Vibration state testing method and system for satellite solar wing assembly

Abstract

The invention relates to the technical field of structural component testing, and specifically to a vibration state testing method and system for satellite solar wing assemblies, comprising: using a vibration testing system to capture the wing edge and central axis vibration signals of each solar wing assembly in real time; analyzing and extracting the edge resonance signal of each assembly and the time series characteristics before reaching resonance, as well as the central axis resonance signal and the corresponding time series variation characteristics; based on the resonance signal samples and the corresponding time series variation characteristics, in conjunction with resonance probability identification information, training a resonance monitoring module, and monitoring and warning the solar wing parallel assembly, thereby solving the technical problem of being unable to identify and prevent potential structural failure and performance loss of the solar wing assembly due to resonance in real time, and realizing the technical effect of accurately measuring the vibration characteristics of the solar wing parallel assembly, warning the potential resonance risk in advance, improving the overall structural stability, and thus strengthening the refined management of the vibration behavior of the solar wing parallel assembly.

Description

Technical Field

[0001] The present invention relates to the technical field related to structural component testing, and in particular to a vibration state testing method and system for satellite solar wing assemblies. Background Art

[0002] The satellite solar wing assembly is a key part of providing power to the satellite. It is usually made of lightweight and flexible materials so that it can be folded and stored and unfolded after launch. Since the satellite solar wing assembly is usually used in an environment with aerodynamic forces, thruster jet impact, etc., the satellite solar wing assembly is prone to vibration. If the satellite solar wing assembly vibrates too much, it may cause structural fatigue damage to the satellite solar wing assembly, thereby affecting its service life.

[0003] In summary, the prior art has the technical problem of being unable to identify and prevent in real time the potential structural failure and performance loss of solar wing assemblies caused by resonance. Summary of the invention

[0004] The present application aims to solve the technical problem in the prior art of being unable to identify and prevent in real time the potential structural failure and performance loss of solar wing assemblies caused by resonance by providing a vibration state testing method and system for satellite solar wing assemblies.

[0005] In view of the above problems, the technical solution to implement this application is:

[0006] In one aspect, the present application provides a vibration state test method for a satellite solar wing assembly, wherein the method comprises: obtaining a solar wing parallel assembly, wherein the solar wing parallel assembly is an assembly structure in which a plurality of solar wing assemblies are arranged in parallel;

[0007] Performing a vibration state test on the solar wing parallel assembly by a vibration test system, and recording a real-time wing vibration signal of each solar wing assembly, wherein the real-time wing vibration signal includes a wing edge vibration signal and a wing mid-axis vibration signal;

[0008] By analyzing the wing edge vibration signal of each solar wing assembly, the wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics before each solar wing assembly reaches the wing edge resonance state are obtained;

[0009] By analyzing the vibration signal of the wing center axis of each solar wing assembly, the wing center axis resonance signal of each solar wing assembly and the center axis-time sequence signal change characteristics before each solar wing assembly reaches the wing center axis resonance state are obtained;

[0010] Based on the wing edge resonance signal, samples of the edge-timing variation characteristics, the wing central axis resonance signal, samples of the central axis-timing signal variation characteristics, and identification information identifying the resonance probability, a resonance monitoring module is trained and embedded in the resonance monitoring module to monitor and warn the solar wing parallel components.

[0011] On the other hand, the present application provides a vibration state test system for a satellite solar wing assembly, wherein the system comprises: an assembly acquisition module, used to acquire a solar wing parallel assembly, wherein the solar wing parallel assembly is an assembly structure in which a plurality of solar wing assemblies are arranged in parallel;

[0012] A vibration state test module, used for performing a vibration state test on the solar wing parallel assembly through a vibration test system, and recording a real-time wing vibration signal of each solar wing assembly, wherein the real-time wing vibration signal includes a wing edge vibration signal and a wing mid-axis vibration signal;

[0013] A resonance signal acquisition module is used to acquire the wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics of each solar wing assembly before reaching the wing edge resonance state by analyzing the wing edge vibration signal of each solar wing assembly;

[0014] A signal analysis module is used to analyze the vibration signal of the wing center axis of each solar wing assembly to obtain the wing center axis resonance signal of each solar wing assembly and the center axis-time sequence signal change characteristics before each solar wing assembly reaches the wing center axis resonance state;

[0015] The monitoring and early warning module is used to train a resonance monitoring module based on the wing edge resonance signal, samples of the edge-timing change characteristics, the wing central axis resonance signal, samples of the central axis-timing signal change characteristics, and identification information identifying the resonance probability, and embed the resonance monitoring module to monitor and warn the solar wing parallel components.

[0016] In summary, one or more technical solutions provided in the present application achieve the technical effect of precisely measuring the vibration characteristics of the solar wing parallel assemblies, providing early warning of potential resonance risks, improving the overall structural stability, and thereby strengthening the refined management of the vibration behavior of the solar wing parallel assemblies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic flow chart of a vibration state test method of a satellite solar wing assembly is provided for this application;

[0018] Figure 2 A flow chart of a feedback optimization resonance monitoring module in a vibration state testing method of a satellite solar wing assembly is provided for this application;

[0019] Figure 3 A structural schematic diagram of a vibration state testing system for a satellite solar wing assembly is provided for this application.

