Unmanned cluster system reliability digital twin model evolution method
By constructing agents and screening evolutionary elements, capturing evolutionary opportunities and formulating strategies, the gap in reliability of unmanned cluster systems is solved, and the orderly evolution update of the reliability of unmanned cluster systems is achieved, ensuring the system stability and evolution support.
Patent Information
- Application Number
- CN202411178078.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-26
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-26
AI Technical Summary
There is a gap in reliability for existing unmanned cluster systems, mainly focusing on functional performance descriptions but not achieving balance, and lack of effective theoretical methods to ensure the reliability and stability of the system.
A digital twin evolution method for the reliability of unmanned cluster systems is proposed. By constructing node agents, sub-system agents and component agents, combining system failure occurrence and propagation, screening evolutionary elements, capturing evolutionary timing, and based on agent dynamics The knowledge base develops evolutionary strategies, including correction of attribute parameters, adjusting interactions, and increasing or decreasing node agents.
The orderly evolution and update of the digital twin model of the unmanned cluster system has been realized, which enhances the reliability of the system and ensures its stable operation and evolution support under multi-level characteristics.
Smart Images

Figure CN118917103A_ABST
Abstract
Description
Technical Field
[0001] The present invention provides an evolution method of a reliability digital twin model of an unmanned cluster system, and belongs to the technical field of reliability engineering. Background Art
[0002] At present, thanks to the rapid development of sensor networks, distributed communications and big data technologies, the application characteristics of digital twins have gradually transitioned to multi-level and multi-factor. The digital twin of the unmanned cluster system is a technical means to create a virtual simulation model of the cluster physical entity in a digital way, and simulate, verify, predict and control the entire life cycle process of the physical entity with the help of historical monitoring information, real-time operation data and algorithm models. Its basic workflow during operation can be divided into four parts: information collection, state perception, operation and maintenance decision-making, and implementation. From a virtual perspective, the realization of state perception and operation and maintenance decision-making is usually based on digital twins, and its essence is the aggregation of a series of models and model evolution processes. At the same time, the content involved in cluster state perception includes functional performance, reliability and other aspects. Among them, reliability is to capture the overall and local degradation fault information of the system through the model to achieve the processing and display of relevant data.
[0003] However, the relevant research on the digital twin of the unmanned cluster system has not yet reached a balance at this stage, and most of the work only focuses on describing the functional performance of the system and its evolution process. In order to achieve accurate operation and maintenance of the unmanned cluster system and ensure its reliable and stable operation, it is necessary to propose corresponding theoretical methods on the existing basis to make up for the gap in the field of reliability digital twins. To this end, the present invention comprehensively considers the typical characteristics of unmanned cluster systems such as distributed deployment, reconfiguration, and autonomous collaboration, and invents an evolution method for the reliability digital twin model of unmanned cluster systems, which can provide technical support for the development and application of digital twin technology in the field of reliability of unmanned cluster systems. Summary of the invention
[0004] The purpose of the present invention is to provide an evolution method for the digital twin model of the reliability of unmanned cluster systems, so as to effectively integrate the evolution elements, evolution timing and evolution strategy of the digital twin model of the reliability of unmanned cluster systems, realize the evolution and update of the digital twin model, and further support the construction of the digital twin of the unmanned cluster system.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] A reliability digital twin model evolution method for unmanned cluster systems mainly includes the following steps:
[0007] S100: Constructing node agents, subsystem agents and component agents in sequence to describe the components and functional principles of the unmanned swarm system;
[0008] S200: Focusing on the occurrence and propagation of system failures, screening the evolution factors of the reliability digital twin model of unmanned swarm systems;
[0009] S300: Compare model simulation data with real-time monitoring information to capture the evolution timing of the reliability digital twin model of the unmanned swarm system;
[0010] S400: Based on the dynamic knowledge base of intelligent agents, formulate the evolution strategy of the reliability digital twin model of unmanned cluster system:
[0011] S401: Modify the agent attribute parameters;
[0012] S402: Adjusting the interaction relationship between intelligent agents;
[0013] S403: Add or delete node agents.
