On-orbit health assessment method and terminal based on time-varying topology structure

Abstract

The application discloses a kind of constellation on-orbit health evaluation method and terminal based on time-varying topology structure, steps include: satellite health state evaluation in constellation;Inter-satellite communication link health evaluation;Constellation task efficiency evaluation;Calculate constellation health degree value to judge health grade.The application obtains constellation health state information and constellation health degree data, can determine the degree of influence of each level fault on constellation health, so that targeted maintenance strategy can be specified, with the advantages of objective and comprehensive, high reliability, strong robustness and the like.

Description

Technical Field

[0001] This invention belongs to the field of spacecraft health management applications, specifically referring to a method and terminal for on-orbit health assessment of star clusters based on time-varying topology. Background Technology

[0002] Satellite constellations (or simply constellations) offer advantages over individual microsatellites in terms of operational efficiency, flexibility, and system robustness, and have become one of the trends in future space technology development. However, with the dramatic increase in the size of constellations in orbit, their on-orbit operational stability, safety, and reliability are all challenged. Therefore, the need for health management technologies for satellite constellations is urgent, and health status assessment technology is a crucial prerequisite for this technology.

[0003] Generally, satellite constellations consist of multiple satellites working together. The failure of a few satellites does not necessarily mean the failure of the entire constellation. Furthermore, the complex structure of constellation systems means that numerous factors influence their health status, and limited onboard resources cannot handle complex calculations. These constraints make satellite constellation health assessment a challenge in the field of health management. However, traditional health assessment techniques are generally applied to spacecraft attitude control systems. Their basic principle relies on mathematical models of the attitude control system and expert knowledge for evaluation; or on large amounts of real data from existing spacecraft, using intelligent algorithms such as neural networks, Bayesian networks, and support vector machines for extensive simulation and evaluation. These two methods clearly cannot meet the health assessment needs of satellite constellations. Although satellite constellation health management is currently a research hotspot in predictive and health management technologies, there are currently no established examples of constellation health assessment models, either domestically or internationally. Summary of the Invention

[0004] To address the shortcomings of the existing technologies, the present invention aims to provide a method and terminal for on-orbit health assessment of satellite constellations based on time-varying topology. The method of the present invention acquires satellite constellation health status information and satellite constellation health data, and can determine the degree of impact of various levels of faults (different numbers of satellite faults, different numbers of communication link faults) on satellite constellation health, thereby enabling targeted specification of maintenance strategies. It has advantages such as objectivity, comprehensiveness, high reliability, and strong robustness.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] The present invention provides a method for on-orbit health assessment of star clusters based on time-varying topology, comprising the following steps:

[0007] (1) Satellite health status assessment in the constellation: Establish a satellite reliability model and use the method of topological node importance to describe the correlation between satellites and constellation health, thereby assessing the impact of all satellites on constellation health;

[0008] (2) Health assessment of inter-satellite communication links: The importance weight of links in the topology is confirmed, the evaluation index is natural connectivity, and the reliability of inter-satellite links is calculated by reliability analysis, thereby assessing the impact of all inter-satellite links on the health of the constellation.

[0009] (3) Evaluation of the effectiveness of the constellation mission: Based on the error terms of the three-axis position, three-axis attitude and energy consumption of each satellite in the constellation, the entropy weight method is used to calculate the weight of each index, thereby calculating the mission effectiveness;

[0010] (4) Calculate the health value of the satellite cluster: After the satellite fault is completed and the weight is changed, the health value of the satellite cluster is calculated based on the evaluation results obtained in steps (1) to (3) to determine the health level.

[0011] Furthermore, the health status is divided into three levels: Health (H), Sub-Health (SH), and Fault (F).

[0012] Furthermore, in the health level, the value of healthy (H) is [0.8000, 1.0000], the value of sub-healthy (SH) is [0.4500, 0.7999], and the value of faulty (F) is [0.0000, 0.4499].

[0013] Furthermore, step (1) specifically includes:

[0014] (11) During the satellite's lifetime T, the Weibull distribution was used respectively. and normal distribution Fault simulation is performed, where t is the evaluation time; η is the scale parameter, representing the characteristic lifetime; β is the shape parameter; μ is the expected value; σ is the standard deviation; and e is the natural exponent. The resulting satellite reliability model is:

[0015]

[0016] In the formula, R vi Indicates satellite v in the constellation i The reliability values, i = 1, 2, ..., N;

[0017] (12) Assume that at time t, the constellation contains N satellites and M communication links. Let the number of satellites v be V. i and satellite v j The shortest path between them is d ij Then the satellite density can be obtained.

[0018] (13) Combining the ratio of the number of corresponding satellites in the shortest path of all satellite pairs in the constellation topology, i.e., the satellite betweenness number Solve for satellite vi Importance is After normalization, the weight of the satellite's importance in the constellation is: Among them, g jk For satellite v j With satellite v k The number of shortest paths, g jk (i) is g jk via satellite v i Quantity, It is satellite v i For satellite v j With satellite v k The contribution of betweenness reflects satellite position information;

[0019] (14) Combining the satellite reliability model obtained in step (11) and the satellite importance weight values ​​in the constellation obtained in step (13), output the health status of all satellites in the constellation:

[0020]

[0021] Furthermore, step (2) specifically includes:

[0022] (21) Analyze whether there are communication links between satellites and construct the inter-satellite correlation matrix A(G)={a ij} N×N That is, if satellite v i and satellite v j There is a link between them, a ij =1; if satellite v i and satellite v j There is no link between them, a ij =0; the natural connectivity is the average of all eigenvalues ​​of the interstellar correlation matrix A(G) with respect to the natural logarithm and natural exponent.

[0023]

[0024] In the formula, λ p It is the p-th eigenvalue of A(G), and its natural connectivity is... The larger the value, the more robust the connection.

[0025] (22) When link e j After failure, the recalculated natural connectivity is used If expressed as such, then the importance of the link is... Normalization can yield the link importance weight values.

[0026] (23) During link analysis, it is assumed that satellites are visible to each other. Only the reliability of communication link transmission needs to be considered. The reliability of inter-satellite links is expressed as:

[0027]

[0028] In the formula, λ ej Indicates link e j The failure rate, j = 1, 2, ..., M; combining the inter-satellite link reliability and the link importance weight value obtained in step (22), the health of the inter-satellite communication link is obtained as follows:

[0029]

[0030] Furthermore, step (3) specifically includes:

[0031] (31) Compare the ideal three-axis position, three-axis attitude and energy consumption of the satellite during its motion with their actual values, select the obtained error as the secondary index, and the satellites in the constellation as the primary index. Based on the analytic hierarchy process, establish the system structure of the performance evaluation index; where the constellation performance is the highest layer of the structure, the primary index is the middle layer of the structure, and the secondary index is the lowest layer of the structure. Draw the hierarchical structure diagram.

[0032] (32) Based on the performance evaluation index system structure, the weight of each index is determined using the entropy weight method; assuming that the error of each secondary index is E ij The normalized value is z ij The weight value of the secondary indicator is The weights of the primary indicators can be obtained from the following formula:

[0033]

[0034] z i =1-r i

[0035]

[0036] In the formula, r i This represents the information entropy, or uncertainty, of each secondary indicator; z i This represents the information utility value; ω i The weight values ​​for the primary indicators;

[0037] (33) Combining the performance evaluation index system model obtained in step (31) and the index parameters obtained in step (32), output the overall performance of the star cluster.

[0038] Furthermore, step (4) specifically includes:

[0039] (41) Assume X = (x1)r ,…,x k r ) is factor x i The state vector of (i = 1, ..., k), W = (ω1, ..., ωk) k ) is factor x i A constant weight vector S(X) = (S1(X), ..., Sk) is formed by the constant weight vectors (i = 1, ..., k). k If (X) is the state-variable weight vector, then factor x i The variable weight vectors (i = 1, ..., k) are obtained by the Hadamard product, i.e.:

[0040]

[0041] In the formula, The weights are determined based on the actual situation. When it is determined that a momentary fault has occurred in the star cluster, the weight values ​​ω of steps (13), (22) and (32) are adjusted according to the different faults using the above formula (1) to achieve variable weights.

[0042] (42) After the weight transfer is completed, the current health H of the constellation is taken into account. swarm (t) is affected by the health status of the previous time step, i.e., there is a temporal dependency. Using a nonlinear weighting method, the cluster health function is obtained:

[0043]

[0044] In the formula, σ is the time series function, and k1, k2, k3 are the influencing parameters;

[0045] (43) By using the first exponential smoothing method, the health value of the star cluster is fitted to estimate σ=1-α,α∈[0,1] and k1+k2+k3=α; according to the star cluster health function in step (41), the health of the star cluster at time t in the life cycle is calculated, and the corresponding health level is obtained.

[0046] The health assessment method of this invention fully considers the impact of inter-satellite communication and cluster performance on the health status of a satellite constellation. Based on the time-varying topology of the constellation, a multi-level health assessment model is established. By calculating the health of satellites, inter-satellite communication, and cluster performance separately, and combining this with the variable weighting principle, the health status of the constellation can be assessed. Applying this method to modular microsatellite constellations performing autonomous stitching missions can provide the prerequisites for constellation health management and improve on-orbit operational reliability.

[0047] The present invention also provides a health assessment terminal, comprising:

[0048] One or more processors;

[0049] Memory, used to store one or more programs;

[0050] When the one or more programs are executed by the one or more processors, the one or more processors implement the methods described above.

[0051] The beneficial effects of this invention are:

[0052] 1. This invention defines the health of a satellite cluster from the perspectives of satellites, multiple satellites, and multiple satellite groups. By establishing a health assessment model that integrates satellites, inter-satellite communication links, and satellite cluster mission performance, it fully considers the impact of inter-satellite communication and cluster performance on the health status of the satellite cluster.

[0053] 2. Based on time-varying topology networks, this invention demonstrates the evolution of satellite cluster health over time. By understanding the satellite cluster's operational lifespan through the health curve, the health status of the cluster can be determined, making cluster health assessment practically valuable.

[0054] 3. The introduction of communication links in this invention reduces the subjectivity of the assessment; based on the health assessment results, the degree of impact of various levels of failure (different numbers of satellite failures, different numbers of communication link failures) on the constellation's health can be determined, thus allowing for targeted maintenance strategies. In future space maintenance management, this technology will greatly enhance the intelligence and safety of spacecraft constellations. Attached Figure Description

[0055] Figure 1 This is a flowchart of the evaluation method of the present invention.

[0056] Figure 2 This is a flowchart of a satellite health assessment process based on reliability analysis.

[0057] Figure 3 This is a flowchart for the health assessment of inter-satellite communication links based on natural connectivity.

[0058] Figure 4 This is a flowchart of the star cluster performance evaluation based on the entropy weight method.

[0059] Figure 5 This is a flowchart of a non-linear weighted comprehensive evaluation process. Detailed Implementation

[0060] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0061] In this invention, satellites: a satellite constellation mainly consists of multiple satellites, which act as nodes in the constellation topology. Inter-satellite communication links: satellites in the constellation maintain communication connections to work collaboratively, acting as links in the constellation topology. Constellation mission effectiveness: a constellation performs specific tasks based on the specific payloads of its satellites. Constellations that fail to meet mission requirements are considered unhealthy; therefore, constellation mission effectiveness is also an important factor in assessing constellation health.

[0062] Reference Figure 1 As shown, the present invention provides a method for on-orbit health assessment of star clusters based on time-varying topology, comprising the following steps:

[0063] (1) Satellite health status assessment in the constellation: A satellite reliability model is established using reliability analysis, and the correlation between satellites and constellation health is described using the method of topological node importance, thereby assessing the impact of all satellites on constellation health;

[0064] The health status is divided into three levels: Health (H), Sub-Health (SH), and Fault (F).

[0065] In the health level classification, the values ​​for healthy (H) are [0.8000, 1.0000], sub-healthy (SH) are [0.4500, 0.7999], and faulty (F) are [0.0000, 0.4499].

[0066] Specifically, such as Figure 2 As shown, step (1) specifically includes:

[0067] (11) During the satellite's lifetime T, the Weibull distribution was used respectively. and normal distribution Fault simulation is performed, where t is the evaluation time; η is the scale parameter, representing the characteristic lifetime; β is the shape parameter; μ is the expected value; σ is the standard deviation; and e is the natural exponent. The resulting satellite reliability model is:

[0068]

[0069] In the formula, R vi Indicates satellite v in the constellation i The reliability values, i = 1, 2, ..., N;

[0070] (12) Assume that at time t, the constellation contains N satellites and M communication links. Let the number of satellites v be V. i and satellite v j The shortest path between them is d ij Then the satellite density can be obtained.

[0071] (13) Combining the ratio of the number of corresponding satellites in the shortest path of all satellite pairs in the constellation topology, i.e., the satellite betweenness number Solve for satellite v i Importance is After normalization, the weight of the satellite's importance in the constellation is: Among them, g jk For satellite v j With satellite v k The number of shortest paths, g jk (i) is g jk via satellite v i Quantity, It is satellite v i For satellite v j With satellite v k The contribution of betweenness reflects satellite position information;

[0072] (14) Combining the satellite reliability model obtained in step (11) and the satellite importance weight values ​​in the constellation obtained in step (13), output the health status of all satellites in the constellation:

[0073]

[0074] (2) Health assessment of inter-satellite communication links: The weight values ​​of the importance of links in the topology are determined by a resilience-based approach. The evaluation index is natural connectivity (which reflects the redundancy of inter-satellite alternative paths and indicates the degree to which inter-satellite connections can still be established after a link failure). Reliability analysis is used to calculate the reliability of inter-satellite links, thereby assessing the impact of all inter-satellite links on the health of the constellation.

[0075] Among them, such as Figure 3 As shown, step (2) specifically includes:

[0076] (21) Analyze whether there are communication links between satellites and construct the inter-satellite correlation matrix A(G)={a ij} N×N That is, if satellite v i and satellite v j There is a link between them, a ij =1; if satellite v i and satellite v j There is no link between them, a ij =0; the natural connectivity is the average of all eigenvalues ​​of the interstellar correlation matrix A(G) with respect to the natural logarithm and natural exponent.

[0077]

[0078] In the formula, λ pIt is the p-th eigenvalue of A(G), and its natural connectivity is... The larger the value, the more robust the connection.

[0079] (22) When link e j After failure, the recalculated natural connectivity is used If expressed as such, then the importance of the link is... Normalization can yield the link importance weight values.

[0080] (23) During link analysis, it is assumed that satellites are visible to each other. Only the reliability of communication link transmission needs to be considered. The reliability of inter-satellite links is expressed as:

[0081]

[0082] In the formula, λ ej Indicates link e j The failure rate, j = 1, 2, ..., M; combining the inter-satellite link reliability and the link importance weight value obtained in step (22), the health of the inter-satellite communication link is obtained as follows:

[0083]

[0084] (3) Evaluation of the effectiveness of the constellation mission: Based on the error terms of the three-axis position, three-axis attitude and energy consumption of each satellite in the constellation, the entropy weight method is used to calculate the weight of each index, thereby calculating the mission effectiveness;

[0085] Among them, such as Figure 4 As shown, step (3) specifically includes:

[0086] (31) Compare the ideal three-axis position, three-axis attitude and energy consumption of the satellite during its motion with their actual values, select the obtained error as the secondary index, and the satellites in the constellation as the primary index. Based on the analytic hierarchy process, establish the system structure of the performance evaluation index; where constellation performance is the highest layer of the structure, the primary index is the middle layer of the structure, and the secondary index is the lowest layer of the structure. Draw the hierarchical structure diagram.

[0087] (32) Based on the performance evaluation index system structure, the weight of each index is determined using the entropy weight method; assuming that the error of each secondary index is E ij The normalized value is z ij The weight value of the secondary indicator is The weights of the primary indicators can be obtained from the following formula:

[0088]

[0089] z i =1-r i

[0090]

[0091] In the formula, r i This represents the information entropy, or uncertainty, of each secondary indicator; z i This represents the information utility value; ω i The weight values ​​for the primary indicators;

[0092] (33) Combining the performance evaluation index system model obtained in step (31) and the index parameters obtained in step (32), output the overall performance of the star cluster.

[0093] (4) Calculate the health value of the satellite cluster: After the satellite fault is completed and the weight is changed, the health value of the satellite cluster is calculated based on the evaluation results obtained in steps (1) to (3) to determine the health level.

[0094] Among them, such as Figure 5 As shown, step (4) specifically includes:

[0095] (41) Since the health status of each satellite in the constellation is dynamic, when a satellite experiences a transient failure, the effect of low-weight factors often becomes significant. Therefore, the importance weight values ​​need to be adjusted according to different failures; assuming X = (x1 r ,…,x k r ) is factor x i (the state vectors of i = 1, ..., k, W = (ω1, ..., ω) k ) is factor x i (i = 1, ..., k constant weight vectors, S(X) = (S1(X), ..., S k If (X) is the state-variable weight vector, then factor x i The variable weight vectors (i = 1, ..., k) are obtained by the Hadamard product, i.e.:

[0096]

[0097] In the formula, The weights are determined based on the actual situation. When it is determined that a momentary fault has occurred in the star cluster, the weight values ​​ω of steps (13), (22) and (32) are adjusted according to the different faults using the above formula (1) to achieve variable weights.

[0098] (42) After the weight transfer is completed, the current health H of the constellation is taken into account. swarm (t) is affected by the health status of the previous time step, i.e., there is a temporal dependency. Using a nonlinear weighting method, the cluster health function is obtained:

[0099]

[0100] In the formula, σ is the time series function, and k1, k2, k3 are the influencing parameters;

[0101] (43) By using the first exponential smoothing method, the health value of the star cluster is fitted to estimate σ=1-α,α∈[0,1] and k1+k2+k3=α; according to the star cluster health function in step (41), the health of the star cluster at time t in the life cycle is calculated, and the corresponding health level is obtained.

[0102] In addition, the present invention also provides a health assessment terminal, comprising:

[0103] One or more processors;

[0104] Memory, used to store one or more programs;

[0105] When the one or more programs are executed by the one or more processors, the one or more processors implement the methods described above.

[0106] This invention has many specific applications. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A method for on-orbit health assessment of star clusters based on time-varying topology, characterized by the following steps: include: (1) Satellite health status assessment in the constellation: Establish a satellite reliability model and use the method of topological node importance to describe the correlation between satellites and constellation health, thereby assessing the impact of all satellites on constellation health; (2) Health assessment of inter-satellite communication links: The importance weight of links in the topology is confirmed, the evaluation index is natural connectivity, and the reliability of inter-satellite links is calculated by reliability analysis, thereby assessing the impact of all inter-satellite links on the health of the constellation. (3) Evaluation of the effectiveness of the constellation mission: Based on the error terms of the three-axis position, three-axis attitude and energy consumption of each satellite in the constellation, the entropy weight method is used to calculate the weight of each index, thereby calculating the mission effectiveness; (4) Calculate the health value of the satellite cluster: After the satellite fault is completed and the weight is changed, the health value of the satellite cluster is calculated based on the evaluation results obtained in steps (1) to (3) to determine the health level. Step (1) specifically includes: (11) During the satellite's lifetime T, the Weibull distribution was used respectively. and normal distribution Fault simulation is performed, where t is the evaluation time; η is the scale parameter, representing the characteristic lifetime; β is the shape parameter; μ is the expected value; σ is the standard deviation; and e is the natural exponent. The resulting satellite reliability model is: In the formula, R vi Indicates satellite v in the constellation i The reliability values, i = 1, 2, ..., N; (12) Assume that at time t, the constellation contains N satellites and M communication links. Let the number of satellites v be V. i and satellite v j The shortest path between them is d ij Then the satellite density can be obtained. (13) Combining the ratio of the number of corresponding satellites in the shortest path of all satellite pairs in the constellation topology, i.e., the satellite betweenness number Solve for satellite v i Importance is After normalization, the weight of the satellite's importance in the constellation is: Among them, g jk For satellite v j With satellite v k The number of shortest paths, g jk (i) is g jk via satellite v i Quantity, It is satellite v i For satellite v j With satellite v k The contribution of betweenness reflects satellite position information; (14) Combining the satellite reliability model obtained in step (11) and the satellite importance weight values ​​in the constellation obtained in step (13), output the health status of all satellites in the constellation:

2. The method for on-orbit health assessment of star clusters based on time-varying topology according to claim 1, characterized in that, The health status is divided into three levels: healthy, sub-healthy, and faulty.

3. The method for on-orbit health assessment of star clusters based on time-varying topology according to claim 2, characterized in that, The health levels are defined as follows: healthy (0.8000, 1.0000), sub-healthy (0.4500, 0.7999), and faulty (0.0000, 0.4499).

4. The method for on-orbit health assessment of star clusters based on time-varying topology according to claim 1, characterized in that, Step (2) specifically includes: (21) Analyze whether there are communication links between satellites and construct the inter-satellite correlation matrix A(G)={a ij } N×N That is, if satellite v i and satellite v j There is a link between them, a ij =1; if satellite v i and satellite v j There is no link between them, a ij =0; the natural connectivity is the average of all eigenvalues ​​of the interstellar correlation matrix A(G) with respect to the natural logarithm and natural exponent. In the formula, λ p It is the p-th eigenvalue of A(G), and its natural connectivity is... The larger the value, the more robust the connection. (22) When link e j After failure, the recalculated natural connectivity is used If expressed as such, then the importance of the link is... Normalization can yield the link importance weight values. (23) During link analysis, it is assumed that satellites are visible to each other. Only the reliability of communication link transmission needs to be considered. The reliability of inter-satellite links is expressed as: In the formula, λ ej Indicates link e j The failure rate, j = 1, 2, ..., M; combining the inter-satellite link reliability and the link importance weight value obtained in step (22), the health of the inter-satellite communication link is obtained as follows:

5. The method for on-orbit health assessment of star clusters based on time-varying topology according to claim 4, characterized in that, Step (3) specifically includes: (31) Compare the ideal three-axis position, three-axis attitude and energy consumption of the satellite during its motion with their actual values, select the obtained error as a secondary indicator, and the satellites in the constellation as primary indicators. Based on the analytic hierarchy process, establish a performance evaluation index system structure. (32) Based on the performance evaluation index system structure, the weight of each index is determined using the entropy weight method; assuming that the error of each secondary index is E ij The normalized value is z ij The weight value of the secondary indicator is The weights of the primary indicators can be obtained from the following formula: z i ＝1-r i In the formula, r i This represents the information entropy, or uncertainty, of each secondary indicator; z i This represents the information utility value; ω i The weight values ​​for the primary indicators; (33) Combining the performance evaluation index system model obtained in step (31) and the index parameters obtained in step (32), output the overall performance of the star cluster.

6. The method for on-orbit health assessment of star clusters based on time-varying topology according to claim 5, characterized in that, Step (4) specifically includes: (41) Assume X = (x1) r ,…,x k r ) is factor x i State vectors, i = 1, ..., k, W = (ω1, ..., ω k ) is factor x i The constant weight vector, S(X)=(S1(X),…,S k If (X) is the state-variable weight vector, then factor x i The variable weight vector is obtained by the Hadamard product, that is: In the formula, The weights are determined based on the actual situation. When it is determined that a momentary fault has occurred in the star cluster, the weight values ​​ω of steps (13), (22) and (32) are adjusted according to the different faults using the above formula (1) to achieve variable weights. (42) After the weight transfer is completed, the current health H of the constellation is taken into account. swarm (t) is affected by the health status of the previous time step, i.e., there is a temporal dependency. Using a nonlinear weighting method, the cluster health function is obtained: In the formula, σ is the time series function, and k1, k2, k3 are the influencing parameters; (43) By using the first exponential smoothing method, the health value of the star cluster is fitted to estimate σ=1-α,α∈[0,1] and k1+k2+k3=α; according to the star cluster health function in step (41), the health of the star cluster at time t in the life cycle is calculated, and the corresponding health level is obtained.

7. A health assessment terminal, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method as described in any one of claims 1-6.

Images