CN111444648A - Method for quickly estimating structural dynamic characteristics of modular spacecraft - Google Patents
Method for quickly estimating structural dynamic characteristics of modular spacecraft Download PDFInfo
- Publication number
- CN111444648A CN111444648A CN202010212207.0A CN202010212207A CN111444648A CN 111444648 A CN111444648 A CN 111444648A CN 202010212207 A CN202010212207 A CN 202010212207A CN 111444648 A CN111444648 A CN 111444648A
- Authority
- CN
- China
- Prior art keywords
- model
- spacecraft
- module
- simplified model
- refined
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for quickly estimating the structural dynamic characteristics of a modular spacecraft, and belongs to the field of aircraft structure analysis. The invention aims to solve the problems of insufficient calculation precision and long time consumption of the existing method for estimating the structural dynamic characteristics of the modular spacecraft, and provides a method for quickly estimating the structural dynamic characteristics of the modular spacecraft. The method takes a modularized spacecraft as an object, and realizes the rapid estimation of the structural dynamic characteristics of the spacecraft configuration to be estimated including the natural frequency and the mode vibration type on the premise of knowing the splicing combination mode of the spacecraft configuration module to be estimated, the number of the formed modules and the pretightening force of the butt joint device.
Description
Technical Field
The invention relates to a method for quickly estimating the structural dynamic characteristics of a modular spacecraft, and belongs to the field of aircraft structure analysis.
Background
In order to solve the outstanding problems of long task response time, low component reuse rate and the like of the traditional spacecraft, the design idea of the modular spacecraft is provided, and the modular spacecraft separates the components and subsystems of the satellite to form a standardized module which can be formed by multiple times of launching and on-orbit assembly. Kortmann et al analyzed the system composition, design concept, and advantages of a modular spacecraft over conventional spacecraft in detail.
When the modularized spacecraft runs in orbit, the spacecraft module can realize movement and module splicing combination by means of the mechanical arm and the standard docking device, so that spacecraft configurations with various module compositions and different splicing modes are formed, and the requirements of different types of space in-orbit tasks are met. Adomeit et al describe a technical approach for modular spacecraft to form different spacecraft configurations in space.
Compared with the traditional spacecraft, the modularized spacecraft can be designed to obtain various spacecraft configurations, and the iteration and the cyclicity in the design stage are more prominent. In the design of the modular spacecraft, the reasonability of a configuration design scheme of the modular spacecraft needs to be judged by combining the structural dynamic characteristics of the configuration including the natural frequency and the mode shape. Local nonlinear factors such as contact, friction and the like are generated in the modular spacecraft structure due to the existence of the pretightening force of the butting device. When a spacecraft system containing a connecting structure is subjected to dynamics analysis in engineering, in order to reduce the computation time and simplify the complexity of the problem, two sides of a model connecting surface are directly and fixedly connected, but the processing method can cause the connecting rigidity to be higher, so that the inherent frequency is higher, and the structural dynamic characteristics of the spacecraft configuration cannot be accurately analyzed. And the refined model is used for dynamic response analysis, and a dynamic characteristic identification method is used for analysis, so that a more accurate result can be obtained, but the overall analysis is time-consuming, and particularly when the spacecraft is complex in configuration and high in degree of freedom, the required calculation and time cost cannot meet the requirement of spacecraft concept design. In addition, due to the characteristics of multiple standard butting devices, adjustable pretightening force and modularization, a large number of configuration schemes need to be demonstrated in the conceptual design stage of the reconfigurable spacecraft. In the future development proposal of the modularized spacecraft, Huang Pan et al also points out the importance of structural dynamic characteristic parameter analysis on the modularized spacecraft configuration and mentions the influence of the structural dynamic characteristic of the modularized spacecraft on the space on-orbit task.
In recent decades, with the trend of the aerospace industry towards commercialization, some aerospace enterprises are seeking to further research related research and development processes to achieve the purposes of optimizing design, improving efficiency and controlling cost. However, no structural dynamic characteristic estimation method which takes a modular spacecraft as a main research object and gives consideration to both analysis efficiency and accuracy exists at present. The modularized spacecraft is an important and novel spacecraft at present, so that the modularized spacecraft has certain value for the research on the modularized spacecraft and can be applied to engineering practices related to the modularized spacecraft.
Disclosure of Invention
The invention aims to solve the problems of insufficient calculation precision and long time consumption of the existing method for estimating the structural dynamic characteristics of the modular spacecraft, and provides a method for quickly estimating the structural dynamic characteristics of the modular spacecraft. The method takes a modularized spacecraft as an object, and realizes the rapid estimation of the structural dynamic characteristics of the spacecraft configuration to be estimated including the natural frequency and the mode vibration type on the premise of knowing the splicing combination mode of the spacecraft configuration module to be estimated, the number of the formed modules and the pretightening force of the butt joint device.
The invention is realized by the following technical scheme.
