A Method and System for Evaluating Modal Parameter Stability Based on Single-Point Excitation and Multi-Point Response
By conducting repeated single-point excitation and multi-point response tests in the modal tests of a large gas turbine rotor, valid data were screened and stability indices were calculated, thus solving the problem of modal parameter stability evaluation, ensuring the reliability and consistency of modal parameters, and supporting the verification of the dynamic model.
Patent Information
- Application Number
- CN202610952113.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-29
- Publication Date
- 2026-11-10
AI Technical Summary
In modal tests of large gas turbine rotors, how to prove that the identified natural frequencies and mode shapes have sufficient repeatability and stability for dynamic model verification, especially in single-point excitation multi-point response modal tests, is a challenge. Existing technologies lack effective methods for evaluating stability and repeatability.
By conducting repeated single-point excitation multi-point response modal tests under fixed test boundary conditions, valid data are screened, relative frequency deviation, mode shape deviation, and node position deviation are calculated, a stability index is constructed, and the stability of the modal parameters is determined to meet the requirements for dynamic model verification.
It enables quantitative stability evaluation of modal parameters, ensuring their reliability and consistency, and can be used to verify the dynamic model of large gas turbine rotors.
Smart Images

Figure CN123065699A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of rotor modal testing, and more specifically, to a method and system for evaluating the stability of modal parameters based on single-point excitation and multi-point response. Background Technology
[0002] The dynamic characteristics of the tie-rod rotor of a large gas turbine directly affect the unit's critical speed, vibration response, shaft stability, and operational safety. Because these rotors typically have large lengths, large spans, complex disk and blade structures, and significant coupling between tie rod preload and the contact interface, theoretical dynamic models alone cannot fully reflect the actual structural state. Therefore, reliable static frequency and modal tests are necessary to obtain parameters such as natural frequencies and mode shapes.
[0003] Modeling of the contact interface between the tie rod and rotor, the virtual material layer method, the thin-layer unit method, and the molecular dynamics contact model all require verification through experimental values of natural frequencies and calculations from theoretical models. Therefore, modal testing is not merely a simple testing step, but a fundamental step in the correction and verification of rotor dynamics models and designs.
[0004] The reliability of modal test results for large gas turbine rotors cannot be judged solely by the peak frequency response or a single natural frequency from a single test. Factors such as hoisting boundaries, hammer impact angle, sensor installation quality, signal noise, and data processing parameters can all affect modal identification results. Without repeatability and stability evaluation, even if the first three natural frequencies are obtained, it is difficult to prove that the results can be used for dynamic model verification of the rod rotor. For large gas turbine rotors, demonstrating that the identified natural frequencies and mode shapes possess sufficient repeatability and stability for dynamic model verification during single-point excitation multi-point response modal tests is a key technical challenge. Summary of the Invention
[0005] This invention provides a method and system for evaluating the stability of modal parameters based on single-point excitation and multi-point response, in order to overcome at least one technical problem existing in the prior art.
[0006] In a first aspect, embodiments of the present invention provide a method for evaluating the stability of modal parameters based on single-point excitation and multi-point response, comprising: Set experimental boundary conditions, fix the experimental conditions under the boundary conditions, and conduct m repeated single-point excitation multi-point response modal tests based on the fixed experimental conditions, where m ≥ 3; All experimental data are subjected to validity screening. After each invalid test is removed, a single-point excitation multi-point response modal test is added, and the validity screening is performed on the supplementary experimental data until m valid experimental data are obtained. For each of the m valid tests, the r-th modal frequency and mode shape vector are extracted, and the mode shape vector is normalized. Calculate the relative frequency deviation based on the r-th modal frequency from m valid tests; Based on the normalized results of the r-th order mode shape vectors from all valid tests, the mode shape deviation is calculated. Identify the node positions of the r-th mode shape in all valid tests and calculate the node position deviations; The stability index is calculated based on the relative frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate of the target frequency band. Determine whether the stability index meets the stability requirements. If it does, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable, and the test boundary conditions are set for re-examination.
[0007] Optionally, the test conditions include at least: boundary state, excitation point location, coordinates of N response measurement points, sensor installation method, and data acquisition channel configuration; wherein, N ≥ 2.