[0014] In step S100, based on the multi-level characteristics of the unmanned cluster system, node agents, subsystem agents and component agents are constructed from top to bottom to describe the components and functional principles of the system. Among them, the node agent is used to simulate the operation mode of a single system; the subsystem agent is used to abstract the key subsystems of the node; and the component agent is used to characterize the functional differences of the components contained in the subsystem. On this basis, a dynamic knowledge base is configured for each agent in turn to store the monitored node operation mode, as well as the subsystem and component failure degradation information, and to carry out knowledge generation and updating to support model evolution;
[0015] In step S200, the potential evolution factors of the reliability digital twin model of the unmanned cluster system are determined from the perspectives of model parameter update, interaction relationship adjustment, and agent increase and decrease, centering on the occurrence of component failures and the propagation of failures inside and outside the node. On this basis, the focus is on the parameter correction of the component failure model, the adaptive adjustment of the system failure propagation path, and the online increase and decrease of node agents from the inside out, and the evolution factors are selected in a targeted manner according to the specific mission scenarios of the unmanned cluster system;
[0016] In step S300, on the basis of determining the evolution factors, the real-time monitoring information and the model simulation data are dynamically compared to determine the evolution timing of the reliability digital twin model of the unmanned cluster system. For the parameter correction of the component fault model, the relative percentage error is used to measure the difference between the model simulation data and the actual monitoring data, and the error threshold is used to trigger the update of the intelligent agent attribute parameters; for the adaptive adjustment of the system fault propagation path, the change of the intelligent agent interaction relationship is triggered according to the difference before and after the actual node operation mode; for the online increase and decrease of the node intelligent agent, the increase and deletion of the node intelligent agent is triggered based on the activation of the actual node;
[0017] In step S400, based on the agent dynamic knowledge base and integrating the evolution opportunity, a reliability digital twin model evolution strategy is formulated for the corresponding unmanned cluster system evolution elements:
[0018] In step S401, the relative percentage error measurement result is referred to to trigger the component fault model parameter correction strategy. Based on the Bayesian update theory, the component agent attribute parameters are corrected using the measured component degradation fault information;
[0019] In step S402, the adaptive adjustment strategy of the fault propagation path is triggered according to the difference before and after the actual node operation mode. After the default node operation mode is adjusted, knowledge update must be carried out synchronously. First, the node agent dynamic knowledge base is searched to see whether the operation mode has been stored. If it exists, the agent interaction relationship is adjusted based on the associated relevant knowledge. If it does not exist, the node operation mode related knowledge is generated and updated, and the agent interaction relationship is adjusted based on the updated knowledge, thereby realizing the adaptive adjustment of the fault propagation path.
[0020] In step S403, the online increase and decrease strategy of the node agent is triggered according to the difference before and after the actual node activation state. When it is detected that the actual node is out of the unmanned cluster system, the corresponding node agent and the associated subsystem agent and component agent are adjusted to a silent state, and according to the hot and cold backup status of the removed node, it is determined whether the contained component fault model continues to run with the simulation time. On the contrary, when the node joins the unmanned cluster system, the relevant agent is activated and the knowledge base is copied. The maximum set of node operation modes in the current multi-agent model knowledge base is marked, and the marked knowledge base is copied to the activated node agent and the associated subsystem agent and component agent in turn, and finally the interaction relationship of the agent is updated.
[0021] Compared with the existing model evolution methods, the beneficial effects brought by the present invention are: with the help of the unmanned cluster system reliability digital twin model evolution method developed by the present invention, the model evolution elements can be targetedly selected according to the operating characteristics of the actual unmanned cluster system, and the model evolution timing can be captured based on the real-time monitoring data of the system, and the corresponding reliability digital twin model evolution strategy can be formulated using the dynamic knowledge base, so as to achieve an effective integration of the evolution elements, evolution timing and evolution strategy of the unmanned cluster system reliability digital twin model, realize the orderly evolution and update of the model, and support the construction of the digital twin of the unmanned cluster system. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 A flow chart of an evolution method of a reliability digital twin model of an unmanned cluster system provided by the present invention;
[0023] Figure 2 The UAV cluster reliability digital twin model provided by the present invention. DETAILED DESCRIPTION
[0024] The following will refer to the attached Figure 1 With attached Figure 2 Specific embodiments of the present invention are described in detail. Although specific embodiments of the present invention are discussed, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, the embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the information of the present invention to those skilled in the relevant art.
