Space-based system service flexible optimization method and device based on distributed proxy self-healing

By adopting distributed proxy self-healing methods in space-based systems, a two-layer hypernetwork model and a multi-scale elastic model are built, which solves the problem that traditional space-based systems are difficult to quickly and autonomously restore services when nodes fail, and achieves efficient and rapid service recovery and resource utilization.

CN120151171APending Publication Date: 2025-06-13NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202510299523.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-13
Publication Date
2025-06-13

AI Technical Summary

Technical Problem

When traditional space-based systems face node failure, it is difficult to quickly and independently restore services, resulting in service interruption and failure, and cannot meet users' needs for service stability and reliability.

Method used

Using a distributed proxy self-healing method, a two-layer hypernetwork model and a multi-scale elastic model are built, and an intelligent recovery strategy for module failure and node failure is established, and service recovery is achieved through optimization algorithms.

Benefits of technology

It improves the elastic performance of space-based system services, reduces service recovery time, realizes independent and rapid recovery on track, and improves the system's service recovery effect and resource utilization rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120151171A_ABST
    Figure CN120151171A_ABST
Patent Text Reader

Abstract

The invention discloses a space-based system service flexible optimization method and device based on distributed agent self-healing, and the method comprises the steps: constructing a user service-oriented application scene of a space-based system according to a multifunctional service provided by the space-based system; based on the task network and the resource network, constructing a two-layer super network model according to a mapping relation between the task network and the resource network; constructing a space-based system multi-scale elastic model according to an application scene based on a two-layer hypernetwork model; establishing a space-based resource pool based on a satellite cluster in a resource network according to the ground user density and the satellite node resource carrying condition; establishing a space-based system elastic service recovery strategy according to the space-based system multi-scale elastic model under the constraint of the space-based resource pool; and optimizing and executing the space-based system elastic service recovery strategy by setting an optimization algorithm to finish space-based system service elastic optimization. The method can effectively improve the service elasticity of the system and reduce the service recovery time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of space-based system service elasticity optimization, and specifically relates to a method and device for optimizing the service elasticity of a space-based system based on distributed agent self-healing. Background Art

[0002] Space-based systems can provide services such as navigation, communication, and remote sensing for global users, and have extremely high social and economic value. However, due to being in space orbits, ground facilities are prone to failure due to extreme weather or natural disasters, and satellite nodes are prone to failure due to space environment radiation, etc. They may also be subject to targeted interference or even attacks. In this case, the quality of space-based system services will be greatly reduced, and even problems such as service interruption and failure may occur. In such a situation, traditional space-based systems generally have difficulty in continuing to maintain services. However, with the development of technologies such as software-defined satellites, satellite edge computing, and 6G, the functions of satellites in intelligent space-based systems can be reconfigured on demand. When some nodes fail, the system can still maintain certain performance and not collapse through task migration, that is, the service has elasticity. For users, the stability and reliability of intelligent space-based system services are very important. In addition to measures such as enhancing the capabilities of space-based systems, increasing system redundancy, and designing new architectures, which belong to the pre-launch deployment of satellites, the estimation and maintenance of service stability during the on-orbit stage are also extremely crucial aspects. Because it is inevitable for intelligent space-based systems to fail. However, the complex and changeable space environment, high maintenance costs, and limited on-orbit adjustment capabilities have always been challenges for maintaining the stability of space information services during on-orbit operation. Therefore, people hope that satellites can autonomously and quickly recover services on orbit to maintain service continuity and stability.

[0003] Currently, in order to improve the rapid recovery ability of space-based systems, the traditional methods generally adopted are to optimize network topologies and routing, increase redundant nodes and links, design fault detection mechanisms, etc. However, these strategies are all based on traditional single-layer satellite networks and lack considerations of features such as multi-service mixing, dynamic cascading failures, and underlying resource constraints in existing intelligent space-based systems such as communication-navigation integration and multi-orbit mixing. Most importantly, they cannot accurately and hierarchically and quickly recover failed services autonomously to maintain service stability as fast as possible and with the greatest ability, and reduce the interruption time.

[0004] Therefore, how to invent a method for optimizing the service elasticity of a space-based system, which can effectively improve the service elasticity performance of the system and reduce the service recovery time, has become an urgent problem to be solved. Summary of the Invention

[0005] To this end, the present invention provides a method and device for optimizing the service elasticity of a space-based system based on distributed agent self-healing. By establishing a multi-scale elasticity model of the space-based system under a two-layer hypernetwork model, the granularity of elasticity analysis is divided into three levels: module level, node level, and system level. Under the constraints of the space-based resource pool, intelligent space-based system elasticity service recovery strategies are established for two cases: module failure and node failure. Through a node-level space-based system service elasticity optimization method applicable to self-healing in the space environment and a service recovery algorithm based on the maximum elasticity of the shortest path, the service recovery strategy of the intelligent space-based system is optimized and executed, which can effectively improve the service elasticity performance of the system and reduce the service recovery time.

[0006] To achieve the above object, the present invention provides the following technical solutions: A method for optimizing the service elasticity of a space-based system based on distributed agent self-healing, comprising:

[0007] Construct an application scenario of the space-based system for user services according to the multi-functional services provided by the space-based system;

[0008] Based on the task network and the resource network, construct a two-layer hypernetwork model according to the mapping relationship between the task network and the resource network;

[0009] Based on the two-layer hypernetwork model, construct a multi-scale elasticity model of the space-based system according to the application scenario;

[0010] According to the ground user density and the satellite node resource loading situation, establish a space-based resource pool based on satellite clusters in the resource network;

[0011] Under the constraints of the space-based resource pool, establish a space-based system elasticity service recovery strategy according to the multi-scale elasticity model of the space-based system;

[0012] Optimize and execute the space-based system elasticity service recovery strategy through a set optimization algorithm to complete the optimization of the space-based system service elasticity.

[0013] As a preferred solution of the method for optimizing the service elasticity of a space-based system based on distributed agent self-healing, the granularity of elasticity analysis of the multi-scale elasticity model of the space-based system is divided into module level, node level, and system level;

[0014] The elasticity index expression of the multi-scale elasticity model of the space-based system is:

[0015]

[0016] In the formula, F is the elasticity index; D is the number of state variables of the space-based system; N is the number of nodes of the space-based system; and are the resource states of two satellite nodes for interaction; It is the mutual information in information theory, which measures the influence of one satellite on another satellite.

[0017] As an optimal solution of the space-based system service elasticity optimization method based on distributed agent self-healing, the space-based system elastic service recovery strategy includes: module failure recovery strategy and node failure recovery strategy;

[0018] The module failure recovery strategy is: when the set module on the set node fails, the agent of the intelligent satellite judges whether there is a backup redundant set module in the set node; if so, start the backup redundant set module; if not, request the node in the cluster that has the set module, and maintain the service through task migration;

[0019] The node failure recovery strategy is: detect the node status through the heartbeat mechanism; if the node fails, initiate a service recovery process in the space-based resource pool through the distributed intelligent satellite agent strategy.

