: From the slave star twins selected in step S3.2, select the slave star twin with the largest number of cores of the central processor, and divide it into the corresponding master star twin sub-area; calculate the remaining available capacity of the server of the digital twin platform again; S3.4: Repeat S3.2 and S3.3 until there is no slave twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the master twin is located; Step 4: Divide the average number of CPU cores of a server of a single digital twin platform by the average number of CPU cores of a single slave satellite twin to obtain the estimated number of slave satellite twins that can be accommodated by the server of a single digital twin platform; divide the average number of CPU cores of a server of a single digital twin platform by the average number of CPU cores of a single ground station twin to obtain the estimated number of ground station twins that can be accommodated by the server of a single digital twin platform; Step 5: Divide the satellite twin sub-areas according to the satellite twin capacity, digital twin platform capacity and the orbital similarity of the actual satellite; Step 6: Divide the ground station twin sub-areas according to the ground station twin capacity, digital twin platform capacity and the Euclidean distance of the actual ground station.

2. The sub-area division method of the digital twin platform of the space computing system according to claim 1 is characterized in that: The sorting by capacity in step 2 specifically includes: the ground station twin, the master star twin, the slave star twin, and the server of the digital twin platform are first sorted from large to small according to the number of cores of the central processing unit; if the number of cores of the central processing unit is the same, they are sorted from large to small according to the memory capacity; if the memory capacity is the same, they are sorted from large to small according to the hard disk capacity; if the hard disk capacity is the same, they are sorted randomly.

3. The sub-area division method of the digital twin platform of the space computing system according to claim 1 is characterized in that: The remaining available capacity of S3.1 refers to the designed available capacity of the digital twin platform server minus the designed capacity of the twin of the server that has been divided into the current digital twin platform.

4. The sub-area division method of the digital twin platform of the space computing system according to claim 1 is characterized in that: The step five includes the following sub-steps: S5.1: Divide the number of all unpartitioned slave twins by the estimated number of slave twins that can be accommodated by the server of a single digital twin platform obtained in step 4 to obtain the number of unpartitioned slave twin partitions Num_fs; From all unpartitioned slave star twins, select the slave star twins with the highest number of CPU cores in descending order, and assign each of them to a server of the digital twin platform as the core of their respective slave star twin sub-areas; S5.2: Calculate the remaining available capacity of the server of the digital twin platform, and take the minimum values ​​of the number of cores, memory capacity, and hard disk capacity of the central processing unit to obtain the minimum value of the remaining available capacity of the server of the digital twin platform; S5.3: Filter by capacity, and select the first Num_fs slave star twins sorted by capacity from the slave star twins whose capacities are all less than the minimum value of the remaining available capacity; S5.4: Calculate the orbital similarities between the first Num_fs slave star twins and the core of each slave star twin sub-region respectively, and form an orbital similarity reciprocal matrix with Num_fs rows and Num_fs columns by the reciprocals of the orbital similarities; S5.5: Use the Hungarian algorithm to solve the similarity inverse matrix, so that each slave star twin sub-region is assigned to a slave star twin, and the sum of the similarity inverses of each slave star twin and the corresponding slave star twin sub-region core is the smallest; S5.6: Repeat S5.2 to S5.5 until there is no slave twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the core of the slave twin sub-area is located; S5.7: Determine whether there are any undivided slave star twins. If yes, return to S5.1; if not, end the loop; at this time, all slave star twins are divided into the slave star twin sub-area.

5. The sub-area division method of the digital twin platform of the space computing system according to claim 4 is characterized in that: The orbital similarity is calculated using six orbital numbers.

6. The sub-area division method of the digital twin platform of the space computing system according to claim 1 is characterized in that: The step six includes the following sub-steps: S6.1: Divide the number of ground station twins by the estimated number of ground station twins that can be accommodated by the server of a single digital twin platform to obtain the number of ground station twin partitions Num_gs; select the ground station twins with the highest number of cores of the Num_gs central processors in descending order, and divide each of them into a server of a digital twin platform as the core of the ground station twin sub-area where they are located; S6.2: Calculate the remaining available capacity of the servers of the digital twin platform, and take the minimum value of the remaining available capacity of the servers of the digital twin platform; S6.3: Filter by capacity, and select the first Num_gs ground station twins sorted by capacity from the ground station twins whose capacities are all less than the minimum value of the remaining available capacity; S6.4: Calculate the Euclidean distances between the first Num_gs ground station twins and the core of each ground station twin sub-area, respectively, to form a Euclidean distance matrix with Num_gs rows and Num_gs columns; S6.5: Solve the Euclidean distance matrix using the Hungarian algorithm so that each ground station twin sub-area is assigned to a ground station twin, and the sum of the Euclidean distances between each ground station twin and the core of the corresponding ground station twin sub-area is minimized; S6.6: Repeat S6.2 to S6.5 until there is no ground station twin whose capacity is less than the remaining available capacity of the server of the digital twin platform where the core of the ground station twin sub-area is located; S6.7: Determine whether there are any undivided ground station twins. If so, return to S6.1; otherwise, the loop ends, and all ground station twins are now divided into ground station twin sub-areas.

7. The sub-area division method of the digital twin platform of the space computing system according to claim 6 is characterized in that: The Euclidean distance in S6.4 is calculated using the actual longitude and latitude of the two ground stations.

8. The sub-area division method of the digital twin platform of the space computing system according to claim 1 is characterized in that: The slave satellites in the space satellites are computing satellites that receive information from satellites of other systems and perform primary computing; the master satellites in the space satellites are computing satellites that receive information from the slave satellites and perform centralized computing; The information of one slave star is only transmitted to one master star, and one master star can receive information of multiple slave stars.

9. A sub-area division device for a digital twin platform of a space computing system, characterized in that: It includes one or more processors for implementing the sub-area division method of the digital twin platform of the space computing system as described in any one of claims 1 to 8.

Images