[0008] Optionally, the data from m trials are screened for validity, specifically including: The raw test data for each experiment is pre-checked. If there are two consecutive hits, the corresponding test data is deemed invalid and discarded; otherwise, the pre-check is passed. The coherence function and frequency response function of the target frequency band are calculated for the test data that pass the pre-inspection. If both the coherence function and the frequency response function meet the preset requirements, the corresponding test data is determined to be valid; otherwise, the corresponding test data is determined to be invalid and discarded.
[0009] Optionally, the mode shape vector is normalized, specifically including: The mode shape vector is normalized using amplitude normalization: ,in, This represents the r-th mode shape vector of the k-th valid test. It represents the i-th component of the r-th mode vector of the k-th valid test.
[0010] Optionally, the relative frequency deviation is calculated based on the r-th modal frequency from m valid tests, specifically including: Calculate the r-th modal frequency of m valid trials average ; Calculate the relative frequency deviation based on the average value. ,in, This is the error term.
[0011] Optionally, based on the normalized results of the r-th order mode shape vectors from all valid tests, the mode shape deviation is calculated, specifically including: Based on the normalized results of the r-th order mode vectors from all valid experiments, the mode correlation coefficient between any two valid experiments is calculated. ,in, This represents the r-th mode shape vector of the p-th valid test. The normalization result, This represents the r-th mode shape vector of the q-th valid test. The normalization result; The mean value is obtained by averaging the modal correlation coefficients of all pairwise combinations of the r-th order. ; Based on the mean value, the modal shape deviation is calculated as follows: .
[0012] Optionally, identify the nodal positions of the r-th mode shape in all valid tests and calculate the nodal position deviations, specifically including: Based on the normalized result of the r-th order mode shape vector obtained from each test, the mode shape node positions are identified; the node positions are the axial positions where the mode shape curve crosses zero or has the minimum amplitude. Calculate the node position deviation based on the node position. Where L represents the range of axial measuring points on the rotor. and These represent the mode node positions of the p-th and q-th valid tests, respectively.
[0013] Alternatively, the stability index can be expressed as: ,in, , , , These represent the weights of the corresponding indicators. For normalization function, Indicates the relative frequency deviation. Indicates mode shape deviation. Indicates the deviation of the node position. This represents the pass rate of the coherence function in the target frequency band.
[0014] Optionally, before calculating the stability index, the coherence function pass rate of the target frequency band is also calculated, specifically including: Calculate the number of frequency points within the target frequency band that satisfy the coherence function requirements; The proportion of frequency points that meet the coherence function requirements to the total number of frequency points is calculated to obtain the coherence function qualification rate of the target frequency band.
[0015] Secondly, the present invention also provides a modal parameter stability evaluation system based on single-point excitation and multi-point response, comprising: The setting module is used to set the test boundary conditions, fix the test conditions under the boundary conditions, and perform m repeated single-point excitation multi-point response modal tests based on the fixed test conditions, where m ≥ 3; The filtering module is used to filter the validity of all experimental data. After each invalid experiment is removed, a single-point excitation multi-point response modal experiment is added, and the validity of the added experimental data is filtered until m valid experimental data are obtained. The extraction module is used to extract the r-th modal frequency and mode shape vector from m valid tests, and to normalize the mode shape vector. The calculation module is used to calculate the relative frequency deviation based on the r-th modal frequency of m valid tests; it is also used to calculate the mode shape deviation based on the normalized result of the r-th mode shape vector of all valid tests. The identification module is used to identify the node positions of the r-th mode shape in all valid tests and to calculate the node position deviations. The calculation module is also used to calculate the stability index based on the relative frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate of the target frequency band. The judgment module is used to determine whether the stability index meets the stability requirements. If it does, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable and the system returns to set experimental boundary conditions for re-examination.
[0016] The innovative aspects of this invention include, but are not limited to, the following: In this embodiment, under fixed experimental conditions, multiple single-point excitation multi-point response tests are repeatedly performed, and the same-order modal frequencies and mode shape vectors are identified respectively. The same-order frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate constitute a multi-index evaluation of modal parameter stability index. The stability evaluation results are used as a prerequisite criterion for modal parameters to enter the verification of large gas turbine rotor dynamics model. This realizes the quantitative judgment of whether modal parameters can enter the verification process of large gas turbine rotor dynamics model, which is one of the innovations of this embodiment. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A flowchart of an evaluation method provided in an embodiment of the present invention; Figure 2 A flowchart for data validity screening provided in an embodiment of the present invention; Figure 3 A flowchart for calculating mode shape deviation provided in an embodiment of the present invention; Figure 4 Another flowchart of the evaluation method provided in the embodiments of the present invention; Figure 5 This is a schematic diagram of an evaluation system provided in an embodiment of the present invention. Detailed Implementation
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0020] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.