[0020] As an optimal solution of the space-based system service elasticity optimization method based on distributed agent self-healing, during the process of service recovery through the space-based system elastic service recovery strategy, two mechanisms are set for the decision-making and control rights of the service recovery process;

[0021] The first mechanism is: the decision-making and control rights of the service recovery process lie in the node that discovers the failure and the node that initiates the recovery process;

[0022] The second mechanism is: the decision-making and control rights of the service recovery process lie in the preset decision-making satellite in the space-based resource pool.

[0023] As an optimal solution of the space-based system service elasticity optimization method based on distributed agent self-healing, during the process of optimizing and executing the space-based system elastic service recovery strategy through the set optimization algorithm, the optimization process includes a recovery consensus formation stage based on network communities and a service recovery execution stage;

[0024] The steps of the recovery consensus formation stage based on network communities are:

[0025] The control node statistically obtains the number of all failed nodes, and circularly judges the specific failure conditions of each failed node;

[0026] If it is a component failure of a failed node, judge whether there is a replaceable resource component in the failed node. If so, replace it and update the resource network, which will not affect the task network; if not, search for available resource components in the space-based resource pool;

[0027] Starting from the failed node itself, search for recoverable paths within the community through the BSF algorithm; if there are recoverable paths, record the set of recoverable path solutions; if not, further expand the search to between communities;

[0028] Through the inter-community search, if there are recoverable paths, record the set of recoverable path solutions; if not, determine that the failed task cannot be recovered;

[0029] Summarize the recoverable paths of the failed node to obtain the range of recoverable paths;

[0030] The steps of the service recovery execution stage are as follows:

[0031] Find the missing links in the task network and divide the missing links into a starting point group and an ending point group;

[0032] According to the inter-satellite link rate, convert the system and task network into a weighted directed graph;

[0033] Through the Edmond-Karp shortest path algorithm, find the minimum path between the starting point and the ending point of the task network corresponding to the failed node to obtain the shortest path;

[0034] Judge whether the number of paths of the shortest path is greater than 1; if so, map the shortest path to the resource network and calculate the elasticity values respectively; if not, continue to find paths through the Edmond-Karp shortest path algorithm;

[0035] Judge the next failed node; calculate the elasticity of the space-based system according to the resource capabilities of the remaining nodes;

[0036] Convert the obtained path into a configuration strategy of satellite nodes, output and execute to restore the service.

[0037] The present invention also provides a space-based system service elasticity optimization device based on distributed agent self-healing. Based on the above space-based system service elasticity optimization method based on distributed agent self-healing, it includes:

[0038] An application scenario construction module, configured to construct an application scenario of the space-based system for user services according to the multi-functional services provided by the space-based system;

[0039] A two-layer hypernetwork model construction module, configured to construct a two-layer hypernetwork model based on the task network and the resource network according to the mapping relationship between the task network and the resource network;

[0040] A space-based system multi-scale elasticity model construction module, configured to construct a space-based system multi-scale elasticity model based on the two-layer hypernetwork model according to the application scenario;

[0041] The space-based resource pool establishment module is used to establish a space-based resource pool based on satellite clusters in the resource network according to the ground user density and the resource loading situation of satellite nodes;

[0042] The space-based system elastic service recovery strategy establishment module is used to establish a space-based system elastic service recovery strategy according to the multi-scale elastic model of the space-based system under the constraint of the space-based resource pool;

[0043] The space-based system service elasticity optimization module is used to optimize and execute the space-based system elastic service recovery strategy by setting an optimization algorithm to complete the space-based system service elasticity optimization.

[0044] As a preferred solution of the space-based system service elasticity optimization device based on distributed agent self-healing, in the multi-scale elastic model construction module of the space-based system, the elastic analysis granularity of the multi-scale elastic model of the space-based system is divided into module level, node level and system level;

[0045] The elastic index expression of the multi-scale elastic model of the space-based system is:

[0046]

[0047] In the formula, F is the elastic index; D is the number of state variables of the space-based system; N is the number of nodes of the space-based system; and are the resource states of two satellite nodes for interaction; is the mutual information in information theory, which measures the influence of one satellite on another satellite.

[0048] As a preferred solution of the space-based system service elasticity optimization device based on distributed agent self-healing, in the space-based system elastic service recovery strategy establishment module, the space-based system elastic service recovery strategy includes: module failure recovery strategy and node failure recovery strategy;

[0049] The module failure recovery strategy is: when the set module on the set node fails, the agent of the intelligent satellite judges whether there is a backup redundancy of the set module in the set node; if so, start the backup redundancy of the set module; if not, request the node in the cluster that has the set module to maintain the service through task migration;

[0050] The node failure recovery strategy is: detect the node status through the heartbeat mechanism; if the node fails, initiate a service recovery process in the space-based resource pool through the distributed intelligent satellite agent strategy.

[0051] As an optimal solution for the space - based system service elasticity optimization device based on distributed agent self - healing, in the space - based system elastic service recovery policy establishment module, during the process of service recovery through the space - based system elastic service recovery policy, two mechanisms are set for the decision - making and control rights of the service recovery process;

[0052] The first mechanism is that the decision - making and control rights of the service recovery process lie with the node that discovers the failure and the node that initiates the recovery process;

[0053] The second mechanism is that the decision - making and control rights of the service recovery process lie with the preset decision - making satellite in the space - based resource pool.

[0054] As an optimal solution for the space - based system service elasticity optimization device based on distributed agent self - healing, in the space - based system service elasticity optimization module, the optimization process includes a recovery consensus formation stage based on network communities and a service recovery execution stage;

[0055] The optimization sub - module in the recovery consensus formation stage based on network communities includes:

[0056] The failed node data and failure situation statistics sub - module is used to control the node to count and obtain the number of all failed nodes, and cyclically judge the specific failure situation of each failed node;

[0057] The failed node judgment and processing sub - module is used to, if the component of the failed node fails, judge whether there is a replaceable resource component in the failed node. If there is, replace it and update the resource network without affecting the task network; if not, search for available resource components in the space - based resource pool;

[0058] The community search sub - module is used to start from the failed node itself and search for recoverable paths within the community range through the BSF algorithm; if there are recoverable paths, record the set of recoverable path solutions; if not, further expand the search to inter - community search;

[0059] The inter - community search sub - module is used to, through the inter - community search, if there are recoverable paths, record the set of recoverable path solutions; if not, determine that the failed task cannot be recovered;

[0060] The recoverable path range acquisition sub - module is used to summarize the recoverable paths of the failed nodes to obtain the recoverable path range;

[0061] The sub - module in the service recovery execution stage includes:

[0062] The missing link search and grouping sub - module is used to find the missing links from the task network and divide the missing links into a starting point group and an ending point group;

[0063] A weighted direction diagram conversion sub-module, configured to convert a system and task network into a weighted direction diagram according to the inter-satellite link rate;

[0064] A shortest path acquisition sub-module, configured to find a minimum path between the start point and the end point of the task network corresponding to a failed node through the Edmond-Karp shortest path algorithm, and acquire the shortest path;

[0065] A shortest path number judgment sub-module, configured to judge whether the number of paths of the shortest path is greater than 1; if so, map the shortest path to a resource network and calculate the elasticity values respectively; if not, continue to search for paths through the Edmond-Karp shortest path algorithm;

[0066] A failed node loop judgment sub-module, configured to judge the next failed node; calculate the elasticity of the space-based system according to the resource capabilities of the remaining nodes;

[0067] A configuration policy output and execution sub-module, configured to convert the acquired path into a configuration policy of satellite nodes, output and execute it, and resume services.