A method for quickly estimating the structural dynamic characteristics of a modular spacecraft comprises the following steps:
Step 1: performing dynamic response analysis on the refined model of the dual-module spacecraft in finite element software, and performing frequency response function calculation and structural linearization detection processing on acceleration data obtained by the dynamic response analysis to finally complete dynamic characteristic identification; the refined model is a double-module spacecraft refined model with known pretightening force and considering nonlinear factors of contact and friction;
Step 1, comprising the following substeps:
Step 1.1: setting contact, friction properties and pretightening force of a refined model of the double-module spacecraft configuration in finite element software, selecting a proper excitation point to apply excitation, and realizing dynamic response analysis;
The double-module structure is composed of two spacecraft modules, and the modules comprise force bearing structures, standard butt joint devices and internal equipment. Nonlinear factors needing to be considered for refining the model comprise contact and friction, and normal and tangential properties of a contact area are defined according to actual conditions through contact property setting in finite element software;
The pre-tightening force is applied by a temperature load method, and the magnitude of the applied pre-tightening force is controlled by the deformation quantity of the butt joint device;
According to the principle that a structure low-order mode needs to be fully excited in dynamic response analysis, an excitation point is selected at a position capable of exciting the structure low-order mode, wherein the low order is the first five-order mode; the excited signal is characterized by Gaussian white noise, and the excitation cut-off frequency is higher than the natural frequency of a fifth-order mode obtained by modal analysis after the refined model is fixedly connected;
Solving the dynamic response analysis problem by adopting an implicit solver to obtain acceleration data of a refined model measuring point;
Step 1.2: calculating a frequency response function of the dual-module spacecraft structure based on the measuring point acceleration data and the excited signal characteristic data;
The acceleration data and the excited signal characteristic data are time domain data, after the data are converted into frequency domain data through Fourier transform, the self power spectrum and the cross power spectrum are calculated, and finally, the frequency response function of the dual-module spacecraft structure is obtained, wherein the calculation formula of the frequency response function is shown as a formula (1);
In the formula (1), H (omega) is a frequency response function of the dual-module spacecraft structure; g io(omega) is a cross-power spectrum of input excitation signal characteristic data and output acceleration data of the dual-module spacecraft; g oo(omega) is a self-power spectrum of acceleration data output by the dual-module spacecraft;
Step 1.3: performing structural linearization detection on the dual-module spacecraft structure based on the structural frequency response function, and if the structural linearization detection indicates that the structure can be similar to a linear structure, identifying the action characteristic;
the linear detection method comprises time sequence detection, output mean value detection, stability detection and other time domain methods, Hilbert transform detection, frequency response function detection, L issajous detection and other frequency domain methods and amplitude domain methods;
Step 1.4: identifying dynamic characteristic parameters of the structure of the dual-module spacecraft based on the frequency response function;
The input of the dynamic characteristic parameter identification algorithm is a frequency response function of the structure of the dual-module spacecraft, and the output is the structural dynamic characteristics including the inherent frequency and the mode shape of the structure of the dual-module spacecraft, namely the structural dynamic characteristic identification is completed;
Step 2: establishing a simplified model of a refined model of the double-module spacecraft configuration, analyzing the correlation between the simplified model and the refined model, performing pretreatment of correcting the finite element model after obtaining the correlation degree, converting the problem of correcting the finite element model into a standard optimization problem, and optimizing the problem of correcting the finite element model by using an optimization algorithm to realize the equivalence of the simplified model to the refined model;
Step 2.1: simplifying the components of the standard docking device according to the characteristics of a refined model of the double-module spacecraft configuration, directly and fixedly connecting the contact area of the standard docking device, and simultaneously ensuring that the quality characteristic of the configuration is not changed by adding non-structural quality to form a simplified model of the double-module spacecraft;
Step 2.2: performing correlation analysis and inspection on the simplified model and the refined model by taking the structural dynamics characteristics as a standard to obtain the correlation degree;
The structural dynamics characteristics of the simplified model can be directly obtained through finite element modal analysis, and the natural frequency and the modal shape of the simplified model are obtained; the structural dynamics of the refined finite element model are obtained through the step 1;
In the correlation test, the natural frequency and the modal shape are considered at the same time, if the correlation is higher than a preset standard, the model is recognized as a reliable simplified model, namely the equivalent of the simplified model to a refined model can be realized, and at the moment, the step 3 is executed; the simplified model whose correlation does not meet the standard will continue to execute step 2.3, and the formula used in the correlation analysis check is shown in formula (2);
In the formula (2), E freqIs the natural frequency error; f. of l,iAnd f n,iThe ith order intrinsic frequency values correspond to the simplified model and the refined model respectively; MAC ijObtaining modal confidence criterion values of the ith order modal shape of the refined model and the jth order modal shape of the simplified model; phi is a n,iAnd 
Transposing the ith order mode shape and the ith order mode shape of the refined model respectively; phi is a l,jAnd 
Transposing the ith order mode shape and the ith order mode shape of the simplified model respectively;
Step 2.3: introducing a virtual material, and carrying out finite element model correction pretreatment on the simplified model, wherein the treatment method comprises the following steps:
Step 2.3A, taking the formula (3) as a model correction target function, wherein the target function can reflect the difference between a simplified model and a refined model in the natural frequency and the mode shape;
In the formula (3), J is a correlation test result value; j. the design is a square freqAnd J shapeRespectively obtaining a natural frequency correlation value and a modal shape correlation value; n is the order considered by the correlation test; w is a freq,iAnd w shape,iRespectively is a weighted value corresponding to the ith order modal natural frequency and the modal shape; f. of l,iAnd f n,iThe ith order intrinsic frequency values correspond to the simplified model and the refined model respectively; MAC iModal confidence criterion values of ith order modal shape of the simplified model and the refined model;
Step 2.3B, determining a correction area of the simplified model, namely the area where the introduced virtual material is located, wherein the position of the area where the virtual material is located corresponds to the position of a connecting interface of the dual-module spacecraft structure;