[0025] The embodiment of the present invention provides a method for evolving a reliability digital twin model of a drone cluster. The flow chart is as follows: Figure 1 As shown, including:
[0026] S100: Based on the multi-level characteristics of the drone cluster, node agents, subsystem agents and component agents are constructed from top to bottom to describe the components and functional principles of the system. Among them, the node agent is used to simulate the operation mode of a single drone; the subsystem agent is used to abstract the key subsystems of a single drone; the component agent is used to characterize the differences in the functions of the components contained in the drone. On this basis, a dynamic knowledge base is configured for each agent to store the monitored drone operation mode, as well as the fault degradation data of the subsystems and components, and to carry out knowledge generation and updating to support model evolution;
[0027] Example 1: Figure 2 The UAV cluster shown in the figure consists of 3 UAVs. The subsystems of each UAV include the fuselage, attitude control module, collaborative control module, power module and power module. The power module can be further disassembled into electric speed controller, motor and propeller blades. Based on this, the reliability digital twin model of the UAV cluster is constructed. Among them, the node agent is used to simulate the formation flight mode and manual control mode of a single UAV; and 5 subsystem agents are constructed in the node agent to abstract the key subsystems contained in the UAV; further, 3 component agents are constructed in the power module-subsystem agent to characterize the functional differences of the component electric speed controller, motor and propeller blade. On this basis, a dynamic knowledge base is configured for each agent to store the monitored UAV operation mode, as well as the fault degradation data of the subsystems and components, and to generate and update knowledge to support model evolution;
[0028] S200: Focusing on the occurrence of component failures and the propagation of faults inside and outside nodes, we determine the potential evolutionary factors of the UAV cluster reliability digital twin model from the perspectives of model parameter update, interaction relationship adjustment, and agent increase and decrease. On this basis, we focus on the parameter correction of the component failure model, the adaptive adjustment of the system fault propagation path, and the online increase and decrease of node agents from the inside out, and select evolutionary factors in a targeted manner according to the specific UAV cluster mission scenario;
[0029] Example 1: The drone cluster is used to perform joint strike missions. During the formation flight, penetration, attack, and return process, the drone may be shot down by enemy air defense fire. Then, based on the above mission characteristics, from the perspective of agent increase and decrease, the online increase and decrease of node agents is selected as the evolution factor of the drone cluster;
[0030] S300: Based on the determination of evolutionary factors, dynamically compare real-time monitoring information with model simulation data to determine the evolution timing of the UAV cluster reliability digital twin model. For parameter correction of component failure models, use relative percentage error to measure the difference between model simulation data and actual monitoring data, and refer to the error threshold to trigger the update of agent attribute parameters; for adaptive adjustment of system fault propagation path, trigger the change of agent interaction relationship based on the difference before and after the actual node operation mode; for online increase and decrease of node agents, trigger the addition and deletion of node agents based on the activation of actual nodes;
[0031] Continuing from the previous example: The evolutionary elements of the drone cluster are the online addition and reduction of node agents, and the real-time monitoring information and model simulation data are dynamically compared to determine the evolution timing of the drone cluster reliability digital twin model. For the online addition and reduction of node agents, the addition and deletion of node agents are triggered based on whether the actual drone is activated or not;
[0032] S400: Based on the dynamic knowledge base of intelligent agents and integrating evolutionary opportunities, a reliability digital twin model evolution strategy is formulated for the evolutionary elements of drone clusters:
[0033] S401: Referring to the relative percentage error measurement result, trigger the component fault model parameter correction strategy. Based on the Bayesian update theory, use the measured component degradation fault information to correct the component agent attribute parameters;
[0034] S402: Trigger the adaptive adjustment strategy of the fault propagation path based on the difference before and after the actual UAV operation mode. After the default UAV operation mode is adjusted, knowledge update must be carried out synchronously. First, the node agent dynamic knowledge base is retrieved to see if the operation mode has been stored: if it exists, the agent interaction relationship is adjusted based on the associated relevant knowledge; if it does not exist, the UAV operation mode related knowledge is generated and updated, and the agent interaction relationship is adjusted based on the updated knowledge, thereby realizing the adaptive adjustment of the fault propagation path;
[0035] S403: Trigger the online increase and decrease strategy of the node agent based on the difference before and after the actual UAV activation status. When the actual UAV is detected to be out of the UAV cluster, the corresponding node agent and the associated subsystem agent and component agent are adjusted to a silent state, and according to the hot and cold backup status of the removed node, it is determined whether the fault model of the contained component continues to run over the simulation time. On the contrary, when the UAV joins the UAV cluster, the relevant agent is activated and the knowledge base is copied. Mark the maximum set of node operation modes in the current multi-agent model knowledge base, and copy the marked knowledge base to the activated node agent and the associated subsystem agent and component agent in turn, and finally update the interaction relationship between the agents.
[0036] Continuing from the previous example: In the process of performing a joint strike mission, the actual drone swarm detected that a drone was shot down by enemy air defense fire during the return flight, and the drone jumped from the activated state to the inactivated state, thereby triggering the model evolution. Mark this moment, adjust the corresponding node agent and the associated subsystem agent and component agent to the silent state, and update the interaction relationship of the agent. In addition, considering that the drone is in a cold backup state because it was completely shot down, the component failure model contained in it will no longer run continuously with the simulation time.
[0037] Although the embodiments of the present invention are described above in conjunction with the accompanying drawings, the present invention is not limited to the above specific embodiments and application fields, and the above specific embodiments are only illustrative and instructive, rather than restrictive. A person of ordinary skill in the art can also make many forms under the guidance of this specification and without departing from the scope of protection of the claims of the present invention, all of which belong to the protection of the present invention.
Claims
1. A reliability digital twin model evolution method for unmanned cluster systems, characterized in that: include: S100: Constructing node agents, subsystem agents and component agents in sequence to describe the components and functional principles of the unmanned swarm system; S200: Focusing on the occurrence and propagation of system failures, screen the evolution factors of the reliability digital twin model of the unmanned cluster system; determine the potential evolution factors of the reliability digital twin model of the unmanned cluster system from the perspectives of model parameter update, interaction relationship adjustment, and increase and decrease of intelligent agents; on this basis, focus on the parameter correction of the component failure model, the adaptive adjustment of the system failure propagation path, and the online increase and decrease of node intelligent agents from the inside out, and select the evolution factors in a targeted manner according to the specific mission scenarios of the unmanned cluster system; S300: Compare model simulation data with real-time monitoring information to capture the evolution timing of the reliability digital twin model of the unmanned cluster system; for parameter correction of the component failure model, use relative percentage error to measure the difference between model simulation data and actual monitoring data, and refer to the error threshold to trigger the update of the intelligent agent attribute parameters; Adaptive adjustment of the system fault propagation path triggers the change of the agent interaction relationship based on the difference before and after the actual node operation mode; for the online increase and decrease of node agents, the increase and deletion of node agents is triggered based on the activation of the actual node; S400: Based on the dynamic knowledge base of intelligent agents, formulate the evolution strategy of the reliability digital twin model of unmanned cluster system: S401: Correcting the agent attribute parameters; S402: Adjusting the interaction relationship between intelligent agents; S403: Add or delete node agents.
2. According to claim 1, the method for evolving a reliability digital twin model of an unmanned cluster system is characterized by: In the "adjusting the interaction relationship between intelligent agents" described in S402, the fault propagation path adaptive adjustment strategy is triggered according to the difference before and after the actual node operation mode; After the default node operation mode is adjusted, knowledge update needs to be carried out synchronously. First, the node agent dynamic knowledge base is searched to see whether the operation mode has been stored. If it exists, the agent interaction relationship is adjusted based on the associated relevant knowledge. If it does not exist, the knowledge related to the node operation mode is generated and updated, and the interaction relationship between the intelligent agents is adjusted based on the updated knowledge, thereby realizing adaptive adjustment of the fault propagation path.
3. The method for evolving a reliability digital twin model of an unmanned cluster system according to claim 1, characterized in that: In the "addition and deletion of node agents" described in S403, the node agent online addition and reduction strategy is triggered according to the difference before and after the actual node activation state; when it is detected that the actual node is out of the unmanned cluster system, the corresponding node agent and the associated subsystem agent and component agent are adjusted to a silent state, and according to the hot and cold backup states of the removed nodes, it is determined whether the fault model of the contained components continues to run with the simulation time; On the contrary, when a node joins the unmanned cluster system, it activates the relevant agents and copies the knowledge base; Mark the maximum set of node operation modes in the current multi-agent model knowledge base, and copy the marked knowledge base to the activated node agents and the associated subsystem agents and component agents in turn, and finally update the interaction relationship between the agents.
Citation Information
Patent Citations
Self-similarity modeling and self-organization layered aggregation method based on CPS manufacturing assembly and unmanned production line self-similarity fractal reconstruction system
CN113836824A
Simulation deduction system based on digital twinning
CN117150757A
Systems and methods for the digital verification of industrial construction execution
US20190138667A1