[0068] The present invention has the following advantages: the present invention constructs an application scenario of user-oriented services of the space-based system according to the multifunctional services provided by the space-based system; based on the task network and the resource network, a two-layer super network model is constructed according to the mapping relationship between the task network and the resource network; based on the two-layer super network model, a multi-scale elastic model of the space-based system is constructed according to the application scenario; according to the density of ground users and the satellite node resource carrying situation, a space-based resource pool based on a satellite cluster is established in the resource network; under the constraint of the space-based resource pool, according to the multi-scale elastic model of the space-based system, a space-based system elastic service recovery strategy is established; by setting an optimization algorithm, the space-based system elastic service recovery strategy is optimized and executed to complete the space-based system service elasticity optimization. The present invention solves the problem that the space-based system is overly dependent on ground operation and control and cannot manually restore failed nodes to restore services. Through the distributed agent self-healing method, the space-based system is able to rely only on the remaining capacity of the system and try its best to restore the service before the node fails. Compared with the existing ground manual control or the method triggered based on preset conditions, the intelligent space-based system has a higher system service recovery effect and less recovery time than the traditional space-based system because it can autonomously perform service recovery on orbit. The present invention can divide the effectiveness of space-based system services, the part of failure location and the degree of service recovery into finer granularity through three-level elastic division of module level, node level and system level, which can locate more accurately and increase the utilization rate of system resources. The present invention can realize unified modeling of service elasticity of heterogeneous hybrid space-based systems that provide different services such as navigation, communication and remote sensing by modeling the performance and elasticity of intelligent space-based systems based on information entropy and mutual information. The elastic model can not only reflect the recoverability of service capabilities, but also describe the dependence of elasticity on network connectivity. The present invention establishes two recovery modes for module failure and node failure, and also establishes a recovery mechanism for two situations of total resource failure and node interruption and loss of connection in the node failure mode, which solves the problem that the space-based system does not rely on human control and the satellite can self-heal on orbit. The present invention proposes a service recovery algorithm based on the shortest path and maximum elasticity. Compared with a random recovery algorithm, the algorithm has a shorter service recovery time and a faster recovery speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0070] The structures, ratios, sizes, etc. shown in this specification are only used to match the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the implementation conditions of the present invention. Therefore, they do not have substantial technical significance. Any modification of the structure, change in the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0071] Figure 1 It is a schematic flow diagram of the space - based system service elasticity optimization method based on distributed agent self - healing provided in Embodiment 1 of the present invention;

[0072] Figure 2 It is a schematic diagram of the intelligent space - based system resource level division in the space - based system service elasticity optimization method based on distributed agent self - healing provided in Embodiment 1 of the present invention;

[0073] Figure 3 It is a schematic diagram of the two - layer network packet mapping of the intelligent space - based system in the space - based system service elasticity optimization method based on distributed agent self - healing provided in Embodiment 1 of the present invention;

[0074] Figure 4 It is a schematic diagram of the service recovery of the intelligent space - based system in the space - based system service elasticity optimization method based on distributed agent self - healing provided in Embodiment 1 of the present invention;

[0075] Figure 5 It is a schematic flow diagram of the recovery consensus formation algorithm based on distributed agents in the space - based system service elasticity optimization method based on distributed agent self - healing provided in Embodiment 1 of the present invention;

[0076] Figure 6 It is a schematic flow diagram of the service elasticity optimization algorithm based on the maximum elasticity of the shortest path in the space - based system service elasticity optimization method based on distributed agent self - healing provided in Embodiment 1 of the present invention;

[0077] Figure 7 It is a schematic diagram of the elasticity ability test based on three artificial synthetic networks in a possible embodiment provided in Embodiment 1 of the present invention;

[0078] Figure 8 It is a schematic diagram of the node failure process of the space information network in a possible embodiment provided in Embodiment 1 of the present invention;

[0079] Figure 9 It is a schematic diagram of the architecture of the space - based system service elasticity optimization device based on distributed agent self - healing provided in Embodiment 2 of the present invention. Detailed implementation manners

[0080] The following specific embodiments illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0081] Embodiment 1

[0082] Refer to Figure 1 , Embodiment 1 of the present invention provides a method for optimizing the service elasticity of a space-based system based on distributed agent self-healing, including the following steps:

[0083] S1. According to the multi-functional services provided by the space-based system, construct an application scenario of the space-based system for user services;

[0084] S2. Based on the task network and the resource network, construct a two-layer hypernetwork model according to the mapping relationship between the task network and the resource network;

[0085] S3. Based on the two-layer hypernetwork model, construct a multi-scale elasticity model of the space-based system according to the application scenario;

[0086] S4. According to the ground user density and the satellite node resource loading situation, establish a space-based resource pool based on satellite clusters in the resource network;

[0087] S5. Under the constraint of the space-based resource pool, establish an elastic service recovery strategy for the space-based system according to the multi-scale elasticity model of the space-based system;

[0088] S6. Optimize and execute the elastic service recovery strategy of the space-based system by setting an optimization algorithm to complete the optimization of the service elasticity of the space-based system.

[0089] In this embodiment, in step S1, according to the multi-functional services provided by the space-based system, construct an application scenario of the space-based system for user services;

[0090] Specifically, the intelligent space-based system provides services such as navigation, communication, and remote sensing in an integrated manner, and the service execution carriers are specific sensing, computing, and other resource modules of the satellite; according to the multi-functional services provided by the space-based system, establish an application scenario of the intelligent space-based system for user services.

[0091] In this embodiment, in step S2, based on the task network and the resource network, construct a two-layer hypernetwork model according to the mapping relationship between the task network and the resource network;

[0092] Specifically, satellites in the intelligent space-based system collaborate with each other according to mission requirements, which can not only enhance service capabilities as needed but also improve service elasticity. For example Figure 3 As shown, the resource network configures, organizes, and schedules satellite resources according to the task network to provide resource services for regional users.