Step 2.3C determines the correction parameters of the simplified model: obtaining sensitivity matrix values of different parameters to natural frequency and modal shape by calculating an attribute parameter sensitivity matrix of the virtual material, sequencing the attribute parameters of the virtual material according to the size of elements of the sensitivity matrix, and calculating the sensitivity matrix according to a formula shown in a formula (4);
In the formula (4), [ S ] ]A sensitivity matrix corresponding to the attribute parameters of the virtual material; f. of 1To f 5The natural frequencies of 1 st order to 5 th order of the simplified model respectively; MAC 1To MAC 5Modal confidence criterion values of 1 st to 5 th order modal vibration modes of the simplified model and the refined model are respectively; p is a radical of mIs the mth virtual material attribute parameter;
After the virtual material attribute parameters are sequenced, selecting 3 attribute parameters with large sensitivity matrix element values as correction parameters of the simplified model; the virtual material attribute parameters comprise 6 matrix element values in a two-dimensional anisotropic material parameter matrix, and the formula of the two-dimensional anisotropic material parameter matrix is as follows
In the formula (5), [ M ] ]A two-dimensional anisotropic material parameter matrix; g 11To G 33Respectively 6 matrix element values in the material parameter matrix;
Step 2.4: converting the finite element model correction problem into a standard optimization problem, and solving the standard optimization problem by using an optimization algorithm;
The standard optimization problem comprises an objective function, an optimization variable and a constraint condition, wherein the objective function is the model correction objective function determined in the step 2.3A and is shown as the formula (3); the optimization variables are correction parameters of the simplified model selected in the step 2.3C; determining constraint conditions by combining with the actual conditions of the model; the criteria optimization problem can be expressed as shown in the following equation (6)
In the formula (6), p 1To p 3Respectively representing correction parameters of 3 simplified models, namely optimization variables; j is an objective function; lb iAnd ub iRespectively taking the value of the ith optimization variable as a lower bound and an upper bound;
Optimizing parameter values of optimized variables by an optimization algorithm to change the simplified model, performing modal analysis on the changed simplified model to obtain the inherent frequency and the modal shape corresponding to the simplified model, substituting the newly obtained inherent frequency and the modal shape into an objective function calculation formula (3), changing the value of the objective function, and judging whether the simplified model meets the correlation requirement with the refined model by using an objective function value to meet the execution step 2.2; step 2.3 is not performed;
And step 3: forming a data set to construct a proxy model by inputting the pretightening force data corresponding to the step 1 and the optimization variable obtained in the step 2, namely correction parameter data, and calculating the precision of the proxy model, wherein the proxy model meeting the precision requirement can estimate the value of the correction parameter under the condition of knowing the pretightening force;
Step 3, comprising the following substeps:
Step 3.1: obtaining a data set required by constructing a proxy model;
Repeating the step 1 and the step 2, obtaining correction parameter values corresponding to different pretightening force values, and storing the pretightening force data and the correction parameter data in pairs into a data set;
Step 3.2: constructing an agent model;
Dividing the data set obtained in the step 3.1 into two parts, namely a training set and a verification set, wherein the paired data in the training set is more than the paired data in the verification set; training input of the agent model is a pretightening force value, output of the agent model is a correction parameter value, and the agent model is constructed by using a training set;
The method also comprises the step of carrying out precision verification on the proxy model obtained in the step 3.2;
And (4) inputting the pretightening force values in the verification set obtained in the step (3.2) into the proxy model one by one. Obtaining the estimation result of the proxy model for the correction parameters, calculating the estimation precision of the proxy model by combining the correction parameter values in the verification set, wherein the calculation method of the estimation precision can adopt a method of maximum cross-correlation entropy or relative root mean square error, and if the precision meets the requirement, executing the step 4; if the precision can not meet the requirement, executing the step 3.1 to carry out point addition reconstruction;
And 4, step 4: under the condition that the number of modules of the to-be-estimated modular spacecraft configuration, the splicing relation and the pretightening force value of the docking device are known, the structure dynamic characteristic of the to-be-estimated modular spacecraft is quickly estimated;
Step 4, comprising the following substeps:
Step 4.1: constructing a new simplified model by combining information of the modular spacecraft to be estimated;
Establishing a new simplified model based on the module composition number and the module splicing relation of the to-be-estimated modular spacecraft, wherein the module connecting surfaces are all processed in a fixed connection mode;
Step 4.2: taking the pretightening force value of the modular spacecraft to be estimated as the proxy model obtained in the step 3 for input, obtaining an estimated value of the proxy model for the new correction parameters, and applying the estimated value of the new correction parameters to the new simplified model established in the step 4.1;
Step 4.3: carrying out modal analysis on the simplified model of the modular spacecraft to be estimated after the step 4.2 is completed, and obtaining structural dynamic characteristic data;
The modal analysis is completed based on finite element analysis software, and the obtained structural dynamic characteristic data comprises the natural frequency and the modal shape of the modular spacecraft to be estimated.
Advantageous effects
Compared with the existing analysis method based on a refined model and a fixed connection simplified model without correction processing, the method for quickly estimating the structural dynamic characteristics of the modular spacecraft, disclosed by the invention, has the following beneficial effects:
1. The accuracy of the structural dynamic characteristic data of the reconfigurable spacecraft, which is obtained by the method, is higher than that of the structural dynamic characteristic data obtained by using a fixed connection simplified model without correction, and the requirement of the structural dynamic characteristic analysis accuracy in a concept design stage can be met;
2. The calculation time consumption required by the method is far lower than that required by structural dynamic characteristic analysis based on a refined model, the requirement of a concept design stage on the rapidity of the structural dynamic characteristic analysis is met, and the efficiency of concept design is ensured;
3. The method for quickly estimating the structural dynamic characteristics aims at the characteristics that the docking device of the reconfigurable spacecraft is unified, and the contained modules are numerous, can be widely applied to the configuration scheme of the reconfigurable spacecraft, and provides implementation conditions for design iteration and optimization of the configuration scheme.
