A port- hamiltonian structure preserving automatic order reduction method and system
By employing an automatic order selection model reduction method based on port-Hamilton structures, the problem of missing weak port high-energy low-damping flexible modes in existing technologies is solved. This method achieves simultaneous consideration of the energy and residual vibration characteristics of multiple flexible attachments during the order reduction process, thereby improving the attitude control and simulation accuracy of large flexible spacecraft.
Patent Information
- Application Number
- CN202611446671.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-17
- Publication Date
- 2026-11-10
AI Technical Summary
Existing spacecraft model order reduction methods tend to miss weak port high-energy low-damping flexible modes, and cannot simultaneously take into account the energy of multiple flexible attachments, residual vibration characteristics and port Hamiltonian structure fidelity, making it difficult to meet the requirements of high-precision attitude and vibration coupling simulation and attitude control design for large flexible spacecraft.
The port-Hamilton structure-maintained automatic order selection model reduction method is adopted. By converting the full-order flexible dynamic model into the port-Hamilton form, separating the energy storage, interconnection structure and dissipation characteristics, dividing the structure into groups, calculating the energy coverage capture rate and the low-damped residual vibration proxy capture rate, constructing the projection matrix and performing projection transformation, the target reduced-order model is obtained.
While reducing the model order, the energy distribution and low-damped residual vibration modes of the main flexible accessory structure group are maintained, which improves the engineering usability and computational efficiency of the reduced-order model, ensures the response characteristics of the control port, and meets the requirements of attitude control and residual vibration simulation.
Smart Images

Figure CN123069795A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of spacecraft dynamics modeling and model order reduction technology, and more specifically, to a method and system for automatic order selection and model order reduction of port-Hamilton structure. Background Technology
[0002] Large flexible spacecraft are core space platforms for space exploration, Earth observation, communication relay, and on-orbit servicing. They need to achieve high-precision attitude pointing, payload operation, and attitude maneuver control over long periods in the complex microgravity environment. To reduce launch weight, expand deployment size, and improve payload capacity, spacecraft accessories such as solar arrays, thin-film sails, deployable trusses, and flexible antennas generally adopt lightweight and flexible configurations. While these flexible structures offer advantages such as lightweight design and large-scale deployment, they can induce significant elastic vibrations, low-damped long-term residual vibrations, and platform-accessory rigid-flexible coupling dynamics problems. The dynamic models of spacecraft under attitude maneuvering, payload pointing, and orbit maintenance conditions exhibit high-order, strongly coupled, and multimodal characteristics, including low-frequency dense modes of solar array / thin-film sail and local vibration modes of truss / antenna. They also exhibit rigid-flexible coupling effects, uneven energy distribution among flexible components, and large differences in control port responses. Directly using full-order high-precision models for controller design, frequency domain analysis, attitude and vibration co-simulation, and online real-time calculation is extremely difficult. Furthermore, the persistent low-damped residual vibration after the maneuver will degrade attitude stability, payload line-of-sight pointing accuracy, antenna beam stability, and even threaten the safety of flexible appendage structures.
[0003] Current model reduction techniques cannot quantitatively characterize the degree to which the reduced model retains the dynamics of each independent flexible attachment, and cannot specifically suppress the risk of low-damped residual vibration during the modal screening stage. The screening mechanism has inherent defects, and it is very easy to miss key flexible modes with weak control port response but high energy and low damping, resulting in distortion of residual vibration prediction in the reduced model and loss of dynamic characteristics of flexible attachments, which makes it difficult to meet the needs of high-precision attitude and vibration coupling simulation and attitude control design for large flexible spacecraft. Summary of the Invention
[0004] The problem solved by this invention is that existing spacecraft model order reduction methods tend to miss weak port high-energy low-damping flexible modes, and cannot simultaneously take into account the energy of multiple flexible attachments, residual vibration characteristics and port Hamiltonian structure fidelity.
[0005] To address the aforementioned problems, this invention provides a method, system, electronic device, and storage medium for automatic order reduction of port-Hamilton structure-preserving automatic order selection models.
[0006] In a first aspect, the present invention provides a method for reducing the order of a port-Hamilton structure-preserving automatic order selection model, comprising:
[0007] Obtain the full-order flexible dynamic model of the large flexible spacecraft, and perform port-Hamilton structure transformation on the full-order flexible dynamic model to obtain the port-Hamilton system;
[0008] The flexible modes in the port-Hamilton system are divided into structural groups to obtain several flexible accessory structural groups.
[0009] The unselected flexible modes are added to the current selected mode set. The energy coverage capture rate and low-damped residual vibration proxy capture rate of each flexible accessory structure group in the current selected mode set are calculated until each flexible accessory structure group meets the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold, and the port error meets the preset condition. Then the current selected mode set is the target selected mode set.
[0010] Based on the selected mode set of the target, a projection matrix is constructed, and the port-Hamilton system is transformed by projection to obtain the target reduced-order model.
[0011] Optionally, obtaining the full-order flexible dynamic model of the large flexible spacecraft and performing a port-Hamilton structure transformation on the full-order flexible dynamic model to obtain a port-Hamilton system includes:
[0012] Acquire a large flexible spacecraft, which includes a central platform, solar arrays, thin-film sails, deployable trusses, flexible antennas, and locally flexible accessories;
[0013] The central platform, the solar array, the thin-film sail, the deployable truss, the flexible antenna, and the local flexible accessories are processed by finite element modeling, modal synthesis, or equivalent high-order modeling to obtain a full-order flexible dynamic model, wherein the full-order flexible dynamic model includes a generalized flexible coordinate vector and a mass matrix;
[0014] The generalized momentum is obtained by multiplying the mass matrix by the derivative of the generalized flexible coordinate vector;
[0015] The state variables are obtained by combining the generalized flexible coordinate vector and the generalized momentum;
[0016] Based on the state variables, the full-order flexible dynamics model is equivalent to a port-Hamilton form, thus obtaining the port-Hamilton system.
[0017] Optionally, the process of dividing the flexible modes in the port-Hamilton system into structural groups yields several flexible accessory structural groups, including:
[0018] Based on the spacecraft physical structure, finite element substructure, or modal affiliation of the large flexible spacecraft, the solar array, thin-film sail, deployable truss, flexible antenna, and local flexible attachments in the flexible modes are divided into structural groups and assigned to different structural groups to obtain several flexible attachment structural groups.