[0093] In the mission-oriented intelligent space-based system, there is also a community structure similar to that in complex networks. A typical example is a satellite cluster that collaborates on-demand for regional services in the entire network or a satellite group that executes distributed tasks. The characteristics of such satellite clusters are that the internal network connections within the satellite clusters participating in collaboration are tight, with similar functions or the same goals, and relatively few interconnections with other satellites. This is a very typical community feature in complex networks. The community structure in the present invention is different from the resource pool in the resource network and is distributed in the task network.

[0094] In this embodiment, in step S3, based on the two-layer hypernetwork model, a multi-scale elasticity model of the space-based system is constructed according to the application scenario;

[0095] Specifically, as Figure 2 shown, the elasticity analysis granularity of the multi-scale elasticity model of the space-based system is divided into module level, node level, and system level;

[0096] Among them, at the module level, the on-board computer or the master control module of each satellite node will monitor the health status of each module of the satellite in real time and maintain the resource usage, remaining amount, and health status of each resource module.

[0097] At the node level, for a satellite composed of various modules, it is equivalent to a small subsystem, and the satellite node maintains the health status of its own resources. At the same time, it periodically confirms the online status of nodes within the satellite cluster that constitutes the resource pool through a heartbeat mechanism.

[0098] At the system level, the space-based system, as a space information system, is composed of a large number of satellites interconnected by inter-satellite links. From the perspective of the system level, the performance change of each of its satellite nodes follows the following equation:

[0099]

[0100] In the formula, F(x i ) is the auto-dynamic function of the satellite; FF(x i , x j ) is the interaction function of the satellites; x i , x j are different resource blocks in the satellite node; N is the number of resource blocks; A ij is the adjacency matrix of the network topology.

[0101] The elastic performance of the space-based system is jointly simulated through the auto-dynamic function and the interaction function of the satellite. The interaction function is related to the entire space-based system, especially the network topology of each node and the system. Considering that the interaction function needs to establish a deterministic and monotonic mapping relationship between the elastic ability and performance, and can also uniformly describe the multiple service performances of the intelligent space-based system, the present invention will, based on space information theory, model the performance and elasticity of the intelligent space-based system through information entropy and mutual information. Since the intelligent space-based system provides integrated services such as navigation, communication, and remote sensing, and the performance attributes of different functional domains are different, in order to enable the elasticity to reflect both the recoverability of the service ability and describe the dependence of the elasticity on network connectivity, the present invention proposes a method based on a probability model from the perspective of space information theory to uniformly quantify the service elastic ability of the intelligent space-based system.

[0102] In this embodiment, the elastic index expression of the multi-scale elastic model of the space-based system is:

[0103]

[0104] In the formula, F is the elastic index; D is the number of state variables of the space-based system; N is the number of nodes of the space-based system; and are the resource states of two satellite nodes for interaction; is the mutual information in information theory, which measures the influence of one satellite on another satellite.

[0105]

[0106] In the formula, is the information entropy of the resource state ; is the conditional entropy of the mutual information dependence between the resource states x i and x j ;

[0107] The performance state θ of a node providing services is related to its resource ability, defined as the probability The information dependence relationship is Through the mapping of the service elastic performance of the intelligent space-based system by information entropy and mutual information, the probability model based on resource ability solves the problem of differences between the performances of different functions, and can also quantify the importance of the network connectivity of the intelligent space-based resources to service elasticity.

[0108] In this embodiment, the architecture of the intelligent space-based system based on the hypernetwork maps the task network to the resource network and then to the physical network, realizing a finer-grained elastic decomposition of services. Through information entropy and mutual information, the performance of multiple system services is unified, and the overall effective dynamics of the networked system are mapped to its topological structure, realizing the conversion from a complex network to the elasticity of a multi-layer network. More importantly, the granularity of elastic modeling is established in the resource module of the nodes of the intelligent space-based system. Finer modeling enables more accurate elastic analysis and is also a means of efficient resource utilization.

[0109] In this embodiment, in step S4, according to the ground user density and the satellite node resource loading situation, a space-based resource pool based on satellite clusters is established in the resource network;

[0110] Specifically, as Figure 3 the satellite grouping in is carried out according to the user service requirements in different regions. By constructing a space-based resource pool in a collaborative manner, the nodes in the resource pool provide better services through mutual cooperation and resource sharing, such as distributed radar and cooperative communication. Through the resource organization mode of the resource pool, the service efficiency can be improved. In addition, the mode of the resource pool can also promote cooperation and reduce the impact of fault diffusion. Especially in the case where ground operation and control are not required, the intelligent space-based system can continue to maintain services autonomously through this method.

[0111] In this embodiment, in step S5, under the constraint of the space-based resource pool, according to the multi-scale elastic model of the space-based system, an elastic service recovery strategy for the space-based system is established;

[0112] Specifically, due to the inevitability of space-based system failures and node failures, the key to elastic optimization lies in the service recovery process after the system encounters node failures. Following the principle of trying to recover autonomously, relying on the remaining capabilities of the space-based system to achieve the maximum possible maintenance and recovery of the service capabilities of the space-based system. The present invention will focus on the immediacy of service recovery because in some emergencies, such as disaster recovery, emergency rescue, and even war scenarios, the primary factor to consider is the recovery time, followed by the recovery effect. Therefore, it is very crucial to design and optimize a recovery strategy that can quickly restore the service performance of the intelligent space-based system.

[0113] In this embodiment, each intelligent satellite has an intelligent agent, such as a health monitoring and management system. That is, through certain technical means, the intelligent satellite agent can sense and locate the failure of the node itself module. If the satellite completely fails and goes offline, this information can be identified by the agents of adjacent satellites through means such as heartbeat detection. The present invention considers the service failure scenarios as shown in Figure 4 Figure 4The process of task flow interruption in the task network caused by the failure of resource nodes in the resource network is shown. In the figure, the resource node S 1 fails due to interference or attack. From the perspective of the task network, the related tasks will be affected or even interrupted. For example Figure 4 the task flow b in the task network shown in. At this time, T 1 、T 2 、T 3 and T 4 Since they are intelligent satellites, they can redefine their functions as needed to restore interrupted services. After judging that the system fails, in order to optimize the service resilience of the space-based system, the following two types of recovery are proposed:

[0114] Module failure: When a certain module on a single node fails, the agent of the intelligent satellite first judges whether there is a backup redundant module in the node. If so, the backup can be started. This step is available in the key core modules of most current satellites; if not, it requests the nodes in the cluster that have this module and maintains the service through task migration;

[0115] Node failure: The difference between node failure and module failure is that all modules of the entire node can no longer provide services. Generally speaking, a node undertakes multiple services. Therefore, to maintain the service, it is necessary to migrate as many services as possible undertaken by this node to other nodes. To achieve batch service migration of nodes, the space-based system service resilience optimization method and device based on distributed agent self-healing proposed by the present invention are required.