Drawings
FIG. 1 is a basic outline diagram of a dual-module spacecraft in an embodiment 1 of a method for rapidly estimating structural dynamic characteristics of a modular spacecraft of the present invention;
Fig. 2 is a basic outline diagram of a seven-module spacecraft for performing fast estimation of structural dynamic characteristics in an embodiment 1 of a method for fast estimation of structural dynamic characteristics of a modular spacecraft of the present invention;
FIG. 3 is a general flow chart of a method for rapidly estimating structural dynamic characteristics of a modular spacecraft of the present invention;
FIG. 4 is a schematic diagram of a refined model of a dual-module spacecraft in embodiment 1 of the method for rapidly estimating structural dynamic characteristics of a modular spacecraft of the present invention;
FIG. 5 is a diagram showing the result of linear detection of a dual-module spacecraft structure in embodiment 1 of the method for rapidly estimating the structural dynamic characteristics of a modular spacecraft of the present invention;
Fig. 6 is a schematic view of a correction region of a simplified model of a dual-module spacecraft in an embodiment 1 of a method for rapidly estimating structural dynamic characteristics of a modular spacecraft of the present invention.
Detailed Description
In order to better illustrate the objects and advantages of the invention, the present method is used to quickly evaluate the structural and dynamic characteristics of a seven-module spacecraft, which is a modular spacecraft to be evaluated, and the present invention is explained in detail through this complete process.
Example 1
The configuration of the two-module spacecraft and the configuration of the seven-module spacecraft related in the embodiment are respectively shown in fig. 1 and fig. 2, and the spacecraft forming modules in fig. 1 and fig. 2 both adopt hexahedral modules, so that the characteristics of flexible assembly, large design space and on-orbit service oriented of the modularized spacecraft compared with the traditional spacecraft are highlighted, and the hexahedral modularized spacecraft is the most widely used spacecraft module in the related industries. According to the method for quickly estimating the structural dynamic characteristics of the modular spacecraft, the composition characteristics of the modular spacecraft are fully considered, and compared with a traditional analysis method based on a direct-connection simplified model or a refined model, the method gives consideration to the analysis precision and efficiency. The design efficiency of the modular spacecraft in the conceptual design stage is effectively improved, design iteration and optimization work are supported, and the workload of designers is reduced.
The specific implementation steps are shown in a flow chart in fig. 3, and include:
Step I: in this embodiment, the range of the pretightening force of the docking device corresponding to the modular spacecraft is set to 250N to 1200N, the pretightening force is selected to be 250N, a refined model of the dual-module spacecraft is subjected to dynamic response analysis, the acceleration data obtained by the dynamic response analysis is subjected to frequency response function calculation and structure linearization detection processing, and finally, dynamic characteristic identification is completed, and the refined model of the dual-module spacecraft used for the dynamic response analysis is shown in fig. 4.
Step I.1: applying 250N pre-tightening force to the module butt joint device through a temperature load method, setting contact and friction properties of a refined model of the double-module spacecraft configuration in finite element software, selecting a proper excitation point to apply excitation, and realizing dynamic response analysis;
The dynamic response analysis is performed in the ABAQUS finite element analysis software, which in this example is divided into two analysis steps, a general static analysis step and an implicit kinetic analysis step in sequence. In the general static analysis step, the boundary condition of the given spacecraft is a side fixed support, and the pretightening force of the butt joint device is applied through the temperature load; in the implicit dynamics analysis step, a point is selected on the unfixed side of the spacecraft as an excitation point, excitation signal characteristics are selected from Gaussian white noise excitation, the cut-off frequency is 350Hz and is higher than the frequency of a fifth-order mode obtained by modal analysis after a refined model is fixedly connected, and the analysis output is set as acceleration data.
Step I.2: and calculating the frequency response function of the dual-module spacecraft structure through the data obtained by finite element dynamic response analysis and the excited signal characteristic data.
In this embodiment, the frequency response function is calculated by the method of formula (1), so that the frequency response function of the dual-module spacecraft structure under the pre-tightening force of 250N can be obtained.
Step I.3: by using the frequency response function obtained in the step i.2, the dual-module spacecraft structure is confirmed to be capable of identifying the dynamic characteristic parameters by a linearization inspection method based on Hlibert transformation in the embodiment;
The principle of the Hlibert transformation linearization detection method is shown as the formula (7);
Re(H(ω))＝HI{Im(H(ω))} (7)
In the formula (7), H (omega) is a frequency response function of the dual-module spacecraft; re (H (omega)) represents the real part of the frequency response function of the dual-module spacecraft; HI { Im (H (omega)) } represents Hlibert transform of the imaginary part of the frequency response function of the dual-module spacecraft; the specific calculation method of the Hlibert transformation is shown as a formula (8);
In equation (8), HI { f (t) } represents the Hlibert transform of the real time function f (t);
When the pre-tightening force is 250N, the result obtained by the structural linearization inspection of the dual-module spacecraft is shown in FIG. 5, the Hlibert transformation matching degree of the real part and the imaginary part in FIG. 5 is high, and it is shown that the dual-module spacecraft when the pre-tightening force is 250N can be approximately of a linear structure and can perform dynamic characteristic parameter identification.
Step I.4: and (4) performing action characteristic parameter identification on the structure based on the frequency response function of the dual-module spacecraft obtained in the step I.2.
In this embodiment, a PloyMAX identification method is used to identify the motion characteristic parameters, and the natural frequency and the mode shape corresponding to the dual-module spacecraft configuration with the pretightening force of 250N are obtained.
Step II: establishing a simplified model of a refined model of the double-module spacecraft configuration, analyzing the correlation between the simplified model and the refined model, performing pretreatment of correcting the finite element model after obtaining the correlation degree, converting the problem of correcting the finite element model into a standard optimization problem, and optimizing the problem of correcting the finite element model by using an optimization algorithm to realize the equivalence of the simplified model to the refined model;
Step II, 1: in the embodiment, a simplified model of the double-module spacecraft is established according to the refined model characteristics of the double-module spacecraft configuration, and the quality characteristics of the spacecraft configuration are ensured to be unchanged by adding non-structural mass.