[0020] Explanation of reference numerals: component acquisition module M100, vibration state test module M200, resonance signal acquisition module M300, signal analysis module M400, monitoring and early warning module M500. DETAILED DESCRIPTION

[0021] Embodiment 1

[0022] The present application is described in detail below with reference to the accompanying drawings. Figure 1 As shown, the present application provides a vibration state testing method for a satellite solar wing assembly, wherein the method comprises:

[0023] S1: obtaining a solar wing parallel assembly, wherein the solar wing parallel assembly is an assembly structure in which a plurality of solar wing assemblies are arranged in parallel;

[0024] S2: Performing a vibration state test on the solar wing parallel assembly through a vibration test system, and recording a real-time wing vibration signal of each solar wing assembly, wherein the real-time wing vibration signal includes a wing edge vibration signal and a wing mid-axis vibration signal;

[0025] S3: Analyzing the wing edge vibration signal of each solar wing assembly to obtain the wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics before each solar wing assembly reaches the wing edge resonance state;

[0026] During the long-term monitoring of the vibration status of satellite solar wing assemblies, it was found that coupled vibrations may occur between multiple solar wing assemblies installed in parallel, resulting in an increase in the resonance effect, which is often difficult to suppress in time. However, the existing solar wing vibration test method is relatively extensive and fails to fully consider the resonance effect of satellite solar wing assemblies. There is a potential risk of structural failure of solar wing assemblies caused by resonance.

[0027] Based on this, the present application fully reflects the coupling vibration effect between solar wing assemblies, accurately determines and predicts potential resonance frequencies, thereby providing early warning of potential resonance risks and improving the overall structural stability of satellite solar wing assemblies. Specifically, the present application includes obtaining a solar wing parallel assembly to be tested, wherein the solar wing parallel assembly is composed of a plurality of independent solar wing assemblies arranged in parallel, wherein each solar wing unit is in a fan-shaped structure, and together they constitute a large-area energy collection array. Furthermore, the solar wing assemblies are installed in parallel on the plane of the satellite, and may be installed in a horizontal, vertical or any other suitable direction.

[0028] Based on the solar wing parallel assembly, a high-precision vibration test system is installed, and the vibration test system is configured with multiple vibration test sensors for capturing and recording vibration data of the solar wing parallel assembly under working conditions; the vibration test system is started to perform continuous vibration state tests on each solar wing assembly, and the real-time vibration signals of each solar wing assembly in different time periods are collected and recorded, including the vibration signals of the edge area of ​​the solar wing structure and the vibration signals of the central axis area of ​​the wing structure;

[0029] The vibration waveform corresponding to the wing edge vibration signal of each solar wing assembly is analyzed in frequency domain and time domain (Fourier transform, filtering, peak detection and other signal processing techniques) to obtain the wing edge resonance signal of the edge part of each solar wing assembly in a resonant state; further, based on the identification of the wing edge resonance signal of each solar wing assembly, the characteristic change law of the edge vibration evolution over time before the solar wing assembly enters the edge resonance state is explored, that is, the edge-timing change characteristics, which include the growth rate of the vibration amplitude, frequency drift, harmonic component changes, etc.; further, the vibration behavior of the solar wing parallel assembly is analyzed in detail to quantify the influence of the interaction between the components on the resonance effect, thereby improving the accuracy of the vibration state test of the satellite solar wing assembly and providing data support for preventing harmful resonance of the solar wing assembly under adverse working conditions.

[0030] S4: by analyzing the vibration signal of the wing center axis of each solar wing assembly, obtaining the wing center axis resonance signal of each solar wing assembly and the center axis-time sequence signal change characteristics before each solar wing assembly reaches the wing center axis resonance state;

[0031] S5: Based on the samples of the wing edge resonance signal, the edge-timing change characteristics, the wing central axis resonance signal, the samples of the central axis-timing signal change characteristics and the identification information identifying the resonance probability, a resonance monitoring module is trained, and the resonance monitoring module is embedded to monitor and warn the solar wing parallel components.

[0032] In the same way, the vibration waveform corresponding to the wing mid-axis vibration signal of each solar wing assembly is analyzed in the frequency domain and time domain to obtain the wing mid-axis resonance signal of each solar wing assembly in the resonant state. Further, after identifying the wing mid-axis resonance signal of each solar wing assembly, the characteristic change law of the mid-axis vibration evolution over time before the solar wing assembly reaches the wing mid-axis resonance state is explored, that is, the mid-axis-time series signal change characteristics, which include the growth trend of the vibration amplitude, the frequency migration characteristics, the mode conversion, etc.