[0021] This invention discloses a method and system for evaluating the stability of modal parameters based on single-point excitation and multi-point response. These will be described in detail below.
[0022] The reliability of modal test results for large gas turbine rotors cannot be judged solely by the peak frequency response or a single natural frequency from a single test. Factors such as hoisting boundaries, hammer impact angle, sensor installation quality, signal noise, and data processing parameters can all affect modal identification results. Without repeatability and stability evaluation, even if the first three natural frequencies are obtained, it is difficult to prove that the results can be used for verifying the dynamic model of the rod rotor. Therefore, this invention proposes a method for evaluating the stability of modal parameters of large gas turbine rotors based on repeated single-point excitation and multi-point response tests. This method involves conducting at least three repeated static frequency tests under the same test boundary conditions and test conditions to identify the modal frequencies and mode shape vectors of the same order. Then, frequency deviation, mode shape correlation coefficient, modal shape deviation, mode shape node consistency, and frequency response function quality index are calculated to construct a modal parameter stability index. This method uses the stability evaluation results to quantitatively determine whether the modal parameters can be included in the verification process of the large gas turbine rotor dynamic model, providing data support for subsequent steps.
[0023] Figure 1 A flowchart of an evaluation method provided in an embodiment of the present invention is provided below. Figure 1 The modal parameter stability evaluation method based on single-point excitation and multi-point response provided in this embodiment of the invention includes: Step 1: Set the test boundary conditions, fix the test conditions under the boundary conditions, and conduct m repeated single-point excitation multi-point response modal tests based on the fixed test conditions, where m ≥ 3.
[0024] To address the lack of reliability in single-test results, this invention proposes conducting multiple repeated single-point excitation multi-point response modal tests to ensure the credibility of the experimental results. However, if the test conditions change during each test, the test results will also change. This makes it impossible to determine the specific reasons for the differences between multiple test results, thus affecting the assessment of stability.
[0025] To address the aforementioned issues, this invention proposes a method for conducting multiple experiments based on fixed boundary and experimental conditions. In other words, the boundary and experimental conditions are fixed before the experiment begins, thereby ensuring consistency of experimental conditions between each repeated experiment.
[0026] For example, in this embodiment, the boundary condition is set as a low-addition-constraint hoisting boundary. Under this boundary condition, the test conditions are set as follows: the same boundary state (e.g., in the application scenario of a large gas turbine rotor, the hoisting state of the rotor is set to be the same for each test), the same excitation point position, the same coordinates of the same N response measurement points, the same sensor installation method, and the same data acquisition channel configuration. Where N ≥ 2, for example, the value of N can be 27. When the value of N is too large, although it can improve the positioning accuracy, the computational complexity will also increase significantly; while when the value of N is too small, it may cause the spatial positioning accuracy to be too low, resulting in serious distortion.
[0027] Step 2: Perform validity screening on all experimental data. After each invalid test is removed, a single-point excitation multi-point response modal test is added, and the validity screening is performed on the supplementary experimental data until m valid experimental data are obtained.
[0028] To improve the reliability of stability evaluation, it is first necessary to ensure that all data used in the evaluation are valid. Therefore, this invention filters the collected data from m experiments, removing invalid data.
[0029] To avoid insufficient data for final calculations and reduced reliability of results, after each removal of outlier data, a supplementary data collection under the same experimental conditions was performed. At the same time, the supplementary data was subjected to validity screening to prevent unqualified data collection, until m valid experimental data were obtained.
[0030] In one embodiment, a secondary screening method is used to screen the experimental data to ensure that all data used in the final calculation are valid data. Figure 2 A flowchart illustrating data validity screening provided in this embodiment of the invention is provided below. Figure 2 The validity screening of data from m trials includes: Step 21: Perform a pre-check on the original test data for each test. If there is a double hit, the corresponding test data is determined to be invalid and discarded; otherwise, it means that the pre-check has passed.