In the embodiment, based on the characteristics of the refined model, series of characteristics at the connecting device of the double-module spacecraft module are omitted. And (3) directly and fixedly connecting the contact area of the standard butt joint device without considering non-linear factors including friction and contact, and establishing a simplified model in ABAQUS finite element software. The weight of each module in the refined model is 35Kg, and the quality characteristic of the simplified model is ensured to be consistent with that of the refined model by adding non-structural quality to the whole model.
Step II, 2: in this embodiment, the simplified model and the refined model are subjected to correlation test using the structural dynamics characteristics as a standard, and if the correlation test meets the correlation requirement, step iii may be performed, and if the correlation requirement is not met, step ii.3 may be performed.
The method comprises the following steps that a natural frequency and a mode shape of a structure are considered in correlation inspection, a formula used in the correlation inspection is shown in a formula (2), the standard required by the correlation is that the natural frequency error is less than 5%, and the mode shape confidence criterion value is greater than 0.9.
Step II, 3: in this embodiment, formula (3) is used as a model modification objective function, a virtual material with two-dimensional anisotropy of material properties is introduced, and a finite element model modification preprocessing is performed on the simplified model.
The correction function in this embodiment considers the natural frequency and the mode shape, and selects the correction region as shown in fig. 6, where the boundary region in the middle of the module in fig. 6 is the correction region, and corresponds to the connection position of the simplified model module.
As the virtual material is set as a two-dimensional linear anisotropic material, a sensitivity matrix of the attribute parameters is calculated by the formula (4), the attribute parameters of the virtual material are sequenced according to the element size of the sensitivity matrix, and the correction parameters are determined to be 3 diagonal item parameters in the material attribute matrix.
Step II, 4: in the embodiment, a standard optimization problem corresponding to the model correction problem is obtained according to the processing of the step II.3, the expression is shown as the formula (6), the problem is optimized and solved by using a genetic algorithm, the inherent frequency and the modal shape corresponding to the simplified model are obtained, the newly obtained inherent frequency and the modal shape are substituted into the objective function calculation formula (3), the value of the objective function is changed, the objective function value is used for judging whether the simplified model meets the correlation with the refined model or not, and the step II.2 is executed; the step II and the step 3 are not executed.
Step III: in this embodiment, the method includes the steps of repeating step I and step ii with a pretightening force range of 250N to 1200N and a pretightening force interval of 10N, respectively obtaining a pretightening force value and a correction parameter value, forming a data set, constructing a proxy model, performing accuracy verification on the proxy model, and estimating the correction parameter value when the pretightening force is known by using the proxy model satisfying the accuracy requirement.
Step III, 1: in this embodiment, the step I and the step ii are repeated to obtain different pretightening force values and corresponding correction parameter values, and 96 sets of data pairs are counted up, and the two types of data are stored.
Step III, 2: in this embodiment, a Kriging agent model is selected and the agent model is constructed.
Using the data set obtained in step iii.1, the data set is divided into two parts, a training set comprising 90 sets of data and a validation set comprising 6 sets of data. The training input of the Kriging agent model is a pretightening force value, the output of the Kriging agent model is a correction parameter value, and the agent model is constructed by using a training set.
Step III, 3: in the embodiment, the accuracy of the constructed Kriging proxy model is verified by taking the relative root mean square error as a reference, and a calculation formula of the relative root mean square error is shown as a formula (9).
In formula (9), RMSE is relative root mean square error; n is tNumber of pairs of data for validation sets; y is iAnd 
Respectively taking values of correction parameters in the verification set and estimation values of the proxy model for the correction parameters; y is iThe mean value of the values of the correction parameters in the verification set is obtained.
In the embodiment, the accuracy of the proxy model with the relative root mean square error smaller than 3% is set, namely, the accuracy requirement is met. The Kriging model obtained using a training set containing 90 sets of data was precision tested, and the samples tested were validation sets containing 6 sets of data. After verification, the precision of the agent model meets the requirement, and the step IV can be executed; and if the precision can not meet the requirement, executing the step III.1, and performing point addition reconstruction.
Step IV: in this embodiment, the seven-module spacecraft is used as the configuration of the modular spacecraft to be estimated, and the structural dynamic characteristics of the seven-module spacecraft are quickly estimated under the condition that the pretightening force, the module splicing relation and the number of module forming blocks of the design scheme of the seven-module spacecraft are obtained.
Step IV, 1: and combining the design scheme information of the seven-module spacecraft to construct a new simplified model. In this embodiment, the pre-tightening force of the seven-module spacecraft docking device is 505N, the module splicing relationship is shown in fig. 2, and a new simplified model can be constructed in the ABAQUS software based on the module component number and the module splicing relationship.
Step IV, 2: and improving a new simplified model of the seven-module spacecraft by utilizing the estimated value of the correction parameter.
In the embodiment, the pre-tightening force value 505N is input into the Kriging surrogate model obtained by training in the step III to obtain the estimation value of the surrogate model for the new correction parameter, and the estimation value of the new correction parameter is applied to the new simplified model established in the step IV.1.
Step IV, 3: and (3) applying the new correction parameter estimation value obtained in the step (IV.2) to the new simplified model established in the step (IV.1), and carrying out modal analysis in ABAQUS software to obtain the structural dynamic characteristic data corresponding to the seven-module spacecraft.
In the embodiment, the structural dynamic characteristics of the seven-module spacecraft are estimated, the time consumed by single calculation is 70s, and compared with the result obtained by fixedly connecting and simplifying the model, the accuracy of the obtained result reduces the error of the data of the structural dynamic characteristics of the first five orders obtained by calculation in the formula (2) from 20.74% to 3.24%. Advantageous effects 1 and 2 of the present invention are embodied.