[0019] Optionally, the energy coverage capture rate is calculated as follows:
[0020] ,
[0021] In the formula, represents the energy coverage capture rate of the g-th flexible attachment structure group; represents the summation of Hamiltonian energy contributions that belong to both the initial selected mode set S and the g-th flexible attachment structure group; This represents the summation of the Hamiltonian energy contributions of all the flexible modes within the g-th flexible attachment structure group.
[0022] Optionally, the low-damped residual vibration proxy capture rate is specifically calculated as follows:
[0023] Constructing a low-damped residual vibration proxy index ,
[0024] In the formula, Indicates the first Hamiltonian energy contribution of each flexible mode. Indicates the modal damping ratio. It is the smallest positive number. Indicates the first The correlation between a flexible mode and the frequency band of attitude maneuvering or control excitation;
[0025] Based on the low-damped residual vibration proxy index, the low-damped residual vibration proxy capture rate is obtained:
[0026] ,
[0027] In the formula, This represents the low-damping residual vibration surrogate capture rate of the g-th flexible attachment structure group; This represents the sum of residual vibration surrogate energy contributions belonging to the g-th flexible attachment structure group in the initial selected mode set; This represents the total residual vibration surrogate energy contribution within the g-th flexible attachment structure group.
[0028] Optionally, the port error satisfies a preset condition, specifically including:
[0029] Calculate the relative error of the control port frequency domain and the port error quantile value of the intermediate reduced-order model corresponding to the current selected mode set within the control bandwidth;
[0030] If the relative frequency error of the control port is greater than a preset error threshold or the port error quantile is greater than a preset quantile threshold, then according to the order of contribution of each flexible mode to the control port from high to low, the unselected flexible modes are continuously added to the current selected mode set until the relative frequency error of the control port is less than or equal to the preset error threshold and the port error quantile is less than or equal to the preset quantile threshold, and all the flexible accessory structure groups meet the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold.
[0031] Optionally, the step of constructing a projection matrix based on the selected target mode set and performing a projection transformation on the port-Hamilton system to obtain a reduced-order target model includes:
[0032] Based on the selected mode set, a projection matrix is constructed, wherein the projection matrix retains the generalized coordinate component and its corresponding generalized momentum component for each of the flexible modes, so that the Hamiltonian energy function corresponding to the port-Hamilton system after order reduction is kept in the physical form of the sum of kinetic and potential energy.
[0033] The antisymmetric interconnect matrix, positive semidefinite dissipation matrix, positive definite energy matrix and port input matrix in the port-Hamilton system are projected using the projection matrix to obtain the reduced antisymmetric interconnect matrix, positive semidefinite dissipation matrix, positive definite energy matrix and port input matrix.
[0034] Based on the reduced antisymmetric interconnect matrix, the reduced positive semidefinite dissipation matrix, the reduced positive definite energy matrix, and the reduced port input matrix, the reduced port-Hamilton system is reconstructed, and the reduced port-Hamilton system is used as the target reduced model.
[0035] Secondly, the port-Hamilton structure-preserving automatic order selection model reduction system of the present invention includes:
[0036] The data acquisition unit is used to acquire the full-order flexible dynamics model of the large flexible spacecraft, and to perform port-Hamilton structure transformation on the full-order flexible dynamics model to obtain the port-Hamilton system.
[0037] The data analysis unit is used to divide the flexible modes in the port-Hamilton system into structural groups to obtain several flexible accessory structural groups.
[0038] The data determination unit is used to supplement the currently selected mode set with the flexible modes that have not been selected, calculate the energy coverage capture rate and low damping residual vibration proxy capture rate of each flexible accessory structure group in the current selected mode set, until each flexible accessory structure group meets the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold, and the port error meets the preset condition, then the current selected mode set is the target selected mode set;
[0039] The data processing unit is used to construct a projection matrix based on the selected target mode set, and to perform a projection transformation on the port-Hamilton system to obtain a reduced-order target model.
[0040] Thirdly, the electronic device of the present invention includes: a processor and a memory, the memory being used to store a computer program;
[0041] When the computer program is loaded by the processor, it causes the processor to execute the aforementioned port-Hamilton structure-preserving automatic order selection model reduction method.
[0042] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described port-Hamilton structure-preserving automatic order selection model reduction method.
[0043] This invention discloses an automatic order selection model reduction method, system, electronic device, and storage medium for port-Hamilton structure preservation. By converting a full-order flexible dynamic model into a port-Hamilton form, it separates the model's energy storage (Q matrix), interconnect structure (J matrix), and dissipation characteristics (R matrix). This allows the order reduction process to be based on mathematical compression of the physical structure rather than simple curve fitting, avoiding the destruction of the system's physical meaning by traditional order reduction methods. The flexible modes are divided into structural groups, overcoming the limitation of traditional mode truncation based solely on frequency sorting. By calculating the energy coverage capture rate of each flexible mode within a specific structural group, it ensures that key components such as solar panels and antennas retain sufficient energy representativeness in the reduced-order model, preventing the omission of high-energy modes with weak ports. By constructing a low-damped residual vibration proxy capture rate, the risk of persistent residual vibration in each flexible mode is quantified, solving the problem of difficulty in maintaining residual vibration after attitude maneuvers in large flexible spacecraft. A step-by-step order selection strategy is adopted, balancing the vibration characteristics of attachments and control simulation requirements at the same order, improving the engineering usability of the reduced-order model. Finally, based on the selected mode set, a projection matrix is constructed to project and transform the port-Hamilton system to obtain the target reduced-order model. This reduces the number of order retained in the model while maintaining the same simulation accuracy, thus improving computational efficiency. Attached Figure Description
[0044] Figure 1This is a flowchart illustrating the automatic order selection model reduction method for port-Hamilton structure preservation in an embodiment of the present invention.
[0045] Figure 2 This is a comparison chart of the retained orders under the automatic order selection of various methods in the embodiments of the present invention;
[0046] Figure 3 This is a comparison chart of the residual vibration RMS error of each method in the embodiments of the present invention;
[0047] Figure 4 This is a comparison chart of group-level free decay errors for each method in the embodiments of the present invention;
[0048] Figure 5 This is a comparison chart of the displacement response curves of key flexible attachments in embodiments of the present invention;
[0049] Figure 6 This is a comparison chart of the free decay response curves of key flexible accessories in embodiments of the present invention;
[0050] Figure 7 This is a comparison chart of port L2 errors of various methods in the embodiments of the present invention at the same order. Detailed Implementation
[0051] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.
[0052] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.
[0053] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.