[0031] First, determine if there are any data points of double-impact in the original experimental data. If so, the experimental data is deemed invalid and discarded; otherwise, it indicates that the experimental data has passed the preliminary inspection and can proceed to the next step of quality screening. Here, the presence of double-impact can be determined by detecting the time-domain waveform. For example, if the time-domain waveform shows a single-impact pulse, then there is no double-impact.
[0032] Step 22: Calculate the coherence function and frequency response function of the target frequency band for the test data that passed the pre-inspection. If both the coherence function and frequency response function meet the preset requirements, the corresponding test data is determined to be valid; otherwise, the corresponding test data is determined to be invalid and discarded.
[0033] To further improve data quality, the coherence function and frequency response function of the target frequency band are calculated for the test data that pass the pre-test, and the validity of the test data is determined based on the coherence function and frequency response function. For example, it is assumed that if the coherence function is greater than the coherence threshold, it means that the coherence function meets the preset requirements; if the frequency response peak is clear, it means that the frequency response function meets the preset requirements. When both are satisfied, the corresponding test data is determined to be valid; otherwise, the corresponding test data is determined to be invalid and discarded. In this way, the reliability of the input data can be fundamentally guaranteed, and invalid data can be avoided from contaminating the subsequent modal parameter extraction and stability evaluation stages.
[0034] The definition of coherence threshold and frequency response peak clarity can be set according to the actual use scenario. For example, the coherence threshold can be set to 0.9, and the frequency response peak clarity can be defined as the absence of multiple peaks within half power bandwidth. This application does not make specific limitations on this.
[0035] Step 3: Extract the r-th modal frequency and mode shape vector from each of the m valid tests, and normalize the mode shape vector.
[0036] A mode includes at least frequencies and mode shapes. In this step, the r-th order modal frequency and mode shape vectors are extracted for each valid test. Here, r-th order refers to the target modal order to be evaluated, ensuring that the same modal parameters are extracted for all valid tests in each stability evaluation. For example, when identifying the first three modal parameters, r can take values of 1, 2, and 3. When r=1, subsequent steps are performed to obtain the stability evaluation results for the first-order mode. Similarly, subsequent steps are performed for the second and third-order modes to obtain the stability evaluation results for each mode.
[0037] To eliminate the influence of sensor amplitude calibration or impact energy differences, the mode shape vector is normalized. In this embodiment, amplitude normalization is used to normalize the mode shape vector, as follows: ,in, This represents the r-th mode shape vector of the k-th valid test. This represents the i-th component of the r-th mode shape vector in the k-th valid test. That is, the maximum absolute value of each component in this mode shape vector is used as the normalization benchmark, so that the normalized mode shape vector elements take values in the range of [-1, 1].
[0038] Step 4: Calculate the relative frequency deviation based on the r-th modal frequency from m valid tests.
[0039] After obtaining the r-th modal frequencies of all valid experiments, the relative deviations of the same-order modal frequencies are calculated to evaluate the consistency of the identification of the same-order natural frequencies in repeated experiments.
[0040] In one embodiment, the relative frequency deviation is defined using the average value of the frequencies of the same order modes. Therefore, it is necessary to first calculate the r-th order mode frequency of m valid trials. average The relative frequency deviation can be calculated from the average value. ,in, This is an error term to avoid division by zero.
[0041] Step 5: Calculate the mode shape deviation based on the normalized results of the r-th order mode shape vectors from all valid tests.
[0042] After obtaining the normalized result of the mode shape vector, the mode shape deviation can be calculated in this step using the normalized result of the mode shape vector.
[0043] In one embodiment, the mode shape deviation is defined using the mode shape correlation coefficient between two tests of the same order mode. Figure 3 Please refer to the flowchart for calculating mode shape deviation provided in this embodiment of the invention. Figure 3 In this embodiment, calculating the mode shape deviation specifically includes: Step 51: Based on the normalized results of the r-th order mode vectors of all valid tests, calculate the mode correlation coefficient between any two valid tests. ,in, This represents the r-th mode shape vector of the p-th valid test. The normalization result, This represents the r-th mode shape vector of the q-th valid test. The normalized result. The closer the value is to 1, the more consistent the vibration modes obtained from the two experiments are.