According to the steps I to IV, the structure dynamic characteristic of the modular spacecraft is quickly estimated, and the beneficial effect 3 of the invention is embodied. The method disclosed by the invention can provide powerful support for the design of the configuration scheme of the modular spacecraft, the rationality analysis, the design iteration of the concept design stage of the modular spacecraft and other aspects, and has wide application prospect and benefit.
The above detailed description is intended to illustrate the objects, aspects and advantages of the present invention, and it should be understood that the above detailed description is only exemplary of the present invention, and is not intended to limit the scope of the present invention, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (2)
1. A method for quickly estimating the structural dynamic characteristics of a modular spacecraft is characterized by comprising the following steps: the method comprises the following steps:
Step 1: performing dynamic response analysis on the refined model of the dual-module spacecraft in finite element software, and performing frequency response function calculation and structural linearization detection processing on acceleration data obtained by the dynamic response analysis to finally complete dynamic characteristic identification; the refined model is a double-module spacecraft refined model with known pretightening force and considering nonlinear factors of contact and friction;
Step 1.1: setting contact, friction properties and pretightening force of a refined model of the double-module spacecraft configuration in finite element software, selecting a proper excitation point to apply excitation, and realizing dynamic response analysis;
The double-module structure is composed of two spacecraft modules, wherein each module comprises a bearing structure, a standard butt joint device and internal equipment; nonlinear factors needing to be considered for refining the model comprise contact and friction, and normal and tangential properties of a contact area are defined according to actual conditions through contact property setting in finite element software;
The pre-tightening force is applied by a temperature load method, and the magnitude of the applied pre-tightening force is controlled by the deformation quantity of the butt joint device;
According to the principle that a structure low-order mode needs to be fully excited in dynamic response analysis, an excitation point is selected at a position capable of exciting the structure low-order mode, wherein the low order is the first five-order mode; the excited signal is characterized by Gaussian white noise, and the excitation cut-off frequency is higher than the natural frequency of a fifth-order mode obtained by modal analysis after the refined model is fixedly connected;
Solving the dynamic response analysis problem by adopting an implicit solver to obtain acceleration data of a refined model measuring point;
Step 1.2: calculating a frequency response function of the dual-module spacecraft structure based on the measuring point acceleration data and the excited signal characteristic data;
The acceleration data and the excited signal characteristic data are time domain data, after the data are converted into frequency domain data through Fourier transform, the self power spectrum and the cross power spectrum are calculated, and finally, the frequency response function of the dual-module spacecraft structure is obtained, wherein the calculation formula of the frequency response function is shown as a formula (1);
In the formula (1), H (omega) is a frequency response function of the dual-module spacecraft structure; g io(omega) is a cross-power spectrum of input excitation signal characteristic data and output acceleration data of the dual-module spacecraft; g oo(omega) is a self-power spectrum of acceleration data output by the dual-module spacecraft;
Step 1.3: performing structural linearization detection on the dual-module spacecraft structure based on the structural frequency response function, and if the structural linearization detection indicates that the structure can be similar to a linear structure, identifying the action characteristic;
the linear detection method comprises time sequence detection, output mean value detection, stability detection and other time domain methods, Hilbert transform detection, frequency response function detection, L issajous detection and other frequency domain methods and amplitude domain methods;
Step 1.4: identifying dynamic characteristic parameters of the structure of the dual-module spacecraft based on the frequency response function;
The input of the dynamic characteristic parameter identification algorithm is a frequency response function of the structure of the dual-module spacecraft, and the output is the structural dynamic characteristics including the inherent frequency and the mode shape of the structure of the dual-module spacecraft, namely the structural dynamic characteristic identification is completed;
Step 2: establishing a simplified model of a refined model of the double-module spacecraft configuration, analyzing the correlation between the simplified model and the refined model, performing pretreatment of correcting the finite element model after obtaining the correlation degree, converting the problem of correcting the finite element model into a standard optimization problem, and optimizing the problem of correcting the finite element model by using an optimization algorithm to realize the equivalence of the simplified model to the refined model;
Step 2.1: simplifying the components of the standard docking device according to the characteristics of a refined model of the double-module spacecraft configuration, directly and fixedly connecting the contact area of the standard docking device, and simultaneously ensuring that the quality characteristic of the configuration is not changed by adding non-structural quality to form a simplified model of the double-module spacecraft;
Step 2.2: performing correlation analysis and inspection on the simplified model and the refined model by taking the structural dynamics characteristics as a standard to obtain the correlation degree;
The structural dynamics characteristics of the simplified model can be directly obtained through finite element modal analysis, and the natural frequency and the modal shape of the simplified model are obtained; the structural dynamics of the refined finite element model are obtained through the step 1;
In the correlation test, the natural frequency and the modal shape are considered at the same time, if the correlation is higher than a preset standard, the model is recognized as a reliable simplified model, namely the equivalent of the simplified model to a refined model can be realized, and at the moment, the step 3 is executed; the simplified model whose correlation does not meet the standard will continue to execute step 2.3, and the formula used in the correlation analysis check is shown in formula (2);
In the formula (2), E freqIs the natural frequency error; f. of l,iAnd f n,iThe ith order intrinsic frequency values correspond to the simplified model and the refined model respectively; MAC ijObtaining modal confidence criterion values of the ith order modal shape of the refined model and the jth order modal shape of the simplified model; phi is a n,iAnd 