[0116] The service recovery of the intelligent space-based system is realized through distributed intelligent satellite agents, without relying on the ground operation and control center, which is an embodiment of service resilience self-healing autonomy. For the case of module failure, the agent of the intelligent satellite itself can detect the failure situation and update the node health status in real time. For the case of node failure, since all satellites are moving at high speed in orbit, the more common situation is that the node suddenly interrupts and loses contact after failure, or all resource modules of the node fail, and only the communication module for measurement and control works normally. The second situation can be regarded as the case where the resource health status monitors a value of zero, which is equivalent to the case of module failure. For the case of sudden interruption and loss of contact, the present invention establishes a heartbeat mechanism based on the status monitoring of the nearest satellite. This mechanism is based on the inter-satellite link module carried by the satellite itself. For example, a general node is equipped with four inter-satellite link modules, two for the same orbit and two for different orbits. Then, in a heartbeat monitoring, responses from four nodes will be obtained. If there is no response twice, it is judged that the corresponding satellite node has interrupted and lost contact. Therefore, through this design, other satellites in a resource pool can autonomously discover the existence of node failure situations in orbit through the heartbeat detection mechanism, and at the same time initiate a service recovery process in the resource pool.

[0117] In this embodiment, in the design of decision-making and control rights for the service recovery process, two mechanisms are provided. One is for the node that discovers the failed node and initiates the recovery process. At this time, only the service of the corresponding failed module or node is recovered. The second mechanism is the preset decision satellite in the resource pool. The decision satellite is in the resource pool and is determined according to the resource ranking of the satellite. Each time, the satellite with the highest resource amount is given priority. Because satellites with more resources generally have more surplus resources for decision-making, which is beneficial to improving the efficiency of service recovery. When the node with the highest resource amount fails, the satellites are sequentially configured as decision-making and control nodes in descending order.

[0118] In this embodiment, in step S6, the space-based system elastic service recovery strategy is optimized and executed by setting an optimization algorithm to complete the elastic optimization of the space-based system service.

[0119] Specifically, for the node selection and task migration of service recovery, the service elastic optimization problem model of the intelligent space-based system is established on the basis of the two-layer network diagram of the space-based system described above, and the scenario is as Figure 4 shown. In this problem context, the problem to be solved by the elastic service recovery method of the intelligent space-based system is that after the task flow of the task network is interrupted due to the failure of some nodes in the resource network, use the shortest possible recovery time to restore the original task network flow to the greatest extent. There are two goals here, namely the shortest recovery time and the maximum recovery effect. The recovery effect refers to the degree of restoring the task network flow.

[0120] For a given two-layer network The recovery strategy is ST and the recovery time is rt. This problem is expressed as follows:

[0121]

[0122] min rt

[0123] In the formula, RL is the elastic performance constructed by the formula min rt. Since this elastic performance can calculate the system service performance, the recovery effect of it is used to represent the recovery of the service performance. The shortest recovery time rt refers to the shortest time required to restore the task flow interrupted due to node or link failure.

[0124] In this embodiment, in terms of the task network, if the community structure has been formed before service recovery, the community needs to be updated on this basis; if the community structure has not been formed, it is necessary to first establish the community structure according to the failure characteristics and then perform service recovery. The significance of constructing the community structure is that, on the one hand, it is to use distributed agents to achieve information synchronization within the entire space-based system, so as to get rid of ground manual control and achieve service self-healing; at the same time, it also reduces the search operation range and the computing space for recovery. Therefore, according to this design, the community structure division of the space-based system is generally completed before the fault to cope with the uncertain fault failure problem.

[0125] The difficulty in community division for service recovery in the intelligent space-based system lies in that it is a heterogeneous information network, different from the traditional homogeneous information network, which is equivalent to multiple types of communities sharing multiple identical features. Although it increases the network type characteristics, it can also more accurately describe the accuracy of the network community divided based on features such as tasks and geography. In the hypernetwork modeling of the intelligent space-based system, the architecture of the intelligent space-based system has been tensor decomposed into resources, grids, time slots, etc. Therefore, the present invention uses the method of tensor factorization to divide the community of the intelligent space-based system. The method based on tensor factorization can be carried out based on the hypergraph, which will not be discussed in detail in the present invention. After the division is completed, the optimization of the service recovery and execution phases can be carried out.

[0126] In this embodiment, the optimization process includes a recovery consensus formation phase based on the network community and a service recovery execution phase;

[0127] As Figure 5 shown, the steps of the recovery consensus formation phase based on the network community are as follows:

[0128] S601. The control node statistically obtains the number of all failed nodes and repeatedly judges the specific failure situation of each failed node;

[0129] Specifically, the control node statistically obtains the number of all failed nodes and repeatedly judges the specific failure situation of each failed node;

[0130] S602. If it is a component failure of a failed node, it is judged whether there is a replaceable resource component in the failed node. If there is, replace it and update the resource network without affecting the task network; if not, search for available resource components in the space-based resource pool;

[0131] Specifically, if it is a component failure of this node, first judge whether there is a replaceable resource component in this node. If there is, directly replace it and update the resource network without affecting the task network; if not, start searching for available resource components in the resource pool;

[0132] S603. Starting from the failed node itself, search for recoverable paths within the community using the BSF algorithm; if there are recoverable paths, record the set of recoverable path solutions; if not, further expand the search to between communities.

[0133] Specifically, use the BSF algorithm to start from itself and search for recoverable paths within the community; if there are feasible paths, record the set of feasible path solutions; if not, further expand the search to between communities.

[0134] S604. Through the search between communities, if there are recoverable paths, record the set of recoverable path solutions; if not, determine that the failed task cannot be recovered.

[0135] Specifically, if a feasible path exists in the search between communities, record the set of feasible path solutions; if not, determine that the failed task cannot be recovered.

[0136] S605. Summarize the recoverable paths of the failed node to obtain the range of recoverable paths.

[0137] Specifically, summarize the range of recovery paths of the failed node. In the recovery consensus formation stage, only the community range for the task flow recovery of the failed node is determined, and the final feasible paths for each task flow need to be further determined.

[0138] As Figure 6 shown, the steps of the service recovery execution stage are as follows:

[0139] S606. Find the missing links in the task network and divide the missing links into a starting point group and an ending point group.

[0140] Specifically, find the missing links from the task network and divide them into a starting point group and an ending point group.

[0141] S607. Convert the system and task network into a weighted directed graph according to the inter-satellite link rate.

[0142] S608. Through the Edmond-Karp shortest path algorithm, find the minimum path between the starting point and the ending point of the task network corresponding to the failed node to obtain the shortest path.

[0143] Specifically, when a specific task is recovered, use the starting point at the failure moment as the new starting point, and then find the feasible path from this starting point to the final destination. At this time, the problem of elastic service recovery is transformed into the problem of the feasible flow set of the shortest path of the network, as follows:

[0144] minl(f)

[0145] In the formula, l(f) is the set of shortest paths;

[0146]

[0147] In the formula, ω ij is the weight; f ij is the path between satellites i and j; f(v i , v j ) is the path between resource nodes vi and vj; c i,j is the c resource capacity between satellites i and j; V RN is the node set of the resource network RN; tk fi is the earliest start time required by the task; te k is the latest completion time required by the task;

[0148] Among them, the first constraint represents the shortest time of the task network, the second represents that each edge must meet the resource capacity constraint, and the third represents that the task must be completed within the required time of the task, otherwise it is considered a task failure and cannot be restored.