[0044] Step 52: Take the average of the modal correlation coefficients of all pairwise combinations of the r-th order to obtain the mean value. .
[0045] Step 53, calculate the modal shape deviation based on the mean value. .
[0046] Assuming three trials are conducted, there are three possible combinations: (1,2), (1,3), and (2,3). Then, in the first mode, step 52 calculates the modal correlation coefficients between each pair of combinations. , and ,right , and Take the average to get the mean. Then the mode shape deviation is .
[0047] Step 6: Identify the node positions of the r-th mode shape for all valid tests and calculate the node position deviations.
[0048] In order to determine whether the modal parameters are stable, in addition to the frequency and mode shape, it is also necessary to see whether the node position has shifted. However, the frequency and mode shape alone cannot detect the node shift. Therefore, this embodiment introduces the calculation of the node position deviation.
[0049] To calculate the nodal position deviation, it is first necessary to identify the nodal positions of the mode shape. In this embodiment, the nodal position is defined as the axial position where the mode shape curve crosses zero or has the minimum amplitude. The nodal positions are identified based on the normalized result of the r-th order mode shape vector obtained from each experiment, for example, through linear interpolation. Once the nodal positions are obtained, the nodal position deviation can be calculated. Where L represents the range of axial measuring points on the rotor. and These represent the mode node positions of the p-th and q-th valid tests, respectively.
[0050] Step 7: Calculate the stability index based on the relative frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate of the target frequency band.
[0051] Once the above deviation indicators are obtained, all indicators can be incorporated into the stability index. However, it should be noted that in calculating the stability index, in addition to the above indicators, the coherence function pass rate of the target frequency band also needs to be considered.
[0052] Figure 4 For another flowchart of the evaluation method provided in this embodiment of the invention, please refer to... Figure 4 Before calculating the stability index, step 61 is included, which calculates the coherence function pass rate of the target frequency band. To calculate the coherence function pass rate of the target frequency band, first, the number of frequency points within the target frequency band that meet the coherence function requirements is calculated. Then, the proportion of the number of frequency points that meet the coherence function requirements to the total number of frequency points is calculated to obtain the coherence function pass rate of the target frequency band. The coherence function requirements can be specifically set according to the actual usage scenario; this invention does not impose specific limitations on this.
[0053] This invention uses the coherence function pass rate of the target frequency band as a confidence index for frequency relative deviation, mode shape deviation, and node position deviation. Together with frequency relative deviation, mode shape deviation, and node position deviation, it constitutes a stability index for modal parameters. This not only avoids the risk of misjudgment by a single index, but also prevents the results from being artificially high due to data defects (such as an abnormality in a certain index being masked by other defects), thereby ensuring the authenticity of the final results.
[0054] In this embodiment, the stability index is defined as: ,in, , , , These represent the weights of the corresponding indicators. For normalization function, Indicates the relative frequency deviation. Indicates mode shape deviation. Indicates the deviation of the node position. This represents the pass rate of the coherence function in the target frequency band.
[0055] Step 8: Determine whether the stability index meets the stability requirements. If it does, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable, and the process returns to setting experimental boundary conditions for re-examination.
[0056] Assuming the stability threshold is T, the stability index is compared with T. If the stability index is greater than T, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable, and the test boundary conditions are set for re-examination.
[0057] The modal parameter stability evaluation method based on single-point excitation and multi-point response provided by this invention repeatedly executes single-point excitation and multi-point response tests under fixed experimental conditions, and identifies the same-order modal frequencies and mode shape vectors respectively. The modal parameter stability index is constructed by using the same-order frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate as a multi-index evaluation. The stability evaluation results are used as a prerequisite criterion for modal parameters to enter the verification of large gas turbine rotor dynamics models, realizing a quantitative judgment on whether modal parameters can enter the verification process of large gas turbine rotor dynamics models.