[0033] The obtained wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics before each solar wing assembly reaches the wing edge resonance state are sorted to form a sample data set corresponding to the wing edge; the obtained wing mid-axis vibration signal of each solar wing assembly and the mid-axis-time sequence signal variation characteristics before each solar wing assembly reaches the wing mid-axis resonance state are sorted to form a sample data set corresponding to the wing mid-axis;

[0034] A complete training data set is created for the sample data set corresponding to the wing edge, the sample data set corresponding to the wing mid-axis, and the corresponding identification information of the identification resonance probability; machine learning or deep learning technology is used to train with the constructed training data set to learn to predict the resonance probability of the wing edge and the wing mid-axis according to the input vibration signal characteristics, and then the trained resonance monitoring module is embedded into the original solar wing parallel component online monitoring system, so that the original solar wing parallel component online monitoring system has the ability to monitor and warn the vibration state of the wing edge and the wing mid-axis in real time, and immediately send out a warning signal when a state close to resonance is detected;

[0035] During actual operation, the vibration signal data of the solar wing parallel components are read in real time and analyzed in real time through the embedded resonance monitoring module. Once the characteristic changes that meet the preset warning conditions are found, the warning mechanism is triggered, thereby avoiding the potential structural failure risk of the solar wing components due to resonance.

[0036] Furthermore, if Figure 2 As shown, the present application method also includes:

[0037] Obtaining parallel installation positions of the plurality of solar wing assemblies;

[0038] Outputting the distance between two adjacent components of the plurality of solar wing components according to the parallel installation positions;

[0039] The resonance influence operators of the plurality of solar wing components are calculated according to the distance between two adjacent components, and then the resonance monitoring module is feedback optimized according to the resonance influence operator.

[0040] The parallel installation positions of multiple solar wing assemblies are optimized for resonance influence operator calculation and resonance monitoring module feedback. Specifically, the positions of multiple solar wing assemblies that are being installed or have been installed are measured, and the installation coordinate points of each solar wing assembly on the fuselage are accurately recorded using tools such as high-precision three-dimensional coordinate measuring equipment or GPS positioning systems, so as to obtain the parallel installation position information of the multiple solar wing assemblies.

[0041] According to the obtained parallel installation positions of the plurality of solar wing assemblies, for any two adjacent solar wing assemblies, respectively calculate the spatial distance between the center points of the adjacent solar wing assemblies and the key fixing points, so as to determine the physical distance between the two adjacent solar wing assemblies;

[0042] Using vibration theory, a mechanical model is established to calculate the dynamic coupling effect between adjacent components based on the spacing between adjacent solar wing components, the structural parameters of the solar wing (such as mass, stiffness, damping, etc.) and the working environment parameters (such as wind load, temperature change, etc.); based on the coupling effect analysis, a resonance influence operator is defined and calculated, which can quantify the possibility and degree of resonance caused by vibration transmission between adjacent components; the calculated resonance influence operator is imported into the resonance monitoring module, and the vibration propagation weight of the fully connected layer in the resonance monitoring module is updated;

[0043] The above steps optimize the model parameters of the resonance monitoring module according to the resonance influence operator, ensuring that the resonance monitoring module can more accurately identify and predict the potential resonance risks caused by the distance between adjacent components. Furthermore, after the resonance monitoring module performs feedback optimization according to the calculated resonance influence operator, the resonance detection accuracy of the resonance monitoring module is further improved.

[0044] Furthermore, the present application method also includes:

[0045] Performing vector conversion on the parallel installation positions of the plurality of solar wing assemblies, randomly selecting a first solar wing assembly and N solar wing assemblies with N different spacings, wherein N is a positive integer greater than or equal to 2;

[0046] The first solar wing assembly is combined with N solar wing assemblies with different spacing sizes to generate N resonance test groups;

[0047] outputting a first wing vibration signal according to the first solar wing assembly;

[0048] Outputting N resonance test signals according to the N resonance test groups;

[0049] The N resonance test signals are compared with the first wing vibration signal, and a first preset spacing is output, wherein the first preset spacing is a spacing at which the resonance effect on the first wing vibration signal is less than a preset resonance effect.

[0050] The parallel installation positions of the plurality of solar wing assemblies are subjected to relevant test analysis. Specifically, the actual installation positions of all solar wing assemblies are measured, and the parallel installation positions of the plurality of solar wing assemblies are converted into mathematical vector expressions, and the relative position coordinates of each solar wing assembly are recorded. Furthermore, a solar wing assembly is randomly selected as the first solar wing assembly; a series of different spacing values ​​are set, which are recorded as d1 , d2 , …, dN , where N is a positive integer greater than or equal to 2, d1 , d2 , …, dN Will be used to construct N different resonance test groups;

[0051] According to the set N different spacing sizes, N solar wing components are selected in sequence, so that the spacing between the N solar wing components and the first solar wing component is respectively d1 , d2 , …, dN Thus, N resonance test groups are created, each resonance test group includes the first solar wing assembly and a specific spacing ( d1 , d2 , …, dN any one of the above) solar wing components; outputting a vibration signal of the first solar wing component under the condition of no external disturbance, recorded as a first wing vibration signal;