[0149] S609. Determine whether the number of paths of the shortest path is greater than 1; if so, map the shortest path to the resource network and calculate the elasticity value respectively; if not, continue to find the path through the Edmond-Karp shortest path algorithm;

[0150] S610. Judge the next failed node; calculate the elasticity of the space-based system according to the resource capabilities of the remaining nodes;

[0151] S611. Convert the obtained path into a configuration strategy of satellite nodes, output and execute it to restore the service.

[0152] In this embodiment, the recovery consensus formation stage and the service recovery execution stage based on the network community are executed by the agent of the intelligent satellite, but they can be executed separately. This reduces the consumption of computing resources of the intelligent satellite. Due to the reason of distributed agent execution, the computational complexity of each agent is only O(N), which can greatly reduce the consumption of satellite computing.

[0153] In a possible embodiment, service elasticity test examples on three types of typical artificial synthetic networks are provided as follows:

[0154] Service elasticity tests are carried out on three types of typical artificial synthetic networks, namely random ER (Erdos–Rényi), scale-free SF (Scale-Free) and small-world SW (Small-World) three network types

[30] . Since a two-layer network needs to be constructed, these three types are respectively used to construct the first layer, and the SF network is used to construct the second layer. The service elasticity test results based on these three types of artificial synthetic networks are shown in Figure 7. By Figure 7It can be seen that when the three types of synthetic networks encounter attacks, as the number of failed nodes increases, the system resilience decreases. During the recovery process, it is the opposite. As the number of recovered nodes increases, the resilience of the system network gradually recovers. This verifies the correctness of the service resilience model regarding resilience.

[0155] Furthermore, the present invention can be tested on an actual space information network. Taking the test of the service recovery algorithm on the super network of the space information network with the number of satellite nodes ranging from 50 to 200 as an example. Since a super network model of the system network and the task network needs to be constructed, for the inter-satellite links of the system network, it is considered that all have communication capabilities under visible conditions. The tasks are randomly configured at the beginning to use satellite resources. The process of random failure of nodes in the space information network is as Figure 8 shown. Figure 8 The color of each node in it represents the connection degree of the node. The more links a satellite is connected to, the greater the node degree. From Figure 8 it can be seen that as the number of failed nodes increases, the number of node connections gradually decreases and the connection degree becomes lower.

[0156] In summary, the present invention constructs an application scenario for the space-based system to serve users based on the multi-functional services provided by the space-based system; constructs a two-layer hypernetwork model based on the task network and the resource network according to the mapping relationship between the task network and the resource network; constructs a multi-scale elastic model of the space-based system based on the two-layer hypernetwork model according to the application scenario; establishes a space-based resource pool based on satellite clusters in the resource network according to the density of ground users and the resource loading of satellite nodes; under the constraint of the space-based resource pool, establishes an elastic service recovery strategy for the space-based system according to the multi-scale elastic model of the space-based system; optimizes and executes the elastic service recovery strategy of the space-based system by setting an optimization algorithm to complete the elastic optimization of the space-based system service. The present invention solves the problem that the space-based system is overly dependent on ground operation and control and cannot manually recover failed nodes to restore services. Through the self-healing method of distributed agents, the space-based system realizes the ability to rely only on the remaining capabilities of the system and tries its best to restore the services before node failure. Compared with the existing methods of ground manual control or triggering based on preset conditions, the intelligent space-based system has higher system service recovery effect and less recovery time than the traditional space-based system because it can autonomously perform service recovery in orbit. Through the three-level elastic division at the module level, node level, and system level, the present invention can divide the effectiveness of the space-based system service, the part of failure location, and the degree of service recovery into finer granularities, enabling more accurate positioning and higher system resource utilization. By modeling the performance and elasticity of the intelligent space-based system based on information entropy and mutual information, the present invention can realize the unified modeling of the service elasticity of heterogeneous hybrid space-based systems providing different services such as navigation, communication, and remote sensing. This elastic model can not only reflect the recoverability of service capabilities but also describe the dependence of elasticity on network connectivity. The present invention establishes two recovery modes for module failure and node failure, and also establishes recovery mechanisms for two situations of full resource failure and node interruption and disconnection in the node failure mode, solving the problem that the space-based system does not rely on manual control and the satellites can autonomously self-heal in orbit. The present invention proposes a service recovery algorithm based on the maximum elasticity of the shortest path. Compared with the random recovery algorithm, this algorithm has less service recovery time and faster recovery speed.

[0157] It should be noted that the method of the embodiments of the present disclosure can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario and completed by the cooperation of multiple devices. In this case of a distributed scenario, one of the multiple devices can only execute one or more steps of the method of the embodiments of the present disclosure, and these multiple devices will interact with each other to complete the described method.

[0158] It should be noted that some embodiments of the present disclosure have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the above embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0159] Embodiment 2

[0160] Referring to Figure 9 , Embodiment 2 of the present invention further provides a space - based system service elasticity optimization device based on distributed agent self - healing, including:

[0161] An application scenario construction module 001, configured to construct an application scenario for the space - based system to serve users according to the multi - functional services provided by the space - based system;

[0162] A two - layer hyper - network model construction module 002, configured to construct a two - layer hyper - network model based on the task network and the resource network according to the mapping relationship between the task network and the resource network;

[0163] A space - based system multi - scale elasticity model construction module 003, configured to construct a space - based system multi - scale elasticity model based on the two - layer hyper - network model according to the application scenario;

[0164] A space - based resource pool establishment module 004, configured to establish a space - based resource pool based on satellite clusters in the resource network according to the ground user density and the resource loading situation of satellite nodes;

[0165] A space - based system elastic service recovery strategy establishment module 005, configured to establish a space - based system elastic service recovery strategy according to the space - based system multi - scale elasticity model under the constraint of the space - based resource pool;

[0166] A space - based system service elasticity optimization module 006, configured to optimize and execute the space - based system elastic service recovery strategy through a set optimization algorithm to complete the space - based system service elasticity optimization.

[0167] In this embodiment, in the space - based system multi - scale elasticity model construction module 003, the elasticity analysis granularity of the space - based system multi - scale elasticity model is divided into module - level, node - level, and system - level;

[0168] The elasticity index expression of the space - based system multi - scale elasticity model is:

[0169]

[0170] Wherein, F is the elasticity index; D is the number of state variables of the space-based system; N is the number of nodes of the space-based system; and are the resource states of two satellite nodes for interaction; is the mutual information in information theory, which measures the influence of one satellite on another satellite.