[0054] It should be noted that the terms "one" and "more" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".
[0055] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.
[0056] In related technologies, large flexible spacecraft are important platforms for missions such as space exploration, Earth observation, communication relay, and on-orbit servicing. Their main task is to stably perform attitude pointing, payload operation, and maneuver control over long periods in complex space environments. Large flexible spacecraft typically operate in microgravity or zero gravity environments, and to reduce launch mass, increase deployment area, and enhance payload capacity, their solar arrays, thin-film sails, space trusses, flexible antennas, and other accessories often employ lightweight and flexible structures. While this flexibility brings advantages such as lightweight structure and large-scale deployment, it also introduces significant problems such as elastic vibration, low-damped residual vibration, and attitude-structure coupling.
[0057] When large flexible spacecraft perform attitude maneuvers, payload pointing, and orbital maneuvers, their dynamic models inevitably exhibit high-order, strongly coupled, and multimodal characteristics. These include, but are not limited to, low-frequency dense flexible modes in large solar arrays and thin-film sails, local vibration modes in support trusses and flexible antennas, rigid-flexible coupling effects between platform attitude and flexible attachments, and uneven energy distribution and port visibility differences among different flexible structure groups. These factors significantly increase the complexity of the full-order dynamic model, making it difficult to directly design controllers, perform frequency domain analysis, attitude-vibration co-simulation, and online calculations based on high-order accurate models. More critically, large flexible spacecraft may still experience persistent low-damped residual vibrations after maneuvers, which can affect attitude stability, payload line-of-sight pointing accuracy, antenna beam stability, and the structural safety of flexible attachments. Therefore, how to reduce the model order while maintaining the energy distribution, low-damped residual vibration modes, and control port response characteristics of the main flexible attachment structure groups is a crucial issue in the dynamic modeling and control applications of large flexible spacecraft.
[0058] Existing commonly used model reduction methods still have many shortcomings: including but not limited to the difficulty in explicitly answering the retention rate of different large flexible attachments after reduction, the difficulty in explicitly taking into account the risk of low-damped residual vibration of flexible attachments during the reduction process, and the risk of a large number of weak-port, high-energy, low-damped flexible attachment modes being missed.
[0059] To address the problems existing in the aforementioned related technologies, this embodiment provides a method, system, electronic device, and storage medium for automatic order reduction of port-Hamilton structure-preserving automatic order selection models.
[0060] Combination Figure 1 As shown in the embodiment of the present invention, an automatic order selection model reduction method for port-Hamilton structure preservation is provided, comprising:
[0061] Obtain the full-order flexible dynamic model of the large flexible spacecraft, and perform port-Hamilton structure transformation on the full-order flexible dynamic model to obtain the port-Hamilton system;
[0062] The flexible modes in the port-Hamilton system are divided into structural groups to obtain several flexible accessory structural groups.
[0063] The unselected flexible modes are added to the current selected mode set. The energy coverage capture rate and low-damped residual vibration proxy capture rate of each flexible accessory structure group in the current selected mode set are calculated until each flexible accessory structure group meets the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold, and the port error meets the preset condition. Then the current selected mode set is the target selected mode set.
[0064] Based on the selected mode set of the target, a projection matrix is constructed, and the port-Hamilton system is transformed by projection to obtain the target reduced-order model.
[0065] Specifically, this embodiment considers the high-precision dynamic modeling and rapid calculation requirements of large flexible spacecraft in attitude maneuvering, load pointing, and attitude-vibration joint simulation. It takes the high-order flexible dynamic model of the large flexible spacecraft as the processing object and represents it as an equivalent port-Hamilton system. Based on the physical affiliation of the main flexible accessories such as solar array, thin-film sail, support truss, and flexible antenna, the structure group is divided. Furthermore, the energy coverage capture rate and low-damped residual vibration proxy capture rate of the structure group are constructed. This solves the problem that traditional low-frequency truncation, control port-guided order reduction, output-guided order reduction, and general input-output fitting order reduction methods are prone to missing weak port but high-energy, low-damped flexible accessory modes. It achieves effective preservation of the residual vibration characteristics of large flexible structure groups at a lower model order.
[0066] This embodiment addresses the characteristics of flexible appendage structures in large flexible spacecraft. It introduces a minimum coverage constraint for structural groups, ensuring sufficient dynamic representativeness of major flexible structural groups such as solar panels, thin-film sails, trusses, and flexible antennas in the reduced-order model. A low-damped residual vibration proxy is constructed, integrating Hamiltonian energy contribution, control bandwidth correlation, and modal damping ratio, to prioritize the retention of key flexible modes prone to sustained vibration after attitude maneuvers. A step-by-step order selection strategy, prioritizing structural groups and correcting port parameters, is designed. While prioritizing the energy coverage and residual vibration proxy coverage requirements of flexible structural groups, control port contribution modes are further supplemented, enabling the reduced-order model to maintain the port-Hamilton structure while satisfying control port error constraints. This improves the applicability of the reduced-order model in attitude control, residual vibration prediction, and engineering simulation of large flexible spacecraft.
[0067] Optionally, the step of obtaining the full-order flexible dynamic model of the large flexible spacecraft and performing a port-Hamilton structure transformation on the full-order flexible dynamic model to obtain a port-Hamilton system includes:
[0068] Acquire a large flexible spacecraft, which includes a central platform, solar arrays, thin-film sails, deployable trusses, flexible antennas, and locally flexible accessories;
[0069] The central platform, the solar array, the thin-film sail, the deployable truss, the flexible antenna, and the local flexible accessories are processed by finite element modeling, modal synthesis, or equivalent high-order modeling to obtain a full-order flexible dynamic model, wherein the full-order flexible dynamic model includes a generalized flexible coordinate vector and a mass matrix;
[0070] The generalized momentum is obtained by multiplying the mass matrix by the derivative of the generalized flexible coordinate vector;
[0071] The state variables are obtained by combining the generalized flexible coordinate vector and the generalized momentum;
[0072] Based on the state variables, the full-order flexible dynamics model is equivalent to a port-Hamilton form, thus obtaining the port-Hamilton system.
[0073] Specifically, this embodiment obtains a full-order flexible dynamic model containing a generalized flexible coordinate vector and a mass matrix by processing the central platform, solar array, thin-film sail, deployable truss, flexible antenna, and local flexible attachments through finite element modeling, modal synthesis, or equivalent high-order modeling. This allows the reduced-order object to fully cover all flexible components of the large flexible spacecraft. The generalized momentum is obtained by multiplying the mass matrix and the derivative of the generalized flexible coordinate vector. The combination of the generalized flexible coordinate vector and the generalized momentum is used as the state variable, and the second-order mechanical system (full-order flexible dynamic model) is equivalently reconstructed into a first-order form, providing a mathematical basis for the port-Hamilton representation. Based on this, the full-order flexible dynamics model is equivalent to a port-Hamilton form, so that the interconnection structure, dissipation characteristics and energy storage of the system are explicitly represented by the antisymmetric interconnection matrix, the semi-positive definite dissipation matrix and the positive definite energy matrix, respectively. This ensures that the order reduction process is based on mathematical compression of the physical structure rather than simple curve fitting. It provides a unified port-Hamilton framework for the subsequent structural grouping of flexible modes, calculation of energy coverage capture rate and low-damped residual vibration surrogate capture rate and construction of projection matrix. This allows the order reduction model to strictly inherit the energy structure, dissipation structure and port power interpretation of the original system, and can be directly used for attitude control and residual vibration simulation.
[0074] In one embodiment, the large flexible spacecraft comprises a central platform, solar panels, a thin-film sail, a deployable truss, a flexible antenna, and locally flexible attachments. After finite element modeling, modal synthesis, or equivalent higher-order modeling, its linearized flexible dynamics can be represented as the following second-order mechanical system:
[0075] ,
[0076] in, It is a generalized flexible coordinate vector. , , These are the mass matrix, damping matrix, and stiffness matrix, respectively. To control the input distribution matrix, This is the attitude control torque or equivalent control port input. Let:
[0077] ,
[0078] And take the state variable:
[0079] ,
[0080] The Hamiltonian energy function of the system can then be written as:
[0081] ,
[0082] Therefore, equation (1) can be written in port-Hamiltonian form:
[0083] ,
[0084] in, For anti-symmetric interconnection matrix, It is a positive semidefinite dissipation matrix. It is a positive definite energy matrix. Input matrix for port, For port output. Matrix Possible options are:
[0085] ,
[0086] This guarantees , , .
[0087] This form can explicitly characterize the relationships between system energy storage, port power exchange, and damping dissipation.
[0088] Optionally, the process of dividing the flexible modes in the port-Hamilton system into structural groups yields several flexible accessory structural groups, including:
[0089] Based on the spacecraft physical structure, finite element substructure, or modal affiliation of the large flexible spacecraft, the solar array, thin-film sail, deployable truss, flexible antenna, and local flexible attachments in the flexible modes are divided into structural groups and assigned to different structural groups to obtain several flexible attachment structural groups.
[0090] Specifically, this embodiment assigns the flexible modes of the solar array, thin-film sail, deployable truss, flexible antenna, and local flexible attachments to different structural groups according to the spacecraft's physical structure, finite element substructure, or modal affixation. This overcomes the limitation of traditional modal truncation, which relies solely on frequency sorting. After division, each mode has a clear energy affixation and residual vibration risk affixation within its group. This allows for independent calculation of energy capture rate and residual vibration proxy capture rate by structural group, preventing weak but high-energy flexible attachments from being overlooked during the order reduction process. This ensures that all major flexible components retain sufficient dynamic representativeness in the order reduction model.
[0091] In one embodiment, to avoid large flexible attachments being completely missed during the order reduction process, the flexible modes are divided into several structural groups based on the spacecraft's physical structure, finite element substructures, or modal affiliation:
[0092] ,
[0093] in, Indicates the first Each flexible structure group. For large solar array spacecraft, the structure group may include solar array groups, flexible antenna groups, and local accessory groups. After the structure groups are divided, each mode not only has frequency, damping, and port response contributions, but also energy attribution and residual vibration risk attribution relative to each structure group.
[0094] Optionally, the energy coverage capture rate is specifically calculated as follows:
[0095] ,
[0096] In the formula, This represents the energy coverage capture rate of the g-th flexible attachment structure group; This represents the summation of Hamiltonian energy contributions that belong to both the initial selected mode set S and the g-th flexible attachment structure group; This represents the summation of the Hamiltonian energy contributions of all the flexible modes within the g-th flexible attachment structure group.
[0097] Specifically, this embodiment performs modal expansion on the full-order transfer function and calculates the Hamiltonian energy contribution of each flexible mode. Then, it normalizes and sums the energy coverage capture rate according to the structure group. The calculation of energy contribution is refined from the global level to the internal structure group of each flexible accessory, which directly reflects the proportion of energy retained by the selected mode in each structure group. This provides a quantitative criterion for priority selection of the structure group and can explicitly characterize the energy retention rate of the reduced-order model for each large flexible accessory. It avoids the overall omission of weak but high-energy flexible accessories during the reduction process.
[0098] In one embodiment, to ensure that the reduced-order model can still be used for attitude control and port response analysis, this embodiment calculates the contribution of each mode to the input and output responses of the control port within a preset control bandwidth. Let the control bandwidth be... , No. The port response contribution corresponding to each mode can be expressed as:
[0099] ,
[0100] in, Indicates the first The port frequency response components correspond to each mode. After normalizing the contributions of all modes, the port contribution ranking within the control bandwidth can be obtained. This index is mainly used for the second-stage port error correction.
[0101] Furthermore, this embodiment not only calculates the global Hamiltonian energy contribution, but also further calculates the modal energy contribution within each flexible structure group. For the first... A set of structures, whose selected modes The corresponding energy coverage capture rate is defined as:
[0102] ,
[0103] in, Indicates the first The mode in the th ... Hamiltonian energy contribution in the structure group, the first Port frequency response components corresponding to each mode It can be obtained from the modal decomposition of a full-order port-Hamilton system or its equivalent state-space model: for the full-order transfer function After modal expansion, the first The transfer function components corresponding to each mode can be expressed as follows: ,in , , Let be the eigenvalues, right eigenvectors, and left eigenvectors corresponding to the mode. For complex conjugate modes, the synthesis contribution of the conjugate mode pair is taken. Let You can get This embodiment, by setting a minimum energy capture rate threshold for each structure group, can prevent certain weak-port but high-energy flexible attachments from being missed in the low-order model.
[0104] Optionally, the low-damped residual vibration proxy capture rate is specifically calculated as follows:
[0105] Constructing a low-damped residual vibration proxy index ,
[0106] In the formula, Indicates the first Hamiltonian energy contribution of each flexible mode. Indicates the modal damping ratio. It is the smallest positive number. Indicates the first The correlation between a flexible mode and the frequency band of attitude maneuvering or control excitation;
[0107] Based on the low-damped residual vibration proxy index, the low-damped residual vibration proxy capture rate is obtained:
[0108] ,
[0109] In the formula, This represents the low-damping residual vibration surrogate capture rate of the g-th flexible attachment structure group; This represents the sum of residual vibration surrogate energy contributions belonging to the g-th flexible attachment structure group in the initial selected mode set; This represents the total residual vibration surrogate energy contribution within the g-th flexible attachment structure group.
[0110] Specifically, in this embodiment, the product of Hamiltonian energy contribution and control bandwidth correlation is used as the numerator, and the sum of modal damping ratio and small positive numbers is used as the denominator. The division of these two values constructs the surrogate capture rate for low-damped residual vibrations. This surrogate capture rate increases with increasing modal energy and frequency band excitation probability, and decreases with increasing modal damping ratio, directly reflecting the risk level of persistent residual vibrations in each flexible mode after the control input ends. The control bandwidth correlation is determined by the distance between the modal natural frequency and the attitude maneuver or control excitation frequency band. It takes a larger value when the modality is within the frequency band and decreases with increasing distance when it is far from the frequency band, thus assigning higher surrogate weights to high-risk low-damped modes within the frequency band. This index is introduced into the structural group priority order selection, so that the order selection process explicitly considers the risk of low-damped residual vibrations of flexible attachments while taking into account the energy coverage of the structural group. It prioritizes the retention of key modes that are easily excited by maneuvers and decay slowly, solving the problem that traditional order reduction methods cannot maintain the residual vibration characteristics of large flexible spacecraft after attitude maneuvers during the order reduction process.
[0111] In one embodiment, since the residual vibration generated by the large flexible attachment after attitude maneuvering is closely related to the modal energy, damping ratio, and control bandwidth, this embodiment first calculates the low-damping residual vibration proxy index, the expression of which is as follows:
[0112] ,
[0113] in, Indicates the first Hamiltonian energy contribution of each mode, Indicates the modal damping ratio. To avoid small positive numbers with a denominator of zero, Indicates the first Correlation between the first mode and the attitude maneuver or control excitation frequency band: Let the first mode be... The natural frequencies of each mode are The frequency band for attitude maneuvering or control excitation is Then we can define:
[0114] ,
[0115] Therefore, when the modal frequency is within the attitude maneuver or control excitation frequency band, Take the larger value; when the modal frequency is far from this frequency band, It decreases with increasing distance. It is used to characterize the likelihood that a mode is excited by attitude maneuvers or control inputs and participates in the residual vibrations after the maneuver. Indicates the modal damping ratio. To avoid small positive numbers with a denominator of zero, this index is used to characterize the risk of a mode producing persistent residual vibrations after the control input ends.
[0116] Furthermore, for the first The low-damped residual vibration surrogate capture rate of a group of structures is defined as follows:
[0117] ,
[0118] Among them, molecules This represents the sum of residual vibrational surrogate energy belonging to the g-th structural group among the currently selected modes; the denominator is... This represents the sum of residual vibrational surrogate energy for all full-order modes in the g-th structure group.
[0119] By simultaneous constraints and The present invention can balance the energy preservation of the structural group and the retention of low-damped residual vibration modes.
[0120] Optionally, the port error satisfies a preset condition, specifically including:
[0121] Calculate the relative error of the control port frequency domain and the port error quantile value of the intermediate reduced-order model corresponding to the current selected mode set within the control bandwidth;
[0122] If the relative frequency error of the control port is greater than a preset error threshold or the port error quantile is greater than a preset quantile threshold, then according to the order of contribution of each flexible mode to the control port from high to low, the unselected flexible modes are continuously added to the current selected mode set until the relative frequency error of the control port is less than or equal to the preset error threshold and the port error quantile is less than or equal to the preset quantile threshold, and all the flexible accessory structure groups meet the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold.
[0123] Specifically, this embodiment first calculates the improvement range through round-by-round traversal and prioritizes the iterative method of selecting the mode with the largest improvement for the non-compliant structural group. This ensures that the growth of the selected mode set is always directly driven by the energy and residual vibration coverage gap of each structural group. It forces the minimum energy capture rate threshold and the minimum residual vibration surrogate capture rate threshold of all flexible accessory structural groups with a relatively small number of orders, thus avoiding the overall omission of high-energy flexible accessories with weak ports from the order selection mechanism. After completing the structural group coverage, the relative error and error quantile of the port frequency domain in the intermediate reduced-order model within the control bandwidth are calculated. Modes are added in descending order of port contribution until the port error meets the preset threshold. The structural group coverage constraint and the control port accuracy constraint are processed in two stages, ensuring that the reduced-order model first guarantees the residual vibration retention capability of each large flexible accessory, and then meets the frequency domain response accuracy requirements of the attitude control port. This balances accessory vibration characteristics and control simulation usability while maintaining the same number of retained orders.
[0124] In one embodiment, the step-by-step order selection process includes two stages of automatic order selection: structure group priority and port correction, as detailed below:
[0125] (1) Prioritize the selection of the order of the structure group:
[0126] Based on the energy capture rate and residual vibration surrogate capture rate of each structural group, modes that can improve the minimum coverage of the structural group are selected sequentially until each structural group satisfies the following constraint:
[0127] ,
[0128] in, and The first Minimum energy capture rate threshold and minimum residual proxy capture rate threshold for each structure group.
[0129] (2) Port error correction:
[0130] After completing the structural group coverage, if the control port error of the reduced-order model still does not meet the requirements, continue to supplement the flexible modes with larger port contributions until the requirements are met:
[0131] ,
[0132] in, Indicates the frequency domain of the control port Relative error Indicates the port error within the control bandwidth. quantile value and This is the preset error threshold.
[0133] By employing a step-by-step order selection mechanism, we first ensure that no large flexible accessory structure group is missed, and then ensure that the input and output errors of the control port are within an acceptable range, thereby simultaneously taking into account the residual vibration retention of the flexible accessories and the usability of the control simulation.
[0134] Optionally, the step of constructing a projection matrix based on the selected target mode set and performing a projection transformation on the port-Hamilton system to obtain a reduced-order target model includes:
[0135] Based on the selected mode set, a projection matrix is constructed, wherein the projection matrix retains the generalized coordinate component and its corresponding generalized momentum component for each of the flexible modes, so that the Hamiltonian energy function corresponding to the port-Hamilton system after order reduction is kept in the physical form of the sum of kinetic and potential energy.
[0136] The antisymmetric interconnect matrix, positive semidefinite dissipation matrix, positive definite energy matrix and port input matrix in the port-Hamilton system are projected using the projection matrix to obtain the reduced antisymmetric interconnect matrix, positive semidefinite dissipation matrix, positive definite energy matrix and port input matrix.
[0137] Based on the reduced antisymmetric interconnect matrix, the reduced positive semidefinite dissipation matrix, the reduced positive definite energy matrix, and the reduced port input matrix, the reduced port-Hamilton system is reconstructed, and the reduced port-Hamilton system is used as the target reduced model.
[0138] Specifically, this embodiment constructs a projection matrix and retains the generalized coordinate components and generalized momentum components corresponding to each flexible mode in pairs. This ensures that the reduced-order Hamiltonian energy function remains in its physical form as the sum of kinetic and potential energy, avoiding the destruction of the energy structure caused by independently selecting state variables. The projection matrix is then used to project the antisymmetric interconnect matrix, positive semi-definite dissipation matrix, positive definite energy matrix, and port input matrix in the port-Hamilton system. This ensures that the reduced-order system matrix strictly satisfies the antisymmetric, positive semi-definite, and positive definite structural constraints, thus inheriting the energy structure, dissipation structure, and port power interpretation of the original system. The transformed matrix is reconstructed into the reduced-order port-Hamilton system and used as the target reduced-order model. This allows the final low-order model to be directly used for controller design and attitude-vibration co-simulation based on passive theory, while reducing the retained order and improving computational efficiency.
[0139] In one embodiment, after obtaining the selected mode set, this embodiment constructs the projection matrix by preserving generalized coordinates and momentum variables in pairs. Because each flexible mode is based on and The system is entered into the reduced-order model in pairs. After the reduction, the system can still be written in port-Hamiltonian form.
[0140] ,
[0141] in:
[0142] ,
[0143] From the port-Hamilton structure, we know that the reduced-order model satisfies:
[0144] ,
[0145] Therefore, the reduced-order model obtained by this invention can maintain the energy structure, dissipation structure and port power interpretation of the original system, which facilitates subsequent attitude-vibration co-simulation and controller design.
[0146] In a preferred embodiment, to verify and demonstrate the effectiveness of the residual vibration pH method for the structural group designed for a large flexible spacecraft model in this embodiment, numerical simulation verification was performed on a simulation platform. The simulation used a 100th-order dual-solar-truss model, and the control port error constraint was set to... .
[0147] The model covers typical structures such as dual-sided solar panels, deployable trusses, flexible antennas, and locally flexible attachments. Comparison methods include the low-frequency truncation method, the control port method, and the output-directed method. Simulation results show that the structural group residual vibration pH method proposed in this embodiment can achieve large flexible attachment structural group coverage and low-damped residual vibration mode retention at a lower order while maintaining the port-Hamilton structure. Compared with the low-frequency truncation method, the control port method, and the output-directed method, this embodiment can better maintain engineering indicators such as residual vibration RMS, group-level free decay response, key flexible attachment displacement response, and key flexible attachment free decay response under the same retention order. This verifies the effectiveness of the designed method.
[0148] Specifically, Figure 2 The comparison of the retained order under automatic order selection by various methods is shown; Figure 3 The residual vibration RMS error of each method is compared. Figure 4 The comparison of group-level free decay errors for each method is presented; Figure 5 The comparison of displacement response curves of key flexible attachments is shown; Figure 6 The free decay response curves of key flexible accessories are compared. Figure 7 The comparison of port L2 errors of various methods at the same order is shown.
[0149] in, Figure 2The coordinate axes are defined as follows: Horizontal axis (X-axis): Four model order reduction algorithms, in order: residual vibration pH method (the method in this embodiment), control port method, output-guided method, and low-frequency truncation method; Vertical axis (Y-axis): Number of modes retained, representing the total number of modes automatically selected by each algorithm under its own preset error constraints, with a value range of 0~100. As shown in the figure, the residual vibration pH method in this embodiment requires only 41 modes, the control port method 50, the output-guided method 85, and the low-frequency truncation method 82. Under unified engineering error constraints (port L2 error ≤ 0.08, port 95% quantile error ≤ 0.60), the number of modes retained required in this embodiment is far lower than the three traditional comparison algorithms, resulting in optimal model compression efficiency and significantly reducing the computational load for subsequent dynamic simulation and controller design.
[0150] Figure 3 Coordinate axis definition: Horizontal axis (X-axis): 4 comparison algorithms, with... Figure 2 The order is completely consistent; the vertical axis (Y-axis) represents the relative error of residual vibration RMS, characterizing the deviation between the root mean square of the residual vibration amplitude predicted by the reduced-order model after the flexible appendage maneuver and the accurate model of the 100th order. The closer the value is to 0, the higher the accuracy of residual vibration retention, ranging from 0 to 1.2. As shown in the figure, the residual vibration pH method of this embodiment is 0.291983, the control port method is 0.814221, the output guiding method is 0.498306, and the low-frequency truncation method is 0.652821; the residual vibration prediction error of this embodiment is significantly lower. Traditional methods are prone to missing low-damped, high-energy flexible modes, and significantly overestimating / underestimating the residual vibration amplitude; this embodiment prioritizes the retention of dangerous modes by using the low-damped residual vibration proxy capture rate, accurately reproducing the continuous micro-vibrations after spacecraft maneuvering, and meeting the simulation requirements of payload line-of-sight and antenna beam stability.
[0151] Figure 4 The coordinate axis is defined as follows: Horizontal axis (X-axis): four comparison algorithms, in the same order as before; Vertical axis (Y-axis): group-level free attenuation relative error, characterizing the simulation deviation of the free vibration attenuation process of independent flexible structural groups such as solar panels, trusses, and flexible antennas under uncontrolled excitation. A smaller value indicates a higher degree of fidelity in the dynamic characteristics of a single structural group, ranging from 0 to 1.2. As shown in the figure, the residual vibration pH method for this embodiment is 0.330887, the control port method is 0.86611, the output guiding method is 0.508404, and the low-frequency truncation method is 0.65005. This embodiment introduces a minimum energy / residual vibration coverage constraint for the flexible accessory structural group, forcing each physical flexible substructure to retain sufficient modes. This avoids the complete distortion of the vibration attenuation law of a single group caused by traditional algorithms that only focus on the control port response and discard weak port accessory modes. This allows for independent analysis of the vibration evolution characteristics of each flexible accessory.
[0152] Figure 5The coordinate axes are defined as follows: Horizontal axis (X-axis): Time, unit s, simulation duration 0~50s, covering the entire attitude maneuver process and the residual vibration stage after maneuver; Vertical axis (Y-axis): Vibration displacement of key flexible accessories (solar fin / flexible antenna), dimensionless amplitude, range -0.03~0.03; Curve markings in this embodiment:
[0153] Dark blue solid line: 100th order full-order high-precision benchmark model;
[0154] Orange dotted line: Residual vibration pH method reduced-order model of the structure group of this invention;
[0155] Green dotted lines: Control port method;
[0156] Red dashed line: Output-oriented method;
[0157] Purple dotted line: Low-frequency truncation method.
[0158] As shown in the figure, the orange curve in this embodiment highly coincides with the full-order reference curve throughout, and the vibration peak, vibration frequency, and phase are perfectly matched; the output guiding method (red line) has a severely distorted amplitude and a shifted vibration period; the low-frequency truncation and control port methods have obvious amplitude attenuation deviations; this proves that the low-order model generated in this embodiment can accurately reproduce the dynamic displacement of the flexible attachments during attitude maneuvering and simulate the pointing accuracy of the support load.
[0159] Figure 6 The coordinate axis is defined as follows: Horizontal axis (X-axis): Time, unit s, simulation duration 0~50s, simulating the free decay stage without excitation after the control torque is removed; Vertical axis (Y-axis): Free decay vibration amplitude of the key flexible attachment, dimensionless, range -0.10~0.20; The curves in this embodiment are similar to... Figure 5 Completely consistent, with the dark blue solid line representing the full-order baseline. As shown in the figure, after the control torque disappears, only the orange curve in this embodiment closely matches the full-order model, fully restoring the long-term residual vibration of the slowly decaying low-damped mode; the output-guided method (red line) almost completely smooths out all vibrations, completely losing the residual vibration characteristics; the vibration decay rate of the low-frequency truncation and control port methods does not match the real model; thus, the effectiveness of the low-damped residual vibration proxy capture rate of this embodiment is verified, solving the engineering pain point of traditional methods losing long-term residual vibration modes.
[0160] Figure 7 The coordinate axis definitions are as follows: Horizontal axis (X-axis): 4 comparison algorithms, in the same order as before; Vertical axis (Y-axis): Control port L2 relative error, using a logarithmic scale (10⁻¹⁰). -3 ~10 -1The smaller the value, the higher the accuracy of the frequency domain response fitting of the control input-attitude output port. The L2 errors of each algorithm port in the figure are as follows: Control port method: 0.000319 (best port fitting accuracy); Low-frequency truncation method: 0.046497; Output guiding method: 0.049704; Residual vibration pH method of the structure group of this invention: 0.067579.
[0161] Depend on Figure 7 It can be seen that this embodiment adopts a two-stage order selection strategy: the first stage prioritizes ensuring the retention of the flexible attachment structure group and residual vibration modes, and the second stage supplements the port modes to meet the error constraints. Therefore, the port accuracy is slightly weaker than the control port method that only optimizes the ports, but still meets the preset engineering threshold of 0.08; the control port method only pursues port accuracy, at the cost of losing a large number of flexible attachment residual vibration modes ( Figure 3 , Figure 4 (The error increases significantly). This embodiment achieves a two-way balance between port control accuracy and residual vibration fidelity of flexible accessories, resulting in higher overall engineering value.
[0162] In summary, the six sets of simulation diagrams in this embodiment were compared and verified from six dimensions: model compression order, residual vibration amplitude error, single structure group attenuation characteristics, maneuver time-domain displacement, free decay long-term vibration, and control port frequency domain accuracy. Traditional order reduction algorithms can only optimize a single port response or low-frequency mode, and cannot take into account the residual vibration of multiple flexible attachments. The port-Hamilton structure preservation automatic order selection method in this embodiment meets the three major engineering requirements of full coverage of structure groups, preservation of low-damped residual vibration modes, and control port error constraints with the fewest modal orders. It is more suitable for the attitude-vibration joint simulation and controller design scenarios of large flexible spacecraft.
[0163] This invention also provides a port-Hamilton structure-preserving automatic order selection model reduction system, comprising:
[0164] The data acquisition unit is used to acquire the full-order flexible dynamics model of the large flexible spacecraft, and to perform port-Hamilton structure transformation on the full-order flexible dynamics model to obtain the port-Hamilton system.
[0165] The data analysis unit is used to divide the flexible modes in the port-Hamilton system into structural groups to obtain several flexible accessory structural groups.
[0166] The data determination unit is used to supplement the currently selected mode set with the flexible modes that have not been selected, calculate the energy coverage capture rate and low damping residual vibration proxy capture rate of each flexible accessory structure group in the current selected mode set, until each flexible accessory structure group meets the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold, and the port error meets the preset condition, then the current selected mode set is the target selected mode set;
[0167] The data processing unit is used to construct a projection matrix based on the selected target mode set, perform projection transformation on the port-Hamilton system, and obtain a reduced-order target model.
[0168] The port-Hamilton structure-preserving automatic order selection model reduction system of the present invention has the same advantages over the prior art as the aforementioned port-Hamilton structure-preserving automatic order selection model reduction method, and will not be repeated here.
[0169] This invention also provides an electronic device, including: a processor and a memory, wherein the memory is used to store computer programs;
[0170] When the computer program is loaded by the processor, it causes the processor to execute the port-Hamilton structure-preserving automatic order selection model reduction method as described above.
[0171] The electronic device of the present invention has the same advantages over the prior art as the aforementioned port-Hamilton structure-preserving automatic order selection model reduction method over the prior art, and will not be repeated here.
[0172] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the port-Hamilton structure-preserving automatic order selection model reduction method as described above.
[0173] The computer-readable storage medium of the present invention has the same advantages over the prior art as the aforementioned port-Hamilton structure-preserving automatic order selection model reduction method over the prior art, and will not be repeated here.
[0174] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.
Claims
1. A method for order reduction of a port-Hamilton structure-preserving automatic order selection model, characterized in that, The method includes: Obtain the full-order flexible dynamic model of the large flexible spacecraft, and perform port-Hamilton structure transformation on the full-order flexible dynamic model to obtain the port-Hamilton system; The flexible modes in the port-Hamilton system are divided into structural groups to obtain several flexible accessory structural groups. The unselected flexible modes are added to the current selected mode set. The energy coverage capture rate and low-damped residual vibration proxy capture rate of each flexible accessory structure group in the current selected mode set are calculated until each flexible accessory structure group meets the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold, and the port error meets the preset condition. Then the current selected mode set is the target selected mode set. Based on the selected mode set of the target, a projection matrix is constructed, and the port-Hamilton system is transformed by projection to obtain the target reduced-order model.
2. The method for reducing the order of a port-Hamilton structure-preserving automatic order selection model according to claim 1, characterized in that, The process of obtaining a full-order flexible dynamic model of a large flexible spacecraft, and then performing a port-Hamilton structure transformation on the full-order flexible dynamic model to obtain a port-Hamilton system includes: Acquire a large flexible spacecraft, which includes a central platform, solar arrays, thin-film sails, deployable trusses, flexible antennas, and locally flexible accessories; The central platform, the solar array, the thin-film sail, the deployable truss, the flexible antenna, and the local flexible accessories are processed by finite element modeling, modal synthesis, or equivalent high-order modeling to obtain a full-order flexible dynamic model, wherein the full-order flexible dynamic model includes a generalized flexible coordinate vector and a mass matrix; The generalized momentum is obtained by multiplying the mass matrix by the derivative of the generalized flexible coordinate vector; The state variables are obtained by combining the generalized flexible coordinate vector and the generalized momentum; Based on the state variables, the full-order flexible dynamics model is equivalent to a port-Hamilton form, thus obtaining the port-Hamilton system.
3. The method for reducing the order of a port-Hamilton structure-preserving automatic order selection model according to claim 1, characterized in that, The process of dividing the flexible modes in the port-Hamilton system into structural groups yields several flexible accessory structural groups, including: Based on the spacecraft physical structure, finite element substructure, or modal affiliation of the large flexible spacecraft, the solar array, thin-film sail, deployable truss, flexible antenna, and local flexible attachments in the flexible modes are divided into structural groups and assigned to different structural groups to obtain several flexible attachment structural groups.
4. The method for reducing the order of a port-Hamilton structure-preserving automatic order selection model according to claim 1, characterized in that, The energy coverage capture rate is calculated as follows: , In the formula, This represents the energy coverage capture rate of the g-th flexible attachment structure group; This represents the summation of Hamiltonian energy contributions that belong to both the initial selected mode set S and the g-th flexible attachment structure group; This represents the summation of the Hamiltonian energy contributions of all the flexible modes within the g-th flexible attachment structure group.
5. The method for reducing the order of a port-Hamilton structure-preserving automatic order selection model according to claim 4, characterized in that, The calculation process for the low-damped residual vibration proxy capture rate is as follows: Constructing a low-damped residual vibration proxy index , In the formula, Indicates the first Hamiltonian energy contribution of each flexible mode. Indicates the modal damping ratio. It is the smallest positive number. Indicates the first The correlation between a flexible mode and the frequency band of attitude maneuvering or control excitation; Based on the low-damped residual vibration proxy index, the low-damped residual vibration proxy capture rate is obtained: , In the formula, This represents the low-damping residual vibration surrogate capture rate of the g-th flexible attachment structure group; This represents the sum of residual vibration surrogate energy contributions belonging to the g-th flexible attachment structure group in the initial selected mode set; This represents the total residual vibration surrogate energy contribution within the g-th flexible attachment structure group.
6. The method for reducing the order of a port-Hamilton structure-preserving automatic order selection model according to claim 1, characterized in that, The port error meets preset conditions, specifically including: Calculate the relative error of the control port frequency domain and the port error quantile value of the intermediate reduced-order model corresponding to the current selected mode set within the control bandwidth; If the relative frequency error of the control port is greater than a preset error threshold or the port error quantile is greater than a preset quantile threshold, then according to the order of contribution of each flexible mode to the control port from high to low, the unselected flexible modes are continuously added to the current selected mode set until the relative frequency error of the control port is less than or equal to the preset error threshold and the port error quantile is less than or equal to the preset quantile threshold, and all the flexible accessory structure groups meet the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold.
7. The method for reducing the order of a port-Hamilton structure-preserving automatic order selection model according to claim 1, characterized in that, The step of constructing a projection matrix based on the selected target mode set and performing a projection transformation on the port-Hamilton system to obtain a reduced-order target model includes: Based on the selected mode set, a projection matrix is constructed, wherein the projection matrix retains the generalized coordinate component and its corresponding generalized momentum component for each of the flexible modes, so that the Hamiltonian energy function corresponding to the port-Hamilton system after order reduction is kept in the physical form of the sum of kinetic and potential energy. The antisymmetric interconnect matrix, positive semidefinite dissipation matrix, positive definite energy matrix and port input matrix in the port-Hamilton system are projected using the projection matrix to obtain the reduced antisymmetric interconnect matrix, positive semidefinite dissipation matrix, positive definite energy matrix and port input matrix. Based on the reduced antisymmetric interconnect matrix, the reduced positive semidefinite dissipation matrix, the reduced positive definite energy matrix, and the reduced port input matrix, the reduced port-Hamilton system is reconstructed, and the reduced port-Hamilton system is used as the target reduced model.
8. A port-Hamilton structure-preserving automatic order selection model reduction system, characterized in that, include: The data acquisition unit is used to acquire the full-order flexible dynamics model of the large flexible spacecraft, and to perform port-Hamilton structure transformation on the full-order flexible dynamics model to obtain the port-Hamilton system. The data analysis unit is used to divide the flexible modes in the port-Hamilton system into structural groups to obtain several flexible accessory structural groups. The data determination unit is used to supplement the currently selected mode set with the flexible modes that have not been selected, calculate the energy coverage capture rate and low damping residual vibration proxy capture rate of each flexible accessory structure group in the current selected mode set, until each flexible accessory structure group meets the preset minimum energy capture rate threshold and minimum residual vibration proxy capture rate threshold, and the port error meets the preset condition, then the current selected mode set is the target selected mode set; The data processing unit is used to construct a projection matrix based on the selected target mode set, perform projection transformation on the port-Hamilton system, and obtain a reduced-order target model.
9. An electronic device, characterized in that, include: Processor and memory, the memory being used to store computer programs; When the computer program is loaded by the processor, it causes the processor to execute the port-Hamilton structure preservation automatic order selection model reduction method as described in any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the port-Hamilton structure preservation automatic order selection model reduction method as described in any one of claims 1-7.