Respectively an ith order mode shape and an ith order mode shape of the refined model Transposing i-order mode shape; phi is a l,jAnd 
Transposing the ith order mode shape and the ith order mode shape of the simplified model respectively;
Step 2.3: introducing a virtual material, and carrying out finite element model correction pretreatment on the simplified model, wherein the treatment method comprises the following steps:
Step 2.3A, taking the formula (3) as a model correction target function, wherein the target function can reflect the difference between a simplified model and a refined model in the natural frequency and the mode shape;
In the formula (3), J is a correlation test result value; j. the design is a square freqAnd J shapeRespectively obtaining a natural frequency correlation value and a modal shape correlation value; n is the order considered by the correlation test; w is a freq,iAnd w shape,iRespectively is a weighted value corresponding to the ith order modal natural frequency and the modal shape; f. of l,iAnd f n,iThe ith order intrinsic frequency values correspond to the simplified model and the refined model respectively; MAC iModal confidence criterion values of ith order modal shape of the simplified model and the refined model;
Step 2.3B, determining a correction area of the simplified model, namely the area where the introduced virtual material is located, wherein the position of the area where the virtual material is located corresponds to the position of a connecting interface of the dual-module spacecraft structure;
Step 2.3C determines the correction parameters of the simplified model: obtaining sensitivity matrix values of different parameters to natural frequency and modal shape by calculating an attribute parameter sensitivity matrix of the virtual material, sequencing the attribute parameters of the virtual material according to the size of elements of the sensitivity matrix, and calculating the sensitivity matrix according to a formula shown in a formula (4);
In the formula (4), [ S ] ]A sensitivity matrix corresponding to the attribute parameters of the virtual material; f. of 1To f 5The natural frequencies of 1 st order to 5 th order of the simplified model respectively; MAC 1To MAC 5Modal confidence criterion values of 1 st to 5 th order modal vibration modes of the simplified model and the refined model are respectively; p is a radical of mIs the mth virtual material attribute parameter;
After the virtual material attribute parameters are sequenced, selecting 3 attribute parameters with large sensitivity matrix element values as correction parameters of the simplified model; the virtual material attribute parameters comprise 6 matrix element values in a two-dimensional anisotropic material parameter matrix, and the formula of the two-dimensional anisotropic material parameter matrix is as follows
In the formula (5), [ M ] ]A two-dimensional anisotropic material parameter matrix; g 11To G 33Respectively 6 matrix element values in the material parameter matrix;
Step 2.4: converting the finite element model correction problem into a standard optimization problem, and solving the standard optimization problem by using an optimization algorithm;
The standard optimization problem comprises an objective function, an optimization variable and a constraint condition, wherein the objective function is the model correction objective function determined in the step 2.3A and is shown as the formula (3); the optimization variables are correction parameters of the simplified model selected in the step 2.3C; determining constraint conditions by combining with the actual conditions of the model; the criteria optimization problem can be expressed as shown in the following equation (6)
In the formula (6), p 1To p 3Respectively representing correction parameters of 3 simplified models, namely optimization variables; j is an objective function; lb iAnd ub iRespectively taking the value of the ith optimization variable as a lower bound and an upper bound;
Optimizing parameter values of optimized variables by an optimization algorithm to change the simplified model, performing modal analysis on the changed simplified model to obtain the inherent frequency and the modal shape corresponding to the simplified model, substituting the newly obtained inherent frequency and the modal shape into an objective function calculation formula (3), changing the value of the objective function, and judging whether the simplified model meets the correlation requirement with the refined model by using an objective function value to meet the execution step 2.2; step 2.3 is not performed;
And step 3: forming a data set to construct a proxy model by inputting the pretightening force data corresponding to the step 1 and the optimization variable obtained in the step 2, namely correction parameter data, and calculating the precision of the proxy model, wherein the proxy model meeting the precision requirement can estimate the value of the correction parameter under the condition of knowing the pretightening force;
Step 3.1: obtaining a data set required by constructing a proxy model;
Repeating the step 1 and the step 2, obtaining correction parameter values corresponding to different pretightening force values, and storing the pretightening force data and the correction parameter data in pairs into a data set;
Step 3.2: constructing an agent model;
Dividing the data set obtained in the step 3.1 into two parts, namely a training set and a verification set, wherein the paired data in the training set is more than the paired data in the verification set; training input of the agent model is a pretightening force value, output of the agent model is a correction parameter value, and the agent model is constructed by using a training set;
And 4, step 4: under the condition that the number of modules of the to-be-estimated modular spacecraft configuration, the splicing relation and the pretightening force value of the docking device are known, the structure dynamic characteristic of the to-be-estimated modular spacecraft is quickly estimated;
Step 4.1: constructing a new simplified model by combining information of the modular spacecraft to be estimated;
Establishing a new simplified model based on the module composition number and the module splicing relation of the to-be-estimated modular spacecraft, wherein the module connecting surfaces are all processed in a fixed connection mode;
Step 4.2: taking the pretightening force value of the modular spacecraft to be estimated as the proxy model obtained in the step 3 for input, obtaining an estimated value of the proxy model for the new correction parameters, and applying the estimated value of the new correction parameters to the new simplified model established in the step 4.1;
Step 4.3: carrying out modal analysis on the simplified model of the modular spacecraft to be estimated after the step 4.2 is completed, and obtaining structural dynamic characteristic data;
The modal analysis is completed based on finite element analysis software, and the obtained structural dynamic characteristic data comprises the natural frequency and the modal shape of the modular spacecraft to be estimated.