[0052] For each resonance test group in the N resonance test groups, observe the resonance signal of the first solar wing component and a solar wing component with a specific spacing, repeat N times in total, and output N resonance test signals, wherein the N resonance test signals correspond to N resonance test groups, and the first resonance test group in the N resonance test groups is the first solar wing component and the first solar wing component with a specific spacing. d1 The second resonance test group of the N resonance test groups is a solar wing assembly having a spacing of d2 The solar wing assembly of the first solar wing assembly, ..., the Nth resonance test group in the N resonance test groups is the first solar wing assembly and the first solar wing assembly are spaced apart by dN Solar wing components;

[0053] Compare and analyze the first wing vibration signal with N resonance test signals one by one, and judge the degree of resonance influence of the first wing vibration signal when it is combined with solar wing assemblies with different spacings by comparing relevant parameters such as vibration frequency, amplitude, phase, etc.; after comparative analysis, find out the minimum spacing value that makes the resonance influence of the first solar wing assembly less than the preset resonance influence threshold value, and the minimum spacing value that makes the resonance influence of the first solar wing assembly less than the preset resonance influence threshold value is the first preset spacing;

[0054] In short, among N test groups, d1 , d2 , …, dN If the resonance test signal corresponding to a certain spacing value in shows that the resonance effect is small and meets the preset safety threshold, it is determined to be the first preset spacing, indicating that under the first preset spacing condition, the resonance effect between the first solar wing assembly and other solar wing assemblies can be effectively suppressed; the optimal parallel installation spacing between the solar wing assemblies is obtained through the above steps, and further, the layout of the solar wing assemblies is verified and optimized, thereby reducing the potential hazards caused by resonance.

[0055] Furthermore, the resonance influence operator of the plurality of solar wing assemblies is calculated according to the distance between two adjacent assemblies. The method of the present application includes:

[0056] According to the spacing between two adjacent components, the solar wing components whose spacing is greater than the first preset spacing are screened, and the resonance influence of the screened solar wing components is respectively identified, and a plurality of resonance influence operators are output;

[0057] The multiple resonance influence operators are fused, and the fusion result is output as the resonance influence operator of the solar wing parallel assembly.

[0058] The resonance influence operator of the plurality of solar wing assemblies is calculated according to the spacing between two adjacent assemblies. Specifically, the actual installation spacing data between all solar wing assemblies are collected; for each pair of two adjacent assemblies, the distance between the two adjacent assemblies is calculated; the spacing between the adjacent assemblies is compared with the first preset spacing, and the solar wing assemblies whose spacing between all adjacent assemblies is greater than the first preset spacing are screened out. The solar wing assemblies whose spacing between two adjacent assemblies is greater than the first preset spacing can generate a small resonance influence;

[0059] In order to further improve the accuracy of vibration state testing of satellite solar wing assemblies, the tiny resonance effects generated need to be further analyzed. Specifically, for each pair of solar wing assemblies whose spacing between two adjacent assemblies after screening is greater than the first preset spacing, vibration analysis is performed to identify the resonance characteristics between the assemblies at the current spacing; the resonance response of each pair of assemblies at a specific frequency is calculated separately, including amplitude increase; the resonance response of each pair of assemblies at a specific frequency is used as the resonance influence operator of each pair of assemblies; and the relevant attributes of each resonance influence operator are recorded, such as peak resonance frequency, resonance gain coefficient, etc.

[0060] For the multiple resonance influence operators, a fusion algorithm is used for comprehensive processing, such as weighted average, maximum influence principle, matrix operation, etc.; the fusion weight is determined according to the distribution characteristics of the solar wing assembly to ensure that the fusion result can truly reflect the resonance risk level of the entire solar wing parallel assembly; after the fusion is completed, the resonance influence operator corresponding to the entire solar wing parallel assembly is obtained, which improves the overall accuracy of the vibration state test of the satellite solar wing assembly.

[0061] Further, according to the wing edge resonance signal, the samples of the edge-time series variation characteristics, the wing mid-axis resonance signal, the samples of the mid-axis-time series signal variation characteristics, and the identification information identifying the resonance probability, a resonance monitoring module is trained. The method of the present application includes:

[0062] Training an edge resonance network layer according to the wing edge resonance signal, the samples of the edge-time variation feature, and identification information identifying the probability of wing edge resonance;

[0063] Training a central axis resonance network layer according to the wing central axis resonance signal, the samples of the central axis-time series signal change characteristics, and identification information identifying the probability of wing central axis resonance;

[0064] Initialize the weight network layer, and fully connect the weight network layer, the edge resonance network layer and the central axis resonance network layer, and generate a resonance monitoring module after training until convergence.

[0065] According to the wing edge resonance signal, the samples of the edge-time series variation characteristics, the wing mid-axis resonance signal, the samples of the mid-axis-time series signal variation characteristics, and the identification information identifying the resonance probability, a resonance monitoring module is trained. Specifically, for the sample data set corresponding to the wing edge, the sample data set corresponding to the wing mid-axis, and the identification information identifying the wing edge resonance probability, an edge resonance network layer is trained, and a network structure of the edge resonance network layer is set, including an input layer, a hidden layer, and an output layer; the collected wing edge resonance signal and edge-time series variation characteristic samples are used as input data, and the corresponding identification information identifying the wing edge resonance probability is used as a supervision signal; the network parameters are gradually updated by using optimization methods such as back propagation algorithm and gradient descent, so that the model output is as close as possible to the identification information identifying the wing edge resonance probability, until the training loss function reaches a preset convergence condition;

[0066] In the same way, according to the wing mid-axis resonance signal, the samples of the mid-axis-timing signal change characteristics and the identification information identifying the wing mid-axis resonance probability, the mid-axis resonance network layer is trained; a new weight network layer is initialized, and the initialized weight network layer is used to integrate the output information of the edge resonance network layer and the mid-axis resonance network layer; the outputs of the edge resonance network layer and the mid-axis resonance network layer are fully connected to the initialized weight network layer, that is, the resonance probability information obtained by the two parts of independent training are fused; the loss function is set until the overall loss function reaches the convergence standard, indicating that the weight network layer can effectively integrate the information from the edge resonance network layer and the mid-axis resonance network layer, and a resonance monitoring module is generated after training to convergence. The resonance monitoring module will be able to monitor the resonance risk of the entire structure of the solar wing parallel assembly in real time according to the resonance signals and timing change characteristics of the wing edge and the wing mid-axis collected in real time.

[0067] Furthermore, the present application method comprises:

[0068] The edge-timing variation characteristics of each solar wing component before reaching the wing edge resonance state are obtained, wherein the edge-timing variation characteristics include the timing signal amplitude change rate and the timing signal frequency change rate of the wing edge of the solar wing component from a normal state to a resonance state.

[0069] By dividing the vibration amplitude difference between consecutive sampling points by the sampling time interval, the wing edge vibration signal is analyzed in the time domain, and the amplitude change of the wing edge vibration signal over time is calculated, thereby obtaining the amplitude change rate of the time series signal; the fast Fourier transform (FFT), wavelet transform and other methods are used for frequency domain analysis to extract the instantaneous frequency information of the signal, and the change rate of the signal frequency over time, that is, the change rate of the time series signal frequency, is calculated;

[0070] Define and identify the threshold for the wing edge to enter the resonant state. For example, when the amplitude change rate reaches the maximum rate or the frequency change rate shows significant characteristics, it can be preliminarily determined that it is close to the resonant state. In the time window before the solar wing component reaches the resonant state, select the timing signal data in the continuous time period. For the signal in each time period, calculate the amplitude change rate and frequency change rate and record them respectively. Integrate the calculated amplitude change rate and frequency change rate data to form an edge-timing change feature set, which includes the amplitude and frequency change rules of continuous time points in the process of the solar wing gradually approaching the resonant state from the normal working state. In addition, draw the edge-timing change features into an image to intuitively show the evolution process of the wing edge from the normal state to the resonant state through the image, and verify the accuracy by comparing with the principles of engineering mechanics. Through the above steps, the edge-timing change characteristics of each solar wing component before reaching the edge resonance state are obtained to provide support for subsequent vibration state testing.

[0071] Furthermore, the application method includes:

[0072] The central axis-timing signal change characteristics of each solar wing assembly before reaching the wing central axis resonance state are obtained, wherein the central axis-timing signal change characteristics include the axial vibration of the timing signal and the radial vibration of the timing signal of the wing central axis of the solar wing assembly from a normal state to a resonance state.

[0073] By dividing the axial displacement difference of adjacent sampling points by the time interval, the wing mid-axis vibration signal is analyzed in time series, and the rate of change of the vibration amplitude along the solar wing mid-axis direction with time is calculated, which is the axial vibration change rate of the time series signal; in the same way, the rate of change of the vibration amplitude in the radial direction with time is analyzed to obtain the radial vibration change rate of the time series signal;

[0074] According to the natural frequency and dynamic principle of the solar wing assembly, the threshold is set to identify when the axial and radial vibration signals tend to the critical state, that is, close to the resonance condition; in the time interval before the solar wing assembly reaches the central axis resonance state, the timing signal data in the corresponding time period is intercepted, and the axial vibration characteristics and radial vibration characteristics from the normal state to the resonance state are analyzed in detail; the extracted axial vibration change rate data and radial vibration change rate data are combined into the central axis-timing signal change characteristics, which reflects the process of the solar wing assembly gradually transitioning from the normal working state to the resonance state. Track the central axis-timing signal change characteristics of each solar wing assembly before reaching the central axis resonance state to provide support for subsequent vibration state testing.

[0075] Furthermore, after obtaining the wing edge resonance signal and the wing mid-axis resonance signal, the method of the present application further includes:

[0076] According to the wing edge resonance signal and the wing mid-axis resonance signal, testing the edge resonance attenuation index and the wing mid-axis resonance attenuation index of the plurality of solar wing assemblies;

[0077] Weighted calculation of the edge resonance attenuation index and the wing center axis resonance attenuation index, and output of the resonance attenuation index;

[0078] When the resonance attenuation index of any solar wing assembly is less than the preset resonance attenuation index, a first reminder message is output to remind that the vibration reduction performance of the corresponding solar wing assembly is abnormal.

[0079] The high-sensitivity vibration sensor array arranged on the satellite solar wing assembly is used to monitor and capture the vibration signals generated by the edge area and the central axis area of ​​the solar wing in real time when it is in operation; the raw data is filtered to accurately extract the wing edge resonance signal and the wing central axis resonance signal of each solar wing assembly;

[0080] According to the wing edge resonance signal and the wing mid-axis resonance signal, with reference to the edge-time series variation characteristics and the mid-axis-time series signal variation characteristics, the performance of each solar wing assembly in terms of vibration attenuation is quantitatively evaluated, and the edge resonance attenuation index and the wing mid-axis resonance attenuation index of each solar wing assembly are obtained through testing. The edge resonance attenuation index and the wing mid-axis resonance attenuation reflect the ability of the solar wing assembly to attenuate the vibration energy to a stable state after experiencing vibration;

[0081] Furthermore, the edge resonance attenuation index and the wing mid-axis resonance attenuation index of each solar wing assembly are weighted in combination with the importance of each solar wing assembly in the structural stability of the overall solar wing parallel assembly; different weight coefficients are assigned according to the principles of structural dynamics, and the resonance attenuation index corresponding to each solar wing assembly is output through weighted calculation; the resonance attenuation index corresponding to each solar wing assembly is compared with a preset standard threshold: if it is found that the resonance attenuation index of a solar wing assembly is lower than the threshold corresponding to the preset resonance attenuation index, an early warning is triggered and a first reminder message is generated, the first reminder message including the solar wing assembly code corresponding to the solar wing assembly with abnormal vibration reduction performance, which is used to remind relevant technical personnel of the abnormal vibration reduction performance; further, corresponding measures are quickly taken according to the first reminder message, such as adjusting the satellite attitude to change the stress distribution, thereby improving the vibration state of the solar wing assembly.

[0082] In summary, the beneficial effects of the embodiments of the present application are:

[0083] 1. The monitoring accuracy of the vibration status of solar wing components has been improved, especially the ability to accurately capture the resonance characteristics of the edge and the central axis, which is conducive to preventing resonance risks in advance, further reducing potential failures caused by vibration, and extending the service life of solar wing components.

[0084] 2. Dynamically adjust the resonance monitoring module according to the actual installation position and spacing to ensure that the vibration status can be accurately reflected under various working conditions.

[0085] 3. Evaluate the vibration reduction performance of each solar wing assembly in real time. When abnormal vibration reduction performance is detected, issue a warning message immediately to facilitate the adoption of countermeasures and improve the safety and reliability of satellite operation.

[0086] 4. The parallel installation positions of multiple solar wing modules are obtained; according to the parallel installation positions, the distance between two adjacent modules among the multiple solar wing modules is output; according to the distance between two adjacent modules, the resonance influence operator of the multiple solar wing modules is calculated, and then the resonance monitoring module is feedback optimized according to the resonance influence operator. According to the resonance influence operator, the model parameters of the resonance monitoring module are optimized to ensure that the resonance monitoring module can more accurately identify and predict the potential resonance risk caused by the distance between adjacent modules. Furthermore, according to the calculated resonance influence operator, after the resonance monitoring module is feedback optimized, the resonance detection accuracy of the resonance monitoring module is further improved.

[0087] Embodiment 2

[0088] Based on the same inventive concept as the vibration state testing method of the satellite solar wing assembly in the aforementioned embodiment, Figure 3 As shown, an embodiment of the present application provides a vibration state testing system for a satellite solar wing assembly, wherein the system comprises:

[0089] The component acquisition module M100 is used to acquire a solar wing parallel component, wherein the solar wing parallel component is a component structure in which a plurality of solar wing components are arranged in parallel;

[0090] A vibration state test module M200 is used to perform a vibration state test on the solar wing parallel assembly through a vibration test system, and record a real-time wing vibration signal of each solar wing assembly, wherein the real-time wing vibration signal includes a wing edge vibration signal and a wing mid-axis vibration signal;

[0091] The resonance signal acquisition module M300 is used to acquire the wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics of each solar wing assembly before reaching the wing edge resonance state by analyzing the wing edge vibration signal of each solar wing assembly;

[0092] The signal analysis module M400 is used to analyze the vibration signal of the wing center axis of each solar wing assembly to obtain the wing center axis resonance signal of each solar wing assembly and the center axis-time sequence signal change characteristics before each solar wing assembly reaches the wing center axis resonance state;

[0093] The monitoring and early warning module M500 is used to train the resonance monitoring module according to the samples of the wing edge resonance signal, the edge-timing change characteristics, the wing central axis resonance signal, the samples of the central axis-timing signal change characteristics and the identification information identifying the resonance probability, and embed the resonance monitoring module to monitor and warn the solar wing parallel components.

[0094] Furthermore, the satellite solar wing assembly vibration state testing system is also used to perform the following method:

[0095] Obtaining parallel installation positions of the plurality of solar wing assemblies;

[0096] Outputting the distance between two adjacent components of the plurality of solar wing components according to the parallel installation positions;

[0097] The resonance influence operators of the plurality of solar wing components are calculated according to the distance between two adjacent components, and then the resonance monitoring module is feedback optimized according to the resonance influence operator.

[0098] Furthermore, the satellite solar wing assembly vibration state testing system is also used to perform the following method:

[0099] Performing vector conversion on the parallel installation positions of the plurality of solar wing assemblies, randomly selecting a first solar wing assembly and N solar wing assemblies with N different spacings, wherein N is a positive integer greater than or equal to 2;

[0100] The first solar wing assembly is combined with N solar wing assemblies with different spacing sizes to generate N resonance test groups;

[0101] outputting a first wing vibration signal according to the first solar wing assembly;

[0102] Outputting N resonance test signals according to the N resonance test groups;

[0103] The N resonance test signals are compared with the first wing vibration signal, and a first preset spacing is output, wherein the first preset spacing is a spacing at which the resonance effect on the first wing vibration signal is less than a preset resonance effect.

[0104] Furthermore, the resonance influence operators of the plurality of solar wing assemblies are calculated according to the distance between two adjacent assemblies, and the vibration state test system of the satellite solar wing assembly is also used to perform the following method:

[0105] According to the spacing between two adjacent components, the solar wing components whose spacing is greater than the first preset spacing are screened, and the resonance influence of the screened solar wing components is respectively identified, and a plurality of resonance influence operators are output;

[0106] The multiple resonance influence operators are fused, and the fusion result is output as the resonance influence operator of the solar wing parallel assembly.

[0107] Furthermore, the monitoring and early warning module M500 is used to execute the following method:

[0108] Training an edge resonance network layer according to the wing edge resonance signal, the samples of the edge-time variation feature, and identification information identifying the probability of wing edge resonance;

[0109] Training a central axis resonance network layer according to the wing central axis resonance signal, the samples of the central axis-time series signal change characteristics, and identification information identifying the probability of wing central axis resonance;

[0110] Initialize the weight network layer, and fully connect the weight network layer, the edge resonance network layer and the central axis resonance network layer, and generate a resonance monitoring module after training until convergence.

[0111] Furthermore, the satellite solar wing assembly vibration state testing system is also used to perform the following method:

[0112] The edge-timing variation characteristics of each solar wing component before reaching the wing edge resonance state are obtained, wherein the edge-timing variation characteristics include the timing signal amplitude change rate and the timing signal frequency change rate of the wing edge of the solar wing component from a normal state to a resonance state.

[0113] Furthermore, the satellite solar wing assembly vibration state testing system is also used to perform the following method:

[0114] The central axis-timing signal change characteristics of each solar wing assembly before reaching the wing central axis resonance state are obtained, wherein the central axis-timing signal change characteristics include the axial vibration of the timing signal and the radial vibration of the timing signal of the wing central axis of the solar wing assembly from a normal state to a resonance state.

[0115] Further, after obtaining the wing edge resonance signal and the wing mid-axis resonance signal, the satellite solar wing assembly vibration state test system is also used to perform the following method:

[0116] According to the wing edge resonance signal and the wing mid-axis resonance signal, testing the edge resonance attenuation index and the wing mid-axis resonance attenuation index of the plurality of solar wing assemblies;

[0117] Weighted calculation of the edge resonance attenuation index and the wing center axis resonance attenuation index, and output of the resonance attenuation index;

[0118] When the resonance attenuation index of any solar wing assembly is less than the preset resonance attenuation index, a first reminder message is output to remind that the vibration reduction performance of the corresponding solar wing assembly is abnormal.

[0119] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here. Furthermore, the above technical solution only reflects the preferred technical solution of the technical solution of the embodiment of the present application, and some changes that may be made by technicians in this technical field to some parts thereof reflect the new principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A method for testing the vibration state of a satellite solar wing assembly, characterized in that: The method comprises: Obtaining a solar wing parallel assembly, wherein the solar wing parallel assembly is an assembly structure in which a plurality of solar wing assemblies are arranged in parallel; Performing a vibration state test on the solar wing parallel assembly by a vibration test system, and recording a real-time wing vibration signal of each solar wing assembly, wherein the real-time wing vibration signal includes a wing edge vibration signal and a wing mid-axis vibration signal; By analyzing the wing edge vibration signal of each solar wing assembly, the wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics before each solar wing assembly reaches the wing edge resonance state are obtained; By analyzing the vibration signal of the wing center axis of each solar wing assembly, the wing center axis resonance signal of each solar wing assembly and the center axis-time sequence signal change characteristics before each solar wing assembly reaches the wing center axis resonance state are obtained; According to the samples of the wing edge resonance signal and the edge-time series variation characteristics, the wing central axis resonance signal and the central axis-time series signal variation characteristics, and the identification information identifying the resonance probability, a resonance monitoring module is trained, and the resonance monitoring module is embedded to monitor and warn the solar wing parallel components; The method further comprises: Obtaining parallel installation positions of the plurality of solar wing assemblies; Outputting the distance between two adjacent components of the plurality of solar wing components according to the parallel installation positions; The resonance influence operators of the plurality of solar wing components are calculated according to the distance between two adjacent components, and then the resonance monitoring module is feedback optimized according to the resonance influence operator.

2. The method according to claim 1, characterized in that The method further comprises: Performing vector conversion on the parallel installation positions of the plurality of solar wing assemblies, randomly selecting a first solar wing assembly and N solar wing assemblies with N different spacings, wherein N is a positive integer greater than or equal to 2; The first solar wing assembly is combined with N solar wing assemblies with different spacing sizes to generate N resonance test groups; outputting a first wing vibration signal according to the first solar wing assembly; Outputting N resonance test signals according to the N resonance test groups; The N resonance test signals are compared with the first wing vibration signal, and a first preset spacing is output, wherein the first preset spacing is a spacing at which the resonance effect on the first wing vibration signal is less than a preset resonance effect.

3. The method according to claim 2, characterized in that The resonance influence operators of the plurality of solar wing assemblies are calculated according to the distance between two adjacent assemblies, and the method includes: According to the spacing between two adjacent components, the solar wing components whose spacing is greater than the first preset spacing are screened, and the resonance influence of the screened solar wing components is respectively identified, and a plurality of resonance influence operators are output; The multiple resonance influence operators are fused, and the fusion result is output as the resonance influence operator of the solar wing parallel assembly.

4. The method according to claim 1, characterized in that According to the wing edge resonance signal, the samples of the edge-time sequence variation characteristics, the wing mid-axis resonance signal, the samples of the mid-axis-time sequence signal variation characteristics, and identification information identifying the resonance probability, a resonance monitoring module is trained, the method comprising: Training an edge resonance network layer according to the wing edge resonance signal, the samples of the edge-time variation characteristics, and identification information identifying the probability of wing edge resonance; Training a central axis resonance network layer according to the wing central axis resonance signal, the samples of the central axis-time series signal change characteristics, and identification information identifying the probability of wing central axis resonance; Initialize the weight network layer, and fully connect the weight network layer, the edge resonance network layer and the central axis resonance network layer, and generate a resonance monitoring module after training until convergence.

5. The method according to claim 1, characterized in that The edge-timing variation characteristics of each solar wing component before reaching the wing edge resonance state are obtained, wherein the edge-timing variation characteristics include the timing signal amplitude change rate and the timing signal frequency change rate of the wing edge of the solar wing component from a normal state to a resonance state.

6. The method according to claim 1, characterized in that The central axis-timing signal change characteristics of each solar wing assembly before reaching the wing central axis resonance state are obtained, wherein the central axis-timing signal change characteristics include the axial vibration of the timing signal and the radial vibration of the timing signal of the wing central axis of the solar wing assembly from a normal state to a resonance state.

7. The method according to claim 1, characterized in that After acquiring the wing edge resonance signal and the wing mid-axis resonance signal, the method further includes: According to the wing edge resonance signal and the wing mid-axis resonance signal, testing the edge resonance attenuation index and the wing mid-axis resonance attenuation index of the plurality of solar wing assemblies; Weighted calculation of the edge resonance attenuation index and the wing mid-axis resonance attenuation index, and output of the resonance attenuation index; When the resonance attenuation index of any solar wing assembly is less than the preset resonance attenuation index, a first reminder message is output to remind that the vibration reduction performance of the corresponding solar wing assembly is abnormal.

8. A vibration state test system for satellite solar wing components, characterized in that: A method for testing the vibration state of a satellite solar wing assembly according to any one of claims 1 to 7, the system comprising: A component acquisition module, used for acquiring a solar wing parallel component, wherein the solar wing parallel component is a component structure in which a plurality of solar wing components are arranged in parallel; A vibration state test module, used for performing a vibration state test on the solar wing parallel assembly through a vibration test system, and recording a real-time wing vibration signal of each solar wing assembly, wherein the real-time wing vibration signal includes a wing edge vibration signal and a wing mid-axis vibration signal; A resonance signal acquisition module is used to acquire the wing edge resonance signal of each solar wing assembly and the edge-time sequence variation characteristics of each solar wing assembly before reaching the wing edge resonance state by analyzing the wing edge vibration signal of each solar wing assembly; A signal analysis module is used to analyze the wing mid-axis vibration signal of each solar wing assembly to obtain the wing mid-axis resonance signal of each solar wing assembly and the mid-axis-timing signal change characteristics before each solar wing assembly reaches the wing mid-axis resonance state; a monitoring and early warning module is used to train a resonance monitoring module based on the wing edge resonance signal, the samples of the edge-timing change characteristics, the wing mid-axis resonance signal, the samples of the mid-axis-timing signal change characteristics and the identification information identifying the resonance probability, and embed the resonance monitoring module to monitor and warn the solar wing parallel assemblies.

Images