Solid carrier rocket modal identification method based on structural strain
By installing strain gauges on solid rocket motors and using a stochastic subspace algorithm to analyze the strain data, the problem of real-time identification of mode shapes during flight was solved, and efficient elastic stability control was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-12
- Publication Date
- 2026-04-03
AI Technical Summary
Existing modal identification methods cannot accurately identify the mode shapes of solid launch vehicles in real time during flight, leading to difficulties in elastic stability control.
Strain gauges are installed at structural feature points of solid rocket launch vehicles. The strain data is analyzed using a stochastic subspace algorithm to obtain the inherent vibration characteristics, including natural frequencies, mode shapes, and damping ratios. Elastic stability control is then performed in conjunction with a dynamic model.
It enables rapid and accurate identification of the mode shapes of solid-propellant launch vehicles during flight, improving the reliability and accuracy of elastic stability control and reducing implementation costs.
Smart Images
Figure CN121782944A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of aircraft attitude control, specifically relating to a modal recognition method for solid rocket launch vehicles based on structural strain. Background Technology
[0002] Solid-propellant launch vehicles and other aircraft exhibit not only rigid body motion but also elastic motion during flight. This elastic motion can cause aerodynamic servoelasticity problems, interfering with the normal operation of the control system. Therefore, when designing the stability of an aircraft, it is essential to fully consider the inherent vibration characteristics (i.e., modal characteristics) of the aircraft structure. Mode shapes describe the relative deformation of various points on the structure when the aircraft vibrates at a specific natural frequency. Using known mode shapes, components caused by elastic vibration can be precisely "stripped" or "filtered" from the mixed signals measured by gyroscopes, thus obtaining rigid body motion signals that reflect the true trajectory and attitude of the aircraft. Therefore, accurately identifying the mode shapes of a rocket during flight is crucial for the elastic stability control of the rocket.
[0003] To accurately identify mode shapes, a sufficient number of measurement points need to be arranged on the structure to form a complete sensing network. As the main structural component and excitation source of the rocket, the engine's dynamic characteristics are crucial. Because solid rocket engines operate in extreme environments of high temperature, high pressure, and high-speed gas flow during flight, any sensors exposed inside the engine are unlikely to survive. Traditional accelerometers and their mounting bases cannot function properly in this environment, making it impossible to install sensors in these critical locations, thus rendering global mode shape identification impossible from the outset. Furthermore, the excitation during flight is random and cannot be precisely measured, unlike the precisely controllable and measurable exciters (such as hammers and vibrators) used in ground modal tests. Many classic modal parameter identification algorithms (such as the frequency response function method) rely on the synchronous and precise measurement of the input excitation and output response. Under natural excitation, the input excitation is unknown, rendering these methods inapplicable. Therefore, existing modal identification methods suffer from the problem of being unable to accurately identify the mode shapes of solid rocket launch vehicles in flight in real time. Summary of the Invention
[0004] This application provides a modal identification method for solid-propellant launch vehicles based on structural strain, aiming to solve the technical problem that existing technologies cannot accurately identify the mode shapes of solid-propellant launch vehicles in flight conditions in real time, thereby achieving elastic stability control of solid-propellant launch vehicles. In a first aspect, embodiments of this application provide a method for modal identification of solid launch vehicles based on structural strain, including: Strain gauges were installed at structural feature points of the solid rocket launch vehicle, and test cables were connected. Connect the test cable to the data acquisition and analysis system to collect strain data of the solid launch vehicle structure during flight tests or ground tests. Based on strain data, a stochastic subspace algorithm is used to obtain the inherent vibration characteristics of the solid launch vehicle structure, including natural frequencies, mode shapes, and damping ratios. Elastic stability control of solid rocket structure is carried out based on the analysis results of inherent vibration characteristics.
[0005] Before installing strain gauges, the process includes the following steps: constructing a dynamic model of the solid rocket, predicting its mode shapes through simulation calculations, and determining the peak and trough positions of the structural bending mode shapes as characteristic points.
[0006] When installing strain gauges at the feature point locations, the strain gauges are arranged in four rows, located in the four positive quadrants of the rocket structure, and the sensitive direction of each strain gauge is along the axial direction of the rocket structure; the measuring points located at the rocket engine are installed with strain gauges embedded in the structure, and a heat-resistant layer is sprayed on them after installation.