2. The method of claim 1 for fast estimation of structural dynamic properties of a modular spacecraft, characterized by: the method also comprises the step of carrying out precision verification on the proxy model obtained in the step 3.2;
Inputting the pretightening force values in the verification set obtained in the step 3.2 into the proxy model one by one; obtaining the estimation result of the proxy model for the correction parameters, calculating the estimation precision of the proxy model by combining the correction parameter values in the verification set, wherein the calculation method of the estimation precision can adopt a method of maximum cross-correlation entropy or relative root mean square error, and if the precision meets the requirement, executing the step 4; and if the precision can not meet the requirement, executing the step 3.1 and carrying out point addition reconstruction.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010042100 | 2020-01-15 | ||
| CN2020100421006 | 2020-01-15 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111444648A true CN111444648A (en) | 2020-07-24 |
| CN111444648B CN111444648B (en) | 2022-02-08 |
Family
ID=71652431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010212207.0A Active CN111444648B (en) | 2020-01-15 | 2020-03-24 | Method for quickly estimating structural dynamic characteristics of modular spacecraft |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111444648B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113408672A (en) * | 2021-08-19 | 2021-09-17 | 中国科学院力学研究所 | Key parameter identification method for aircraft modal test |
| CN116484512A (en) * | 2023-06-22 | 2023-07-25 | 西北工业大学 | Identification method for pre-tightening state of disc-drum rotor of aero-engine |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090306802A1 (en) * | 2008-06-04 | 2009-12-10 | Cone Michael | Method and system for optimizing the vibrational characteristics of a structure |
| CN104833466A (en) * | 2015-04-30 | 2015-08-12 | 北京航空航天大学 | Spacecraft ground test and on-orbit micro-vibration mechanical environment mapping method |
| CN107066701A (en) * | 2017-03-21 | 2017-08-18 | 北京强度环境研究所 | The model building method of dynamics experiment based on spacecraft |
| CN107941441A (en) * | 2017-11-14 | 2018-04-20 | 北京卫星环境工程研究所 | Determine the method that the in-orbit border of simulation influences the in-orbit dynamics of spacecraft |
-
2020
- 2020-03-24 CN CN202010212207.0A patent/CN111444648B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090306802A1 (en) * | 2008-06-04 | 2009-12-10 | Cone Michael | Method and system for optimizing the vibrational characteristics of a structure |
| CN104833466A (en) * | 2015-04-30 | 2015-08-12 | 北京航空航天大学 | Spacecraft ground test and on-orbit micro-vibration mechanical environment mapping method |
| CN107066701A (en) * | 2017-03-21 | 2017-08-18 | 北京强度环境研究所 | The model building method of dynamics experiment based on spacecraft |
| CN107941441A (en) * | 2017-11-14 | 2018-04-20 | 北京卫星环境工程研究所 | Determine the method that the in-orbit border of simulation influences the in-orbit dynamics of spacecraft |
Non-Patent Citations (3)
| Title |
|---|
| MORTEZA IRANZAD等: "Identification of nonlinear bolted lap joint models", 《COMPUTERS AND STRUCTURES》 * |
| 孙志勇: "基于虚拟材料的栓接结合部动力学特性研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 张敬东: "航空电子设备安装架动态特性分析及结构优化", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113408672A (en) * | 2021-08-19 | 2021-09-17 | 中国科学院力学研究所 | Key parameter identification method for aircraft modal test |
| CN113408672B (en) * | 2021-08-19 | 2021-11-09 | 中国科学院力学研究所 | Key parameter identification method for aircraft modal test |
| CN116484512A (en) * | 2023-06-22 | 2023-07-25 | 西北工业大学 | Identification method for pre-tightening state of disc-drum rotor of aero-engine |
| CN116484512B (en) * | 2023-06-22 | 2023-09-01 | 西北工业大学 | Identification method for pre-tightening state of disc-drum rotor of aero-engine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111444648B (en) | 2022-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mannarino et al. | Nonlinear aeroelastic reduced order modeling by recurrent neural networks | |
| Zhang et al. | Efficient method for limit cycle flutter analysis based on nonlinear aerodynamic reduced-order models | |
| CN110210044A (en) | Load prediction method and device of wind generating set | |
| CN111444648B (en) | Method for quickly estimating structural dynamic characteristics of modular spacecraft | |
| CN109711061A (en) | Dynamic assembly suspension Robust Optimization method based on vehicle idle vibration performance | |
| CN105260581A (en) | Method for virtual vibration and impact tests of electromechanical control equipment of ship | |
| CN105631090A (en) | Finite element model optimization device and method | |
| CN114707390B (en) | Layered shell material based constitutive parameter updating hybrid test method and analysis method | |
| CN110631792A (en) | Seismic hybrid test model updating method based on convolutional neural network | |
| Alıcıoğlu et al. | Ambient vibration analysis with subspace methods and automated mode selection: case studies | |
| CN106017953A (en) | Test method and system used for large-scale complex structure test research | |
| Orts et al. | Finite size effects in active microrheology in colloids | |
| CN110765560B (en) | Mechanical mechanism vibration prediction method based on time-varying damping | |
| D'Augustine | MATLODE: A MATLAB ODE solver and sensitivity analysis toolbox | |
| CN115310327A (en) | Simulation analysis method and device for multi-working-condition combination | |
| Blind et al. | Towards Exascale CFD Simulations Using the Discontinuous Galerkin Solver FLEXI | |
| Li et al. | Approach to prove the efficiency of the Monte Carlo method combined with the elementary effect method to quantify uncertainty of a beam structure with piezo-elastic supports | |
| Gomez-Escalonilla et al. | Development of efficient high-fidelity solutions for virtual fatigue testing | |
| Winter et al. | Application of Unsteady Aerodynamic Reduced-Order Modeling Techniques to a Complex Configuration | |
| Michopoulos et al. | Dynamic data driven methodologies for multiphysics system modeling and simulation | |
| Dai et al. | Impacts of the weight coefficient and modal assurance criterion of large structures on observation station selection and optimization | |
| Liu et al. | Efficient training data generation for reduced-order modeling in a transonic flight regime | |
| Moura et al. | On a stochastic regularization technique for ill-conditioned linear systems | |
| Bonfanti et al. | Non-linear Frequency Domain Modelling of a Wave Energy Harvester | |
| Vinjanampati et al. | Methodology for Performing Submodel Analysis for Random Vibration Problems using Modal Analysis Results |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |