Satellite large-scale structure in-orbit deformation distributed measurement adaptive planning method and system

Abstract

The invention provides a satellite large-scale structure in-orbit deformation distributed measurement adaptive planning method and system, and the method comprises the steps: S1, dividing a satellite structure into a grid model, and obtaining the in-orbit temperature distribution and strain distribution of the satellite structure; s2, establishing a search model based on a particle swarm optimization algorithm and Kriging unbiased estimation, and further obtaining an optimal strain layout and an optimal temperature layout; and S3, fusing the strain optimal layout and the temperature optimal layout through K-Means clustering and a node optimization model, performing optimization adjustment according to an adjacent movement strategy, and outputting a final layout. According to the method, a layout scheme meeting strain and temperature measurement conditions is preliminarily determined through a modal method and a particle swarm optimization algorithm, superposition optimization of strain layout and temperature layout is performed by using K-Means, Kriging unbiased estimation and an adjacent movement strategy, and high-precision adaptive planning of in-orbit deformation distributed measurement of a large-scale structure is further realized.

Description

Technical Field

[0001] The present invention relates to the field of satellite technology, and in particular to a method and system for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures. Background Art

[0002] The on-orbit temperature variation of large remote sensing satellites can reach 200-300℃. Large structures such as satellite mounting plates and large array payloads are deformed by changes in external heat flux, causing errors in the payload's line of sight pointing or surface with orbital and seasonal changes, thereby affecting detection performance indicators such as image positioning and antenna gain. Therefore, it is necessary to monitor the on-orbit thermal deformation of large structures in real time and compensate for the thermal deformation error terms.

[0003] Distributed fiber Bragg grating measurement technology can monitor the thermal deformation of large structures in orbit in real time. However, the distribution of thermal deformation of large structures is not uniform due to factors such as materials and heat exchange. Large-spacing uniform layout is prone to ignoring local sharp features. Due to the limitation of on-board resources, high-density layout is not realistic, and high-cost-effective monitoring requires differential planning. Therefore, an adaptive planning method that comprehensively considers the distribution of thermal deformation of large structures in orbit is needed. Summary of the invention

[0004] In view of the defects in the prior art, the object of the present invention is to provide a method and system for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures.

[0005] According to the present invention, a method for adaptive planning of distributed measurement of on-orbit deformation of a large satellite structure is provided, comprising:

[0006] Step S1: Divide the satellite structure into a grid model to obtain the on-orbit temperature distribution and strain distribution of the satellite structure;

[0007] Step S2: A search model is established based on a particle swarm optimization algorithm and Kriging unbiased estimation to further obtain the optimal strain layout and the optimal temperature layout;

[0008] Step S3: The optimal strain layout and the optimal temperature layout are integrated through K-Means clustering and node optimization model, and the optimization and adjustment are made according to the neighboring movement strategy to output the final layout.

[0009] Preferably, step S2 includes the following sub-steps:

[0010] Step S2.1: Obtain the on-orbit temperature distribution and strain distribution h of the satellite structure through thermal environment simulation and modal method i =(p i ,ε i ,T i )(i=1,2,…,N);

[0011] Among them, p is the position data of the node, ε is the strain data of the node, and T is the temperature data of the node;

[0012] Step S2.2: Taking the node strain gradient and node position as the optimization target, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal strain layout H ε ;

[0013] Step S2.3: Taking the node temperature gradient and node position as the optimization target, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal temperature layout H T .

[0014] Preferably, step S2.2 includes taking node strain gradient and node position as optimization targets, constructing a two-layer iteration - node strain gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the optimal strain layout that satisfies the maximum strain fitting error less than a given threshold.

[0015] Preferably, step S2.3 includes taking node temperature gradient and node position as optimization targets, constructing a two-layer iteration - node temperature gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the temperature optimal layout that satisfies the maximum temperature fitting error less than a given threshold.

[0016] Preferably, step S3 includes the following sub-steps:

[0017] Step S3.1: Use the K-Means method to optimize the strain layout H ε and temperature optimal layout H T Clustering is performed, and the clustered areas and non-clustered areas of the two layouts are merged in turn through the node optimization strategy to obtain the fused layout HO;

[0018] Step S3.2: Based on the layout directions v1 and v2, optimize the fiber Bragg grating nodes through the neighboring movement strategy and output the optimized layout HO * ;

[0019] Step S3.3: Build a fitting model through Kriging unbiased estimation, optimize and adjust the out-of-tolerance nodes according to the equivalent spacing, and output the final layout.

[0020] Preferably, step S3.1 includes clustering the optimal strain layout and the optimal temperature layout by using a K-Means algorithm to obtain a cluster of optimal strain layout nodes. and temperature optimal layout node aggregation The clustering overlap optimization, strain-temperature node clustering fusion and non-clustering area node optimization are carried out in sequence.

[0021] Preferably, the step of sequentially fusing the clustered areas and non-clustered areas of the two layouts through the node optimization strategy includes:

[0022] In the clustered area, for overlapping areas and Eliminate duplicate nodes, perform clustered local area fitting based on the node set where the number of nodes in the overlapping area is less than a preset value and the node set in the non-overlapping area, and select nodes in the set where the number of nodes in the overlapping area is greater than a preset value according to the fitting error;

[0023] In the non-clustered area, an overall fit is performed based on the node set whose node number is less than a preset value and the determined node clusters, and nodes in the set whose node number in the non-clustered area is greater than the preset value are preferred.

[0024] Preferably, the adjacent movement strategy includes locking the movement direction based on the number of wiring nodes and the equivalent spacing according to the number of adjacent wiring nodes to determine whether movement is required; and determining the movement position in combination with the equivalent spacing calculated from the adjacent wiring, strain value and temperature value.

[0025] According to the present invention, a distributed measurement adaptive planning system for on-orbit deformation of a large satellite structure is provided, comprising:

[0026] Module M1: Divide the satellite structure into a grid model to obtain the on-orbit temperature distribution and strain distribution of the satellite structure;

[0027] Module M2: A search model is established based on the particle swarm optimization algorithm and Kriging unbiased estimation to further obtain the optimal strain layout and temperature layout;

[0028] Module M3: The optimal strain layout and the optimal temperature layout are integrated through K-Means clustering and node optimization model, and the optimization and adjustment are made according to the neighboring movement strategy to output the final layout.

[0029] Preferably, the module M2 includes the following submodules:

[0030] Module M2.1: Obtaining the on-orbit temperature and strain distribution of satellite structures through thermal environment simulation and modal method i =(p i ,ε i ,T i )(i=1,2,…,N);

[0031] Among them, p is the position data of the node, ε is the strain data of the node, and T is the temperature data of the node;

[0032] Module M2.2: Taking node strain gradient and node position as optimization targets, the optimal strain layout H is determined using particle swarm optimization algorithm and Kriging unbiased estimation ε ;

[0033] Module M2.3: Taking node temperature gradient and node position as optimization targets, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal temperature layout H T .

[0034] Preferably, the module M2.2 includes taking node strain gradient and node position as optimization targets, constructing a two-layer iteration - node strain gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the optimal strain layout that satisfies the maximum strain fitting error less than a given threshold.

[0035] Preferably, the module M2.3 includes taking node temperature gradient and node position as optimization targets, constructing a two-layer iteration - node temperature gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the temperature optimal layout that satisfies the maximum temperature fitting error less than a given threshold.

[0036] Preferably, the module M3 includes the following submodules:

[0037] Module M3.1: Use K-Means method to optimally arrange strains H ε and temperature optimal layout H T Clustering is performed, and the clustered areas and non-clustered areas of the two layouts are merged in turn through the node optimization strategy to obtain the fused layout HO;

[0038] Module M3.2: Based on the layout directions v1 and v2, optimize the fiber Bragg grating nodes through the adjacent movement strategy and output the optimized layout HO * ;

[0039] Module M3.3: Build a fitting model through Kriging unbiased estimation, optimize and adjust the out-of-tolerance nodes according to the equivalent spacing, and output the final layout.

[0040] Preferably, the module M3.1 includes clustering the optimal strain layout and the optimal temperature layout by using the K-Means algorithm to obtain the optimal strain layout node cluster. and temperature optimal layout node aggregation The clustering overlap optimization, strain-temperature node clustering fusion and non-clustering area node optimization are carried out in sequence.

[0041] Preferably, the step of sequentially fusing the clustered areas and non-clustered areas of the two layouts through the node optimization strategy includes:

[0042] In the clustered area, for overlapping areas and Eliminate duplicate nodes, perform clustered local area fitting based on the node set where the number of nodes in the overlapping area is less than a preset value and the node set in the non-overlapping area, and select nodes in the set where the number of nodes in the overlapping area is greater than a preset value according to the fitting error;

[0043] In the non-clustered area, an overall fit is performed based on the node set whose node number is less than a preset value and the determined node clusters, and nodes in the set whose node number in the non-clustered area is greater than the preset value are preferred.

[0044] Preferably, the adjacent movement strategy includes locking the movement direction based on the number of wiring nodes and the equivalent spacing according to the number of adjacent wiring nodes to determine whether movement is required; and determining the movement position in combination with the equivalent spacing calculated from the adjacent wiring, strain value and temperature value.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The present invention preliminarily determines the layout scheme that meets the strain and temperature measurement conditions through the modal method and particle swarm optimization algorithm, and uses K-Means, Kriging unbiased estimation and neighbor movement strategy to perform superimposed optimization of strain layout and temperature layout, further realizing high-precision adaptive planning of distributed measurement of on-track deformation of large structures.

[0047] 2. The present invention performs adaptive planning based on strain distribution and comprehensive consideration of overall and local fitting accuracy, thereby avoiding neglect of local sharp features caused by large-spacing layout, improving accuracy, and having good application prospects.

[0048] Other beneficial effects of the present invention will be explained in the specific implementation manner through the introduction of specific technical features and technical solutions. Through the introduction of these technical features and technical solutions, those skilled in the art should be able to understand the beneficial technical effects brought about by the technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0050] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0051] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0052] Reference Figure 1 As shown, a distributed measurement adaptive planning method for on-orbit deformation of a large satellite structure includes the following steps:

[0053] S1: Using the finite element analysis method, the satellite structure is divided into mesh models with consistent shapes and sizes;

[0054] S2: Comprehensively consider the satellite space environment, structural layout, heat flow transmission and other factors, and obtain the on-orbit temperature distribution and strain distribution of the satellite structure through thermal environment simulation and modal method. i =(p i ,ε i ,T i )(i=1,2,…,N), where p is the position data of the node, ε is the strain data of the node, and T is the temperature data of the node;

[0055] S3: Taking the node strain gradient and node position as the optimization target, the optimal strain layout H is preliminarily determined by particle swarm optimization algorithm and Kriging unbiased estimation. ε ; Construct a two-layer iteration model - node strain gradient iteration (outer layer) and fiber Bragg grating layout iteration (inner layer), and output the optimal strain layout that satisfies the maximum strain fitting error less than a given threshold.

[0056] S4: Taking the node temperature gradient and node position as the optimization target, the optimal temperature layout H is preliminarily determined by particle swarm optimization algorithm and Kriging unbiased estimation. T ; Construct a two-layer iteration model - node temperature gradient iteration (outer layer) and fiber Bragg grating layout iteration (inner layer), and output the temperature optimal layout that satisfies the maximum temperature fitting error less than a given threshold.

[0057] S5: Use K-Means method to optimally arrange the strain H ε and temperature optimal layout H T Clustering is performed to obtain the optimal layout node clustering and temperature optimal layout node aggregation The clustering overlap optimization, strain-temperature node clustering fusion and non-clustering area node optimization are performed in sequence to obtain HO;

[0058] The implementation of the node selection strategy in the cluster area is as follows: and First, duplicate nodes are removed, and local area fitting is performed based on the node set with fewer nodes in the overlapping area and the node set in the non-overlapping area. Nodes in the set with more nodes in the overlapping area are selected according to the fitting error.

[0059] The implementation of the node optimization strategy in the non-clustered area is specifically to perform overall fitting based on the node set with a smaller number of nodes and the determined node clusters, and to optimize the nodes in the set with a larger number of nodes in the non-clustered area.

[0060] S6: Based on the layout directions v1 and v2, optimize the fiber Bragg grating nodes through the adjacent movement strategy, reduce the fiber Bragg grating orthogonal layout lines as much as possible, and output the optimized layout HO * The adjacent movement strategy is designed based on the number of wiring nodes and the equivalent spacing. First, the movement direction is locked according to the number of adjacent wiring nodes to determine whether movement is needed. Secondly, the movement position is determined by combining the equivalent spacing calculated by adjacent wiring, strain value and temperature value.

[0061] S7: Based on the layout output by S6, a fitting model is constructed through Kriging unbiased estimation. For out-of-tolerance nodes, nodes are locally added to the adjacent wiring of the out-of-tolerance nodes based on equivalent spacing for optimization and adjustment, and the final layout is output.

[0062] The present invention preliminarily determines the layout scheme that meets the strain and temperature measurement conditions through the modal method and particle swarm optimization algorithm, and uses K-Means, Kriging unbiased estimation and neighbor movement strategy to perform superimposed optimization of strain layout and temperature layout, further realizing high-precision adaptive planning of distributed measurement of on-orbit deformation of large structures.

[0063] The above is a basic embodiment of the present invention. The technical solution of the present invention is further described below through a preferred embodiment.

[0064] Example 1

[0065] Reference Figure 1 As shown in the figure, an adaptive planning method for distributed measurement of on-orbit deformation of large satellite structures is presented. The overall idea is: obtain the temperature distribution and strain distribution of the satellite structure through thermal environment simulation and modal method; use particle swarm optimization algorithm and Kriging unbiased estimation to preliminarily determine the optimal temperature layout and strain layout; use K-Means clustering and node optimization model to merge the optimal layouts for the two variables, optimize the orthogonal layout lines of the fiber Bragg grating through the neighboring movement strategy, and output the optimal layout.

[0066] The specific steps are as follows:

[0067] Step S1, dividing the satellite structure into grid models with the same shape and size by finite element analysis method;

[0068] Step S2, comprehensively considering the satellite space environment, structural layout, heat flow transmission and other factors, constructing a satellite on-orbit thermal environment model, simulating the extreme environment on the satellite, and outputting the satellite structure temperature distribution;

[0069] Step S3: obtain the satellite structure strain distribution through modal simulation, and record the node position, strain and temperature data as h i =(p i ,ε i ,T i )(i=1,2,…,N), where p is the node position data, ε is the node strain data, T is the node temperature data, and the extreme values ​​of strain and temperature simulation data are ε respectively. - , ε + 、T + and T - .

[0070] Step S4, selecting the distributed fiber grating layout directions v1 and v2;

[0071] Step S5: Search the layout by particle swarm optimization algorithm and Kriging unbiased estimation method, and iterate the output to satisfy max(Δε i |)≤ε lim The optimal layout of strain

[0072] a. Set the strain gradient g ε , and determine the number of nodes;

[0073] b. Set the number of iterations and fitness function minΣ|Δε i |;

[0074] c. Initialize particles, velocity vectors, learning parameters, local optimal positions, and global optimal positions;

[0075] d. According to the layout represented by each particle, the corresponding strain interpolation model is constructed using the Kriging unbiased estimation method to calculate the absolute fitting error |Δε of all nodes. i |, and then calculate the fitness value of each particle;

[0076] e. Update the local optimal position and the global optimal position;

[0077] f. If the layout represented by the global optimal position satisfies max(Δε i |)≤ε lim , the iteration ends and the layout is output, otherwise the particles and their velocity vectors are updated;

[0078] g. Repeat d and e until the iteration ends;

[0079] h. If the layout represented by the global optimal position satisfies max(Δε i |)≤ε lim , output this layout; if it is not satisfied, update the strain gradient g ε =g ε -Σ|Δε i| / N, and determine the number of nodes, and repeat b~g.

[0080] Step S6: Search the layout by particle swarm optimization algorithm and Kriging unbiased estimation method, and iterate the output to satisfy max(|ΔT i |)≤T lim Optimal layout for temperature The specific process of step S6 is the same as that of step S5, except that the basis for calculation and evaluation is the temperature parameter.

[0081] Step S7: Use K-Means algorithm to calculate H ε and H T Perform clustering to obtain multiple strain optimal layout node clusters containing a certain number of nodes and temperature optimal layout node aggregation

[0082] in,

[0083]

[0084] The two layouts are merged using the node optimization model;

[0085] a. H after clustering ε and H T Perform overlap judgment, if and multiple Overlap, multiple The nodes in the area included in the new boundary are considered as a new cluster and delete these Similarly, if and multiple Overlap, multiple The nodes in the area included in the new boundary are considered as a new cluster and delete these Update n1 and n2;

[0086] b. Elimination and Repeating nodes;

[0087] c. Remember and The overlapping area is D jk ,mark and The node sets in this area are and Will and The node set with a smaller number is used as the benchmark set HB jk , a large number of nodes are combined as the candidate set HS jk ;

[0088] d. According to HB jk The nodes in the HS are obtained by Kriging unbiased estimation fitting. jk Corresponding to the interpolation model of the parameters, calculate HS jk The fitting value of each node is compared with the simulation value;

[0089] e. Arrange the parameter fitting errors of each node from large to small. If the accuracy requirement (ε lim or T lim , by HS jk OK), set HS jk The node with the largest fitting error moves to HB jk ; If all nodes meet the accuracy requirements, delete HS jk ,Will HB jk Put into the node set HO, from H ε and H T Delete and Repeat steps d to f until all clusters are processed.

[0090] f. Repeat steps d and e until HS jk The traversal is completed or the accuracy requirement is met.

[0091] g. For non-clustered points and The node set with a smaller number is taken as the benchmark set HB - , a large number of node sets as the candidate set HS - ;

[0092] h. According to HB - The nodes of HO and HS are obtained by Kriging unbiased estimation. - Corresponding to the interpolation model of the parameters, calculate HS - The fitting value of each node is compared with the simulation value;

[0093] i. Arrange the parameter fitting errors of each node from large to small. If the accuracy requirement (ε lim or T lim , by HS - OK), set HS - The node with the largest fitting error moves to HB - ; If all nodes meet the accuracy requirements, delete HS - , HB - The node set HO is added and the fusion of the two layouts is completed.

[0094] j. Repeat steps h and i until HS- The traversal is completed or the accuracy requirement is met.

[0095] Step S8: Optimize the routing based on v1 and v2 through the neighboring movement strategy, and output the optimized layout HO * ;

[0096] a. Establish a strain interpolation model and a temperature interpolation model determined by HO and Kriging unbiased estimation, normalize the strain fitting error and temperature fitting error of each HO node, sum them up and arrange them in ascending order;

[0097] b. Select the first node

[0098] c. Confirm The wiring parallel to v1 and the two adjacent wirings are counted, and the number of nodes on the wiring is counted. If the number of nodes on the wiring is greater than that of the adjacent wiring, its position in the v2 direction is locked.

[0099] d. Confirm The wiring parallel to v2 and the two adjacent wirings are counted, and the number of nodes on the wiring is counted. If the number of nodes on the wiring is greater than that of the adjacent wiring, its position in the direction of v1 is locked.

[0100] e. Determine whether the node has a locked direction:

[0101] i. If All are locked, the node position remains unchanged, and step f is skipped;

[0102] ii. If locking, Unlocked, determine the equivalent spacing of its distance to the adjacent wiring parallel to v1 ( d ε d T Represents p i The strain difference and temperature difference between the strain at the target position and the target position, the target position is the origin at p i The intersection of the direction vector v2(v1) and the adjacent wiring parallel to v1(v2)), if the equivalent spacing is equal, initially move to the adjacent wiring with a larger number of nodes; if the equivalent spacing is unequal, initially move to the adjacent wiring with the closest distance;

[0103] iii. If locking, Unlocked, determine the equivalent spacing between it and its adjacent wiring parallel to v2. If the equivalent spacing is equal, initially move to the adjacent wiring with a larger number of nodes. If the equivalent spacing is not equal, initially move to the adjacent wiring with the closest distance.

[0104] iv. If If they are not locked, determine the equivalent spacing between the distance and the four adjacent wirings. If the equivalent spacing is equal, initially move to the adjacent wiring with more nodes. If the equivalent spacing is not equal, initially move to the adjacent wiring with the closest distance.

[0105] f. Calculate the node parameter fitting error based on the parameter interpolation model and the preliminary adjustment layout. If the accuracy requirement is not met, the node will not be moved; if the accuracy requirement is met, the new node will replace the original node;

[0106] g. Select the next node and repeat c~f until all nodes are traversed and output the optimized layout HO * .

[0107] Step S9: According to HO * Re-obtain the overall fitting model. If all node strain fitting errors and temperature fitting errors meet the accuracy requirements, HO * This is the final output layout; otherwise, add nodes to the adjacent wiring of the out-of-tolerance node according to the equivalent spacing and update the HO * , until the accuracy condition is met.

[0108] The present invention also provides a distributed measurement and adaptive planning system for on-orbit deformation of large satellite structures. The distributed measurement and adaptive planning system for on-orbit deformation of large satellite structures can be realized by executing the process steps of the distributed measurement and adaptive planning method for on-orbit deformation of large satellite structures. That is, those skilled in the art can understand the distributed measurement and adaptive planning method for on-orbit deformation of large satellite structures as a preferred implementation of the distributed measurement and adaptive planning system for on-orbit deformation of large satellite structures.

[0109] Specifically, a distributed measurement adaptive planning system for on-orbit deformation of large satellite structures includes:

[0110] Module M1: Divide the satellite structure into a grid model to obtain the on-orbit temperature distribution and strain distribution of the satellite structure;

[0111] Module M2: A search model is established based on the particle swarm optimization algorithm and Kriging unbiased estimation to further obtain the optimal strain layout and temperature layout;

[0112] Module M3: The optimal strain layout and the optimal temperature layout are integrated through K-Means clustering and node optimization model, and the optimization and adjustment are made according to the neighboring movement strategy to output the final layout.

[0113] The module M2 includes the following submodules:

[0114] Module M2.1: Obtaining the on-orbit temperature and strain distribution of satellite structures through thermal environment simulation and modal method i=(p i ,ε i ,T i )(i=1,2,…,N);

[0115] Among them, p is the position data of the node, ε is the strain data of the node, and T is the temperature data of the node;

[0116] Module M2.2: Taking node strain gradient and node position as optimization targets, the optimal strain layout H is determined using particle swarm optimization algorithm and Kriging unbiased estimation ε ;

[0117] Module M2.3: Taking node temperature gradient and node position as optimization targets, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal temperature layout H T .

[0118] The module M2.2 includes taking node strain gradient and node position as optimization targets, constructing a two-layer iteration - node strain gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the optimal strain layout that satisfies the maximum strain fitting error less than a given threshold.

[0119] The module M2.3 includes taking node temperature gradient and node position as optimization targets, constructing a two-layer iteration - node temperature gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the temperature optimal layout that satisfies the maximum temperature fitting error less than a given threshold.

[0120] The module M3 includes the following submodules:

[0121] Module M3.1: Use K-Means method to optimally arrange strains H ε and temperature optimal layout H T Clustering is performed, and the clustered areas and non-clustered areas of the two layouts are merged in turn through the node optimization strategy to obtain the fused layout HO;

[0122] Module M3.2: Based on the layout directions v1 and v2, optimize the fiber Bragg grating nodes through the adjacent movement strategy and output the optimized layout HO * ;

[0123] Module M3.3: Build a fitting model through Kriging unbiased estimation, optimize and adjust the out-of-tolerance nodes according to the equivalent spacing, and output the final layout.

[0124] The module M3.1 includes clustering the optimal strain layout and the optimal temperature layout by using the K-Means algorithm to obtain the optimal strain layout node cluster and temperature optimal layout node aggregation The clustering overlap optimization, strain-temperature node clustering fusion and non-clustering area node optimization are carried out in sequence.

[0125] The step of sequentially fusing the clustered areas and non-clustered areas of the two layouts through the node optimization strategy includes:

[0126] In the clustered area, for overlapping areas and Eliminate duplicate nodes, perform clustered local area fitting based on the node set where the number of nodes in the overlapping area is less than a preset value and the node set in the non-overlapping area, and select nodes in the set where the number of nodes in the overlapping area is greater than a preset value according to the fitting error;

[0127] In the non-clustered area, an overall fit is performed based on the node set whose node number is less than a preset value and the determined node clusters, and nodes in the set whose node number in the non-clustered area is greater than the preset value are preferred.

[0128] The adjacent movement strategy includes locking the movement direction based on the number of wiring nodes and the equivalent spacing, judging whether movement is required according to the number of adjacent wiring nodes; and determining the movement position in combination with the equivalent spacing calculated from the adjacent wiring, strain value and temperature value.

[0129] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0130] In the description of the present application, it should be understood that the terms "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0131] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. An adaptive planning method for distributed measurement of on-orbit deformation of large satellite structures, characterized in that: include: Step S1: Divide the satellite structure into a grid model to obtain the on-orbit temperature distribution and strain distribution of the satellite structure; Step S2: A search model is established based on a particle swarm optimization algorithm and Kriging unbiased estimation to further obtain the optimal strain layout and the optimal temperature layout; Step S3: The optimal strain layout and the optimal temperature layout are integrated through K-Means clustering and node optimization model, and the optimization and adjustment are made according to the neighboring movement strategy to output the final layout.

2. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 1 is characterized in that: The step S2 comprises the following sub-steps: Step S2.1: Obtain the on-orbit temperature distribution and strain distribution h of the satellite structure through thermal environment simulation and modal method i =(p i ,ε i ,T i )(i=1,2,…,N); Among them, p is the position data of the node, ε is the strain data of the node, and T is the temperature data of the node; Step S2.2: Taking the node strain gradient and node position as the optimization target, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal strain layout H ε ; Step S2.3: Taking the node temperature gradient and node position as the optimization target, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal temperature layout H T .

3. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 2 is characterized in that: The step S2.2 includes taking the node strain gradient and the node position as the optimization targets, constructing a two-layer iteration - node strain gradient iteration and fiber Bragg grating layout iteration model through the particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the optimal strain layout that satisfies the maximum strain fitting error less than a given threshold.

4. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 3 is characterized in that: The step S2.3 includes taking the node temperature gradient and node position as optimization targets, constructing a two-layer iteration - node temperature gradient iteration and fiber Bragg grating layout iteration model through particle swarm optimization algorithm and Kriging unbiased estimation, and outputting the temperature optimal layout that satisfies the maximum temperature fitting error less than a given threshold.

5. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 2, characterized in that: The step S3 comprises the following sub-steps: Step S3.1: Use the K-Means method to optimize the strain layout H ε and temperature optimal layout H T Clustering is performed, and the clustered areas and non-clustered areas of the two layouts are merged in turn through the node optimization strategy to obtain the fused layout HO; Step S3.2: Based on the layout directions v1 and v2, optimize the fiber Bragg grating nodes through the neighboring movement strategy and output the optimized layout HO * ; Step S3.3: Build a fitting model through Kriging unbiased estimation, optimize and adjust the out-of-tolerance nodes according to the equivalent spacing, and output the final layout.

6. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 5, characterized in that: The step S3.1 includes clustering the optimal strain layout and the optimal temperature layout respectively by using the K-Means algorithm to obtain the optimal strain layout node cluster and temperature optimal layout node aggregation The clustering overlap optimization, strain-temperature node clustering fusion and non-clustering area node optimization are carried out in sequence.

7. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 5, characterized in that: The step of sequentially fusing the clustered areas and non-clustered areas of the two layouts through the node optimization strategy includes: In the clustered area, for overlapping areas and Remove duplicate nodes, perform clustered local area fitting based on the node set where the number of nodes in the overlapping area is less than a preset value and the node set in the non-overlapping area, and select the nodes in the set where the number of nodes in the overlapping area is greater than a preset value according to the fitting error; In the non-clustered area, an overall fit is performed based on the node set whose node number is less than a preset value and the determined node clusters, and nodes in the set whose node number in the non-clustered area is greater than the preset value are preferred.

8. The method for adaptive planning of distributed measurement of on-orbit deformation of large satellite structures according to claim 5, characterized in that: The adjacent movement strategy includes locking the movement direction according to the number of adjacent wiring nodes and the equivalent spacing, and determining whether movement is required; The shift position is determined by combining the equivalent spacing calculated from adjacent wiring, strain values, and temperature values.

9. An adaptive planning system for distributed measurement of on-orbit deformation of large satellite structures, characterized in that: include: Module M1: Divide the satellite structure into a grid model to obtain the on-orbit temperature distribution and strain distribution of the satellite structure; Module M2: A search model is established based on the particle swarm optimization algorithm and Kriging unbiased estimation to further obtain the optimal strain layout and temperature layout; Module M3: The optimal strain layout and the optimal temperature layout are integrated through K-Means clustering and node optimization model, and the optimization and adjustment are made according to the neighboring movement strategy to output the final layout.

10. The satellite large structure on-orbit deformation distributed measurement adaptive planning system according to claim 9, characterized in that: The module M2 Includes the following submodules: Module M2.1: Obtaining the on-orbit temperature and strain distribution of satellite structures through thermal environment simulation and modal method i =(p i ,ε i ,T i )(i=1,2,…,N); Among them, p is the position data of the node, ε is the strain data of the node, and T is the temperature data of the node; Module M2.2: Taking node strain gradient and node position as optimization targets, the optimal strain layout H is determined using particle swarm optimization algorithm and Kriging unbiased estimation ε ; Module M2.3: Taking node temperature gradient and node position as optimization targets, the particle swarm optimization algorithm and Kriging unbiased estimation are used to determine the optimal temperature layout H T .

Images