[0007] The random subspace algorithm includes the following steps: A Hankel matrix is constructed based on the strain data sensed by the strain gauges; Construct the Toeplitz matrix from the output covariance matrix; Singular value decomposition is performed on the Toeplitz matrix, which is written as the product of the extended observable matrix and the inverse extended controllable matrix. The state matrix and output matrix of the system are then solved. Perform eigenvalue decomposition on the state matrix A and transform the discrete system into a continuous system. Based on the relationship between the eigenvalues of the continuous-time system state matrix and the system's circular frequency and damping ratio, determine the system's natural frequency, damping ratio, and mode shape. Compare the differences in modal parameters between adjacent order models, eliminate spurious modes, and determine the actual order of the system.
[0008] To determine the system order, the frequency deviation between two adjacent model orders should not exceed 1%, the damping ratio deviation should not exceed 5%, and the mode shape MAC value deviation should not exceed 1%.
[0009] During the results analysis, the natural frequencies identified by the structural strain data are compared with the natural vibration frequencies measured by the gyroscope. When the frequency deviation between the two is less than 2%, the structural strain mode shape is deemed effective and used for the elastic stability control of solid rocket launch vehicles.
[0010] Furthermore, during the results analysis, the inherent vibration characteristics were analyzed at certain time intervals.
[0011] Furthermore, based on the inherent vibration characteristics, a structural modal model is established, the modal slopes are calculated, and the results are used for the elastic stability control of solid rocket launch vehicles.
[0012] Secondly, embodiments of this application provide a solid rocket launch vehicle modal identification system based on structural strain, characterized in that it includes: The product under test is a solid launch vehicle structure. The product under test has multiple structural feature points determined by simulation based on its dynamic model. These structural feature points correspond to the peak and trough positions of the bending vibration mode of the solid launch vehicle. Strain gauges, wherein multiple strain gauges are installed at structural feature points of the product under test, for detecting strain data of the product under test; A test cable, connected to the strain gauge, is used to transmit the strain data; The data acquisition and analysis system is connected to the strain gauge via the test cable and configured to acquire the strain data. Based on the strain data, a random subspace algorithm is used to obtain the inherent vibration characteristics of the solid rocket structure, including natural frequency, mode shape, and damping ratio. Based on the analysis results of the inherent vibration characteristics, elastic stability control is performed on the solid rocket structure.
[0013] Furthermore, the data acquisition and analysis system is configured to execute the following steps of the random subspace algorithm: A Hankel matrix is constructed based on the strain data sensed by the strain gauges; Construct the Toeplitz matrix from the output covariance matrix; Singular value decomposition is performed on the Toeplitz matrix, which is written as the product of the extended observable matrix and the inverse extended controllable matrix. The state matrix and output matrix of the system are then solved. Perform eigenvalue decomposition on the state matrix A and transform the discrete system into a continuous system. Based on the relationship between the eigenvalues of the continuous-time system state matrix and the system's circular frequency and damping ratio, determine the system's natural frequency, damping ratio, and mode shape. Compare the differences in modal parameters between adjacent order models, eliminate spurious modes, and determine the actual order of the system.
[0014] Compared with the prior art, this application has the following advantages: Currently, there is no reliable flight mode identification technology for solid-propellant launch vehicles in China. This invention proposes to identify the flight modes of solid-propellant launch vehicles through structural strain identification, which has certain advanced technology. This invention proposes to conduct modal testing by embedding strain gauges inside the solid engine. Data acquisition can be uniformly executed by the solid-propellant launch vehicle control system or separately executed by external testing equipment, resulting in low implementation cost, good integration effect, and high technical reliability. The random subspace algorithm can directly process the strain time-domain data acquired by the system, which has advantages such as strong noise resistance, high identification accuracy, and good computational stability, enabling rapid and accurate identification of the dynamic parameters of the solid-propellant launch vehicle structure. The modal parameters identified by structural strain can be compared with the natural frequencies sensitive by the solid-propellant launch vehicle gyroscope for confirmation, which can further improve the accuracy of the identification results and has good practical value. Attached Figure Description
[0015] Figure 1 This is a schematic diagram of the solid rocket modal recognition system based on structural strain according to the present invention; Figure 2 This is a flowchart illustrating the modal identification method for solid launch vehicles based on structural strain according to the present invention. Figure 3 This is a schematic diagram of the random subspace algorithm of the present invention.
[0016] In the diagram: 1. Product under test; 2. Strain gauge; 3. Test cable; 4. Data acquisition and analysis system. Detailed Implementation To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.
[0017] In a first aspect, embodiments of this application provide a method for modal identification of solid-propellant launch vehicles based on structural strain. As shown in Figure 1, which is a schematic diagram of the solid-propellant launch vehicle structure of this invention, its components include a test product 1, a strain gauge 2, a test cable 3, and a data acquisition and analysis system 4. The product under test is generally the entire structure of a solid-propellant launch vehicle, including solid rocket engines and other module mechanisms, internal components, etc. The strain gauge is a general-purpose product, typically a resistance strain gauge, but it can also be a semiconductor strain gauge or a fiber optic strain sensor, etc. The test cable refers to a strain mode test cable, which can be integrated into the cable network of the solid rocket launch vehicle control system. The data acquisition and analysis system is a dedicated testing device for the strain modes of solid launch vehicles. It is used to collect and store strain data, analyze the results, obtain modal characteristic parameters, and transmit the results to the central computer. The data acquisition and analysis system can also be integrated into the central computer of solid launch vehicles for integrated design.
[0018] The present invention discloses a modal identification method for solid-propellant launch vehicles based on structural strain. During the flight of the solid-propellant launch vehicle, strain gauges are used to acquire the strain response at structural feature locations in real time. The data is transmitted to a data acquisition and analysis system via a test cable. A random subspace algorithm is used to acquire the natural vibration frequencies and mode shapes of the structure in real time, and this information is fed back to a central computer for elastic stability control. The method specifically includes the following steps: S1. Install strain gauges at key structural features of the product, connect test cables, and conduct effectiveness tests on the strain gauges on the ground to ensure accurate response; this is typically done by tapping.
[0019] Specifically, in a laboratory or ground environment, strain gauges are functionally tested using simulated mechanical excitations (such as impact) to verify their ability to correctly sense and output structural strain signals. This is a quality control measure to ensure that each sensor responds accurately before being put into actual use, avoiding data distortion or control failure due to sensor malfunction.
[0020] Before step S1, a dynamic model of the solid rocket is constructed, simulation calculations are performed, its approximate mode shape is determined, and characteristic points are planned. Strain sensors should be planned and arranged at the peak and trough positions of the structural bending mode.
[0021] Specifically, before installing strain gauges, a dynamic model of the rocket is constructed using computer simulation technology to predict its vibration characteristics (i.e., mode shapes), thereby scientifically selecting the optimal installation location for the strain gauges. Specifically, the focus is on identifying the points of maximum deformation in the rocket structure during bending vibration (such as crests and troughs), as these locations are most sensitive to strain changes and can effectively capture modal data. This is an optimization strategy of "simulation-driven experimentation," aiming to improve the accuracy and efficiency of modal identification.
[0022] Strain gauges are installed at the determined structural feature points of the product. There are four rows of strain gauge sensors, which are generally arranged according to the positive quadrant of the solid rocket launch vehicle (corresponding to the pitch and yaw directions of the spacecraft). The sensitive direction of the strain gauges is along the axial direction of the solid rocket launch vehicle structure (i.e., the nose-to-tail direction). Strain gauges should be embedded in the measuring points at the solid rocket motor locations, and then a heat-resistant layer should be sprayed on.
[0023] Specifically, by symmetrically arranging four rows of axially sensitive strain gauges around the rocket body, the tensile and compressive strains at symmetrical positions during structural bending can be measured simultaneously, directly reflecting the deformation distribution positively correlated with displacement. This design ensures the integrity of the modal identification data source, laying the foundation for subsequently deriving the actual vibration mode model of the rocket from the strain data using a random subspace algorithm. Because solid rocket motors generate extremely high temperatures during operation, the surface temperature of their outer shell far exceeds the tolerance limit of ordinary strain gauges. Therefore, the strain gauges here need to be "embedded," meaning they are first installed or embedded in the inner layer or a specific structural layer of the engine casing, and then a thick heat-resistant layer (such as a thermal insulation coating or ablative material) is sprayed onto the outer surface. This ensures the sensor's functionality while isolating it from the high-temperature environment, guaranteeing its survival and normal operation during flight.
[0024] When a structure undergoes bending deformation, the strain on one side is tensile strain, and the strain on the other side is compressive strain. The structural strain is positively correlated with the displacement. According to the dynamic system equations of an n-degree-of-freedom structure...
[0025] Where M is the mass matrix, C is the damping matrix, K is the stiffness matrix, f(t) is the external excitation vector, and u(t) is the displacement vector. Based on the relationship between strain and displacement u during structural bending deformation... Let B be the strain matrix, from which we can obtain the structural dynamics equations expressed in terms of strain.
[0026] Based on the similarity between the two formulas above, given the known structural strain matrix, modal data of solid rocket launch vehicles can be effectively obtained by collecting the strain data of the structure.
[0027] S2. Connect the test cable to the data acquisition and analysis system to collect the structural strain characteristics during flight or ground testing; Specifically, strain modal testing cables can be integrated into the cable network of a solid-propellant launch vehicle control system. Solid-propellant launch vehicles will generate elastic vibrations under external excitation or natural flight excitation, causing structural deformation. Strain gauges can sense the amount of structural deformation in real time. Through a data acquisition and analysis system, strain data can be acquired, stored, and arranged according to time series. The data can be preprocessed to remove bad data. Data acquisition can be performed uniformly by the solid-propellant launch vehicle control system or separately by external testing equipment to obtain the corresponding data results.
[0028] S3. The inherent vibration characteristics of the aircraft structure, including natural frequency, mode shape and damping ratio, are obtained in real time through the random subspace algorithm; The random subspace method is based on the time-domain data of strain data sensed by strain gauges. Arranged in chronological order, a Hankel matrix is constructed. Singular value decomposition (SVD) is then used to obtain the system's eigenvalues and eigenvectors, thereby determining the modal parameters. In the scheme provided in this application embodiment, the identification of solid rocket launch vehicle flight modal parameters includes five steps: constructing a Hankel matrix, constructing a Toeplitz matrix, singular value decomposition, modal parameter identification, eliminating spurious modes, and determining the system order. Specifically: 1) Constructing the Hankel matrix: Based on the time-domain data of the strain data sensed by the strain gauge, a Hankel matrix is constructed. The Hankel matrix, consisting of 2i+j-1 output vectors, is defined as follows:
[0029] The collected data is divided into two parts: the first part is called "past output," and the second part is called "future output." The constructed matrix is as follows:
[0030] Where Y is the output vector matrix; This is the k-th output vector.
[0031] The acquired strain time-domain data is divided into "past output" and "future output" blocks, primarily to construct the Hankel matrix, thereby efficiently applying the stochastic subspace algorithm to estimate the dynamic characteristics of the solid rocket launch vehicle structure. This block-based approach leverages the temporal correlation of time-series data: "past output" represents historical response data, used to initialize system state estimation and capture the initial vibration behavior of the structure; while "future output" represents future response data, used to verify the accuracy and consistency of state predictions. Through this division, the algorithm can extract the system's state-space model from strain data generated by natural excitations (such as airflow or engine vibration during flight) without relying on known input excitations, thus significantly improving noise resistance and recognition accuracy. This method is particularly suitable for scenarios involving dynamic mass changes during solid rocket launch vehicle flight, ensuring the real-time performance and accuracy of modal identification.
[0032] 2) Constructing the Toeplitz matrix: The Toeplitz matrix is composed of a series of output covariance matrices. In actual testing, only a finite number of data points are typically obtained; therefore, its output covariance matrix is...
[0033] Constructing the Toeplitz matrix from the covariance matrix
[0034]
[0035] The core function of the covariance matrix is to act as a data purifier. It effectively filters out random noise in the time-domain strain data through statistical averaging and reveals the intrinsic correlation between response signals at different times, thereby extracting the essential features determined by the inherent dynamic characteristics of the structure. On this basis, the purpose of constructing the Toeplitz matrix from the covariance matrix is to build a mathematical bridge. Its unique diagonal constant structure can systematically encapsulate this purified covariance information. Then, through techniques such as singular value decomposition, the output data can be converted into a state-space model that can describe the dynamics of the system. Ultimately, this enables high-precision and noise-resistant identification of the rocket's time-varying modal parameters (natural frequency, mode shape, and damping ratio), providing reliable input for elastic stability control.
[0036] 3) Singular Value Decomposition: Perform singular value decomposition on the Toeplitz matrix.
[0037]
[0038]
[0039] Where U is the left singular vector matrix; S is the singular value matrix; and V is the right singular vector matrix.
[0040] The Toeplitz matrix can be written as the product of the extended observable matrix and the inversely extended controllable matrix.
[0041] in , The covariance between any two adjacent output vectors and the same output vector differs by only one state matrix A. Therefore, the state matrix A can be solved by the covariance between two adjacent sets of output vectors, i.e.
[0042] Therefore, by performing singular value decomposition on the first Toeplitz matrix, and then using the constructed second Toeplitz matrix, the state matrix A can be obtained:
[0043] The output matrix C can be obtained from the observable matrix.
[0044] Where G is the control matrix.
[0045] The state matrix A and output matrix C play a central role in the modal identification of solid rocket motors based on structural strain. State matrix A encapsulates the system's inherent dynamic characteristics (such as the coupling relationship between mass, stiffness, and damping). By performing eigenvalue decomposition on A, key modal parameters of the system, including natural frequencies, damping ratios, and mode shapes, can be directly derived. Output matrix C establishes the mapping relationship between the system's internal states and external strain observations, ensuring that the identified modes accurately correspond to the actual measurement data. These matrices are obtained from the strain time-domain data using stochastic subspace algorithms (such as constructing the Toeplitz matrix and singular value decomposition), ultimately providing real-time and accurate dynamic model inputs for elastic stability control.
[0046] 4) Modal parameter identification: Perform eigenvalue decomposition on the previously obtained system state matrix A:
[0047] The discrete system is then transformed into a continuous system. The relationship between the eigenvalues of the continuous-time system state matrix Ac and the system's angular frequency and damping ratio is as follows:
[0048] At this point, the system's frequency, damping ratio, and mode shape have all been determined.
[0049] Where λ is the eigenvalue of the continuous-time system state matrix; ω is the damping ratio; j is the angular frequency; and j is the imaginary unit.
[0050] 5) Eliminating spurious modes and determining the system order: First, damping ratio is used to initially eliminate spurious modes. Under a certain modal state, the modal parameters identified by the higher-order model are compared with those identified by the lower-order model. If the differences in characteristic frequency, damping ratio, and mode shape are less than the preset limit values, the corresponding mode is the mode of the structural system. Generally, it is required that the frequency deviation between two adjacent model orders is not greater than 1%, the damping ratio deviation is not greater than 5%, and the mode shape MAC value deviation is not greater than 1%.
[0051] In modal identification of solid-propellant launch vehicles based on structural strain, eliminating spurious modes and determining the system order are crucial steps to ensure the accuracy and reliability of modal parameter identification results. Because the stochastic subspace algorithm introduces spurious modes during the solution process due to noise, numerical calculation errors, or system nonlinearity, these spurious modes do not correspond to the actual vibration characteristics of the structure. Directly using these spurious modes would lead to misjudgments in the elastic stability control of the solid-propellant launch vehicle. By setting scientific criteria (such as requiring that the frequency deviation between two adjacent model orders is no greater than 1%, the damping ratio deviation is no greater than 5%, and the mode shape MAC value deviation is no greater than 1%), the system can automatically identify and eliminate these spurious components. This allows for the accurate determination of the minimum system order required to describe the structural dynamics, ultimately ensuring that the identified natural frequencies, mode shapes, and damping ratios are accurate and effective, providing a reliable basis for the precise design of the spacecraft control system.
[0052] S4. Apply the results of the inherent vibration characteristic analysis to the elastic stability control of solid rocket structures.
[0053] The analysis results of inherent vibration characteristics obtained based on structural strain data are transmitted to the central computer. During the result analysis, due to the dynamic changes in mass caused by the combustion of solid rocket fuel, and since the natural frequencies of the structure are directly related to its mass and stiffness, vibration characteristic analysis is performed at 2-second intervals to obtain the actual natural vibration frequencies of the structure. This ensures that the acquired modal frequencies accurately reflect the current structural characteristics. The frequencies obtained through structural strain can be further compared and confirmed with the natural vibration frequencies sensed by the gyroscope. When the frequency difference is less than 2%, the structural strain mode shapes can be used for elastic stability control to ensure the accuracy of the results.
[0054] Specifically, based on the strain gauge installation location and strain mode shape analysis results at characteristic frequencies, an actual structural mode shape model of the solid rocket launch vehicle is established, and the mode shape slope is calculated. This result is then used for elastic stability control. During the stability control of the solid rocket launch vehicle, it is necessary to analyze and separate the rigid body motion and structural elastic vibration deformation under natural frequency vibration. After eliminating the motion components caused by elastic vibration, the gyroscope feedback results represent the actual motion of the solid rocket launch vehicle, which is used to calculate motion information such as the trajectory and attitude of the solid rocket launch vehicle.
[0055] In other words, the random subspace algorithm not only yields the frequency, but more importantly, it provides a set of data describing the relative deformation at each measuring point (strain gauge location) on the overall rocket structure at a given natural frequency. Fitting this data into a continuous curve yields the actual modal model of the rocket at that frequency. Calculating the slope at any point on the modal curve provides the modal slope. Its physical significance lies in quantitatively characterizing the angle of bending deformation of the structure at a specific frequency at that location. This key parameter provides a crucial geometric relationship for subsequent elastic stability control. During stability control, the core challenge lies in the fact that the signal fed back by the gyroscope is a mixture of rigid body motion and structural elastic vibration deformation. To address this issue, the system utilizes the identified real-time mode shapes and their slopes to construct an accurate mathematical model. Based on this, advanced observers or filtering algorithms are designed to estimate and eliminate the interfering motion components caused by elastic vibration from the gyroscope's raw mixed signal in real time. After this signal purification, the result fed back by the gyroscope can truly reflect the rigid body motion of the solid rocket. This purified attitude information is the only reliable basis for calculating the precise trajectory and attitude of the solid rocket and generating reliable control commands, fundamentally avoiding harmful coupling between the control loop and the elastic modes, and ensuring flight stability and accuracy.
[0056] Secondly, embodiments of this application also provide a solid rocket modal recognition system based on structural strain.
[0057] Reference Figure 1 , Figure 1 This is a schematic diagram of a solid rocket modal identification system based on structural strain according to the present invention. The system includes: The product under test 1 is generally the entire structure of a solid-propellant launch vehicle, including solid rocket engines and other module mechanisms, internal single units, etc. The product under test 1 has multiple structural feature points determined by simulation based on its dynamic model, and the structural feature points correspond to the peak and trough positions of the bending vibration mode of the solid-propellant launch vehicle. Strain gauge 2 is a general-purpose product, typically a resistance strain gauge, but it can also be a semiconductor strain gauge or a fiber optic strain sensor, etc.; multiple strain gauges 2 are installed at structural feature points of the product under test 1 to detect the strain data of the product under test 1. Test cable 3 is a dedicated strain mode test cable, which can be integrated into the cable network of the solid rocket launch vehicle control system; it connects to the strain gauge 2 and is used to transmit the strain data; The data acquisition and analysis system 4 is a dedicated testing device for the strain modes of solid launch vehicles. It is used to acquire and store strain data, analyze the results, obtain modal characteristic parameters, and transmit the results to the central computer. The data acquisition and analysis system can also be integrated into the central computer of solid launch vehicles for integrated design.
[0058] The data acquisition and analysis system 4 is connected to the strain gauge 2 via the test cable 3, and is configured to acquire the strain data and analyze the strain data based on the random subspace algorithm to obtain the inherent vibration characteristics of the product under test 1. The inherent vibration characteristics include natural frequency, mode shape and damping ratio. The data acquisition and analysis system 4 performs elastic stability control on the solid rocket structure based on the analysis results of the inherent vibration characteristics.
[0059] Furthermore, the data acquisition and analysis system 4 is configured to perform the following steps by executing the random subspace algorithm: Construct the Hankel matrix based on the strain signal; Construct the Toeplitz matrix from the output covariance matrix; Singular value decomposition is performed on the Toeplitz matrix to solve for the system's state matrix and output matrix; The state matrix is decomposed into eigenvalues, and the discrete system is transformed into a continuous system to obtain the natural frequency, damping ratio, and mode shape. Compare the differences in modal parameters between adjacent order models, eliminate spurious modes, and determine the system order; The functional implementation of each component in the above-mentioned solid rocket modal identification system based on structural strain corresponds to the steps in the above-mentioned solid rocket modal identification method based on structural strain. Their functions and implementation processes will not be described in detail here.
[0060] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A modal identification method for solid launch vehicles based on structural strain, characterized in that, Includes the following steps: Strain gauges were installed at structural feature points of the solid rocket launch vehicle, and test cables were connected. Connect the test cable to the data acquisition and analysis system to collect strain data of the solid launch vehicle structure during flight tests or ground tests. Based on strain data, a stochastic subspace algorithm is used to obtain the inherent vibration characteristics of the solid launch vehicle structure, including natural frequencies, mode shapes, and damping ratios. Elastic stability control of solid rocket structure is carried out based on the analysis results of inherent vibration characteristics.
2. The method for modal identification of solid rocket launch vehicles based on structural strain as described in claim 1, characterized in that, Before installing strain gauges, the following steps are also included: constructing a dynamic model of the solid rocket, predicting its mode shape through simulation calculations, and determining the peak and trough positions of the bending mode shape of the solid rocket structure as characteristic points.
3. The method for modal identification of solid rocket launch vehicles based on structural strain as described in claim 2, characterized in that, When installing strain gauges at the feature point locations, the strain gauges are arranged in four rows, located in the four positive quadrants of the rocket structure, and the sensitive direction of each strain gauge is along the axial direction of the rocket structure; the measuring points located at the rocket engine are installed with strain gauges embedded in the structure, and a heat-resistant layer is sprayed on them after installation.
4. The method for modal identification of solid rocket launch vehicles based on structural strain as described in claim 1, characterized in that, The random subspace algorithm includes the following steps: A Hankel matrix is constructed based on the strain data sensed by the strain gauges; Construct the Toeplitz matrix from the output covariance matrix; Singular value decomposition is performed on the Toeplitz matrix, which is written as the product of the extended observable matrix and the inverse extended controllable matrix. The state matrix and output matrix of the system are then solved. Perform eigenvalue decomposition on the state matrix A and transform the discrete system into a continuous system. Based on the relationship between the eigenvalues of the continuous-time system state matrix and the system's circular frequency and damping ratio, determine the system's natural frequency, damping ratio, and mode shape. Compare the differences in modal parameters between adjacent order models, eliminate spurious modes, and determine the actual order of the system.
5. The method for modal identification of solid rocket launch vehicles based on structural strain as described in claim 4, characterized in that, To determine the system order, the frequency deviation between two adjacent model orders should not exceed 1%, the damping ratio deviation should not exceed 5%, and the mode shape MAC value deviation should not exceed 1%.
6. The method for modal identification of solid rocket launch vehicles based on structural strain as described in claim 1, characterized in that, During the results analysis, the natural frequency identified by the structural strain data is compared with the vibration frequency sensitively measured by the gyroscope. When the frequency deviation between the two is less than 2%, the structural strain mode shape is deemed effective and used for the elastic stability control of solid rocket launch vehicles.
7. The method for modal identification of solid launch vehicles based on structural strain as described in claim 6, characterized in that, The inherent vibration characteristics are analyzed at certain time intervals.
8. The method for modal identification of solid rocket launch vehicles based on structural strain as described in claim 6, characterized in that, Based on the inherent vibration characteristics, a structural modal model is established, the mode slope is calculated, and the results are used for the elastic stability control of solid rocket launch vehicles.
9. A modal identification system for solid rocket launch vehicles based on structural strain, characterized in that, include: The product under test is a solid launch vehicle structure. The product under test has multiple structural feature points determined by simulation based on its dynamic model. These structural feature points correspond to the peak and trough positions of the bending vibration mode of the solid launch vehicle. Strain gauges, wherein multiple strain gauges are installed at structural feature points of the product under test, for detecting strain data of the product under test; A test cable, connected to the strain gauge, is used to transmit the strain data; The data acquisition and analysis system is connected to the strain gauge via the test cable and configured to acquire the strain data. Based on the strain data, a random subspace algorithm is used to obtain the inherent vibration characteristics of the solid rocket structure, including natural frequency, mode shape, and damping ratio. Based on the analysis results of the inherent vibration characteristics, elastic stability control is performed on the solid rocket structure.
10. The solid rocket motor mode recognition system based on structural strain as described in claim 9, characterized in that, The data acquisition and analysis system is configured to execute the random subspace algorithm in the following steps: A Hankel matrix is constructed based on the strain data sensed by the strain gauges; Construct the Toeplitz matrix from the output covariance matrix; Singular value decomposition is performed on the Toeplitz matrix, which is written as the product of the extended observable matrix and the inverse extended controllable matrix. The state matrix and output matrix of the system are then solved. Perform eigenvalue decomposition on the state matrix A and transform the discrete system into a continuous system. Based on the relationship between the eigenvalues of the continuous-time system state matrix and the system's circular frequency and damping ratio, determine the system's natural frequency, damping ratio, and mode shape. Compare the differences in modal parameters between adjacent order models, eliminate spurious modes, and determine the actual order of the system.