[0171] In this embodiment, in the space-based system elasticity service recovery policy establishment module 005, the space-based system elasticity service recovery policy includes: a module failure recovery policy and a node failure recovery policy;

[0172] The module failure recovery policy is: when the set module on the set node fails, the agent of the intelligent satellite judges whether there is a backup redundancy of the set module in the set node; if so, start the backup redundancy of the set module; if not, request the node in the cluster that has the set module to maintain the service through task migration;

[0173] The node failure recovery policy is: detect the node status through a heartbeat mechanism; if a node fails, initiate a service recovery process in the space-based resource pool through a distributed intelligent satellite agent policy.

[0174] In this embodiment, in the space-based system elasticity service recovery policy establishment module 005, during the service recovery process through the space-based system elasticity service recovery policy, two mechanisms are set for the decision-making and control rights of the service recovery process;

[0175] The first mechanism is: the decision-making and control rights of the service recovery process lie in the node that discovers the failure and the node that initiates the recovery process;

[0176] The second mechanism is: the decision-making and control rights of the service recovery process lie in the preset decision-making satellite in the space-based resource pool.

[0177] In this embodiment, in the space-based system service elasticity optimization module 006, the optimization process includes a recovery consensus formation stage based on network communities and a service recovery execution stage;

[0178] The optimization sub-module in the recovery consensus formation stage based on network communities includes:

[0179] The failed node data and failure situation statistics sub-module 061 is used to control the node to count and obtain the number of all failed nodes, and loop to judge the specific failure situation of each failed node;

[0180] The failed node judgment and processing sub-module 062 is used to, if a component failure occurs for a failed node, determine whether there is a replacement resource component within the failed node. If there is, replace it and update the resource network without affecting the task network; if not, search for available resource components in the space-based resource pool.

[0181] The community search sub-module 063 is used to start from the failed node itself and search for recoverable paths within the community range through the BSF algorithm; if there are recoverable paths, record the set of recoverable path solutions; if not, further expand the search to between communities.

[0182] The inter-community search sub-module 064 is used to, through the inter-community search, if there are recoverable paths, record the set of recoverable path solutions; if not, determine that the failed task cannot be recovered.

[0183] The recoverable path range acquisition sub-module 065 is used to summarize the recoverable paths of the failed node to obtain the recoverable path range.

[0184] The sub-modules in the service recovery execution phase include:

[0185] The missing link search and grouping sub-module 066 is used to find the missing links in the task network and divide the missing links into a starting point group and an ending point group.

[0186] The weighted direction graph conversion sub-module 067 is used to convert the system and task network into a weighted direction graph according to the inter-satellite link rate.

[0187] The shortest path acquisition sub-module 068 is used to find the minimum path between the starting point and the ending point of the task network corresponding to the failed node through the Edmond-Karp shortest path algorithm to obtain the shortest path.

[0188] The shortest path number judgment sub-module 069 is used to judge whether the number of paths of the shortest path is greater than 1; if so, map the shortest path to the resource network and calculate the elasticity value respectively; if not, continue to search for paths through the Edmond-Karp shortest path algorithm.

[0189] The failed node loop judgment sub-module 0610 is used to judge the next failed node; calculate the elasticity of the space-based system according to the resource capabilities of the remaining nodes.

[0190] The configuration policy output and execution sub-module 0611 is used to convert the obtained path into a configuration policy for satellite nodes, output and execute it to recover the service.

[0191] It should be noted that the information interaction, execution process, etc. between the above-mentioned system modules are based on the same concept as the method embodiment in Embodiment 1 of this application, and the technical effects brought by them are the same as those of the method embodiment of this application. For specific content, please refer to the description in the method embodiment shown above in this application, and details will not be repeated here.

[0192] Embodiment 3

[0193] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which program codes of a method for elastic optimization of space-based system services based on distributed agent self-healing are stored, and the program codes include instructions for executing the method for elastic optimization of space-based system services based on distributed agent self-healing in Embodiment 1 or any possible implementation manner thereof.

[0194] The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center integrating one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive (SSD)).

[0195] Embodiment 4

[0196] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0197] The processor and the memory communicate with each other through a bus; the memory stores program instructions executable by the processor, and the processor can execute the method for elastic optimization of space-based system services based on distributed agent self-healing in Embodiment 1 or any possible implementation manner thereof by calling the program instructions.

[0198] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor, which is implemented by reading software codes stored in the memory. The memory can be integrated in the processor or can exist independently outside the processor.

[0199] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.).

[0200] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing system. They can be concentrated on a single computing system or distributed on a network composed of multiple computing systems. Optionally, they can be implemented with program codes executable by the computing system. Thus, they can be stored in a storage system and executed by the computing system. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0201] Although the present invention has been described in detail above with general descriptions and specific embodiments, based on the present invention, some modifications or improvements can be made, which are obvious to those skilled in the art. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection required by the present invention.

Claims

1. A space-based system service elasticity optimization method based on distributed agent self-healing, characterized in that: include: Based on the multifunctional services provided by the space-based system, build application scenarios for space-based system services for users; Based on the task network and the resource network, a two-layer super network model is constructed according to the mapping relationship between the task network and the resource network; Based on the two-layer hypernetwork model and according to the application scenario, a multi-scale elastic model of the space-based system is constructed; According to the density of ground users and the satellite node resource loading conditions, a satellite cluster-based space-based resource pool is established in the resource network; Under the constraints of the space-based resource pool, establishing a space-based system elastic service recovery strategy according to the space-based system multi-scale elastic model; The elastic service recovery strategy of the space-based system is optimized and executed by setting an optimization algorithm to complete the elasticity optimization of the space-based system service.

2. The method for optimizing space-based system service elasticity based on distributed agent self-healing according to claim 1 is characterized in that: The elasticity analysis granularity of the multi-scale elasticity model of the space-based system is divided into module level, node level and system level; The elastic index expression of the multi-scale elastic model of the space-based system is: Where F is the elasticity index; D is the number of state variables of the space-based system; N is the number of nodes of the space-based system; and The resource status of the two satellite nodes that are interacting; It is the mutual information in information theory, which measures the influence of one satellite on another satellite.

3. The method for optimizing space-based system service elasticity based on distributed agent self-healing according to claim 2 is characterized in that: The space-based system elastic service recovery strategy includes: module failure recovery strategy and node failure recovery strategy; The module failure recovery strategy is as follows: when the setting module on the setting node fails, the agent of the intelligent satellite determines whether there is a backup redundant setting module in the setting node; if yes, the backup redundant setting module is started; if no, the node with the setting module in the cluster is requested to maintain the service through task migration; The node failure recovery strategy is: detecting the node status through a heartbeat mechanism; if a node fails, initiating a service recovery process in the space-based resource pool through a distributed intelligent satellite proxy strategy.

4. The method for optimizing space-based system service elasticity based on distributed agent self-healing according to claim 3 is characterized in that: In the process of service recovery through the space-based system elastic service recovery strategy, two mechanisms are set up for decision-making and control rights of the service recovery process; The first mechanism is that the decision and control of the service recovery process lies with the node that discovers the failure and the node that initiates the recovery process; The second mechanism is that the decision and control of the service recovery process lies with the preset decision satellite in the space-based resource pool.

5. The method for optimizing space-based system service elasticity based on distributed agent self-healing according to claim 4 is characterized in that: In the process of optimizing and executing the elastic service recovery strategy of the space-based system by using the set optimization algorithm, the optimization process includes a recovery consensus formation phase based on the network community and a service recovery execution phase; The steps of the network community-based recovery consensus formation phase are: The control node obtains the number of all failed nodes by counting and cyclically determines the specific failure status of each failed node; If the component of the failed node fails, it is determined whether there is a replacement resource component in the failed node. If so, it is replaced and the resource network is updated without affecting the task network. If not, searching for available resource components in the space-based resource pool; Starting from the failed node itself, the BSF algorithm is used to search for recoverable paths within the community. If there is a recoverable path, the recoverable path solution set is recorded. If not, the search is further expanded to the community. Through the inter-community search, if a recoverable path exists, the recoverable path solution set is recorded; if not, it is determined that the failed task cannot be recovered; Summarize the recoverable paths of failed nodes to obtain the recoverable path range; The steps of the service recovery execution phase are: Find missing links from the task network, and divide the missing links into a starting point group and an end point group; Convert system and mission networks into weighted patterns based on intersatellite link rates; The Edmond-Karp shortest path algorithm is used to find the minimum path between the starting point and the end point of the task network corresponding to the failed node to obtain the shortest path; Determine whether the path number of the shortest path is greater than 1; if so, map the shortest path to the resource network and calculate the elasticity value respectively; if not, continue to search for the path through the Edmond-Karp shortest path algorithm; Determine the next failed node; calculate the resilience of the space-based system based on the resource capabilities of the remaining nodes; The acquired path is converted into the configuration policy of the satellite node, output and executed, and the service is restored.

6. A space-based system service elasticity optimization device based on distributed agent self-healing, adopting a space-based system service elasticity optimization method based on distributed agent self-healing according to any one of claims 1 to 5, characterized in that: include: An application scenario building module is used to build application scenarios of space-based systems for user services based on the multifunctional services provided by the space-based systems; A two-layer super network model construction module is used to construct a two-layer super network model based on a task network and a resource network according to a mapping relationship between the task network and the resource network; A space-based system multi-scale elastic model construction module, used to construct a space-based system multi-scale elastic model based on the two-layer hypernetwork model and according to the application scenario; A space-based resource pool establishment module, used to establish a space-based resource pool based on a satellite cluster in the resource network according to the density of ground users and the satellite node resource loading conditions; A space-based system elastic service recovery strategy establishment module, used to establish a space-based system elastic service recovery strategy according to the space-based system multi-scale elastic model under the constraints of the space-based resource pool; The space-based system service elasticity optimization module is used to optimize and execute the space-based system elastic service recovery strategy by setting an optimization algorithm to complete the space-based system service elasticity optimization.

7. The space-based system service elasticity optimization device based on distributed agent self-healing according to claim 6 is characterized in that: In the multi-scale elastic model construction module of the space-based system, the elastic analysis granularity of the multi-scale elastic model of the space-based system is divided into module level, node level and system level; The elastic index expression of the multi-scale elastic model of the space-based system is: Where F is the elasticity index; D is the number of state variables of the space-based system; N is the number of nodes of the space-based system; and The resource status of the two satellite nodes that are interacting; It is the mutual information in information theory, which measures the influence of one satellite on another satellite.

8. The space-based system service elasticity optimization device based on distributed agent self-healing according to claim 7 is characterized in that: In the space-based system elastic service recovery strategy establishment module, the space-based system elastic service recovery strategy includes: module failure recovery strategy and node failure recovery strategy; The module failure recovery strategy is as follows: when the setting module on the setting node fails, the agent of the intelligent satellite determines whether there is a backup redundant setting module in the setting node; if yes, the backup redundant setting module is started; if no, the node with the setting module in the cluster is requested to maintain the service through task migration; The node failure recovery strategy is: detecting the node status through a heartbeat mechanism; if a node fails, initiating a service recovery process in the space-based resource pool through a distributed intelligent satellite proxy strategy.

9. The space-based system service elasticity optimization device based on distributed agent self-healing according to claim 8, characterized in that: In the space-based system elastic service recovery strategy establishment module, in the process of performing service recovery through the space-based system elastic service recovery strategy, two mechanisms are set for the decision-making and control rights of the service recovery process; The first mechanism is that the decision and control of the service recovery process lies with the node that discovers the failure and the node that initiates the recovery process; The second mechanism is that the decision and control of the service recovery process lies with the preset decision satellite in the space-based resource pool.

10. The space-based system service elasticity optimization device based on distributed agent self-healing according to claim 9, characterized in that: In the space-based system service resilience optimization module, the optimization process includes a recovery consensus formation phase based on the network community and a service recovery execution phase; The optimization submodules of the network community-based recovery consensus formation phase include: The failure node data and failure situation statistics submodule is used to control the node statistics to obtain the number of all failed nodes, and cyclically determine the specific failure situation of each failed node; The failed node judgment processing submodule is used to judge whether there is a replacement resource component in the failed node if the component of the failed node fails, and if so, replace it and update the resource network without affecting the task network; if not, search for available resource components in the space-based resource pool; The community search submodule is used to search for recoverable paths within the community using the BSF algorithm starting from the failed node itself; if there is a recoverable path, the recoverable path solution set is recorded; if not, the search is further expanded to the inter-community search; The inter-community search submodule is used to record the recoverable path solution set if a recoverable path exists through the inter-community search; if not, it is determined that the failed task cannot be recovered; The recoverable path range acquisition submodule is used to summarize the recoverable paths of failed nodes and obtain the recoverable path range; The submodules of the service recovery execution phase include: A missing link search and grouping submodule is used to find missing links from the task network and divide the missing links into a starting point group and an end point group; A weighted pattern conversion submodule is used to convert the system and mission network into a weighted pattern according to the intersatellite link rate; The shortest path acquisition submodule is used to find the minimum path between the starting point and the end point of the task network corresponding to the failed node through the Edmond-Karp shortest path algorithm to obtain the shortest path; The shortest path number determination submodule is used to determine whether the number of paths of the shortest path is greater than 1; if so, the shortest path is mapped to the resource network and the elasticity value is calculated respectively; if not, the path is continued to be searched through the Edmond-Karp shortest path algorithm; The failed node cycle judgment submodule is used to judge the next failed node; the elasticity of the space-based system is calculated according to the resource capacity of the remaining nodes; The configuration strategy output and execution submodule is used to convert the acquired path into the configuration strategy of the satellite node, output and execute it, and restore the service.

Citation Information

Cited By

  • Efficient agent pool management method based on dynamic selection and intelligent optimization

    CN120750935A