[0058] Based on the same inventive concept, this invention also provides a modal parameter stability evaluation system based on single-point excitation and multi-point response. Figure 5 Please refer to the schematic diagram of an evaluation system provided in an embodiment of the present invention. Figure 5 The present invention provides a modal parameter stability evaluation system 100 based on single-point excitation and multi-point response, comprising: The setting module 110 is used to set the test boundary conditions, fix the test conditions under the boundary conditions, and perform m repeated single-point excitation multi-point response modal tests based on the fixed test conditions, where m ≥ 3; The filtering module 120 is used to filter the validity of all experimental data. After each invalid test is removed, a single-point excitation multi-point response modal test is added, and the validity of the added experimental data is filtered until m valid experimental data are obtained. The extraction module 130 is used to extract the r-th modal frequency and mode shape vector from m valid tests, and to normalize the mode shape vector. The calculation module 140 is used to calculate the relative frequency deviation based on the r-th modal frequency of m valid tests; it is also used to calculate the mode shape deviation based on the normalized result of the r-th mode shape vector of all valid tests. The identification module 150 is used to identify the node positions of the r-th mode shape in all valid tests and to calculate the node position deviations. The calculation module 140 is also used to calculate the stability index based on the relative frequency deviation, mode shape deviation, node position deviation and the coherence function qualification rate of the target frequency band; The judgment module 160 is used to determine whether the stability index meets the stability requirements. If it does, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable and the system returns to set experimental boundary conditions for re-examination.
[0059] The above system embodiments correspond to the method embodiments and have the same technical effects as the method embodiments. For details, please refer to the method embodiments, which will not be repeated here.
[0060] Optionally, the filtering module 120 is specifically configured as follows: The raw test data for each experiment is pre-checked. If there are two consecutive hits, the corresponding test data is deemed invalid and discarded; otherwise, the pre-check is passed. For the test data that passes the pre-inspection, calculate the coherence function and frequency response function of the target frequency band. If both the coherence function and frequency response function meet the preset requirements, the corresponding test data is deemed valid; otherwise, the corresponding test data is deemed invalid and discarded.
[0061] Optionally, the extraction module 130 is specifically configured as follows: The mode shape vector is normalized using amplitude normalization: ,in, This represents the r-th mode shape vector of the k-th valid test. It represents the i-th component of the r-th mode vector of the k-th valid test.
[0062] Optionally, the computing module 140 is specifically configured as follows: Calculate the r-th modal frequency of m valid trials average ; Calculate the relative frequency deviation based on the average value. ,in, This is the error term.
[0063] Optionally, the computing module 140 is specifically configured as follows: Based on the normalized results of the r-th order mode vectors from all valid experiments, the mode correlation coefficient between any two valid experiments is calculated. ,in, This represents the r-th mode shape vector of the p-th valid test. The normalization result, This represents the r-th mode shape vector of the q-th valid test. The normalization result; The mean value is obtained by averaging the modal correlation coefficients of all pairwise combinations of the r-th order. ; The modal deviation is calculated based on the mean value. .
[0064] Optionally, the recognition module 150 is specifically configured as follows: Based on the normalized result of the r-th order mode shape vector obtained from each test, the mode shape node positions are identified; the node positions are the axial positions where the mode shape curve crosses zero or has the minimum amplitude. Calculate the node position deviation based on the node position. Where L represents the range of axial measuring points on the rotor. and These represent the mode node positions of the p-th and q-th valid tests, respectively.
[0065] Optionally, the computing module 140 is specifically configured as follows: Calculate the number of frequency points within the target frequency band that satisfy the coherence function requirements; The proportion of frequency points that meet the coherence function requirements to the total number of frequency points is calculated to obtain the coherence function qualification rate of the target frequency band.
[0066] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.
[0067] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.
[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for evaluating the stability of modal parameters based on single-point excitation and multi-point response, characterized in that, include: Set experimental boundary conditions, fix the experimental conditions under the boundary conditions, and conduct m repeated single-point excitation multi-point response modal tests based on the fixed experimental conditions, where m ≥ 3; All experimental data are subjected to validity screening. After each invalid test is removed, a single-point excitation multi-point response modal test is added, and the validity screening is performed on the supplementary experimental data until m valid experimental data are obtained. For each of the m valid tests, the r-th modal frequency and mode shape vector are extracted, and the mode shape vector is normalized. Calculate the relative frequency deviation based on the rth modal frequency from m valid tests; Based on the normalized results of the r-th order mode shape vectors from all valid tests, the mode shape deviation is calculated. Identify the node positions of the r-th mode shape in all valid tests and calculate the node position deviations; The stability index is calculated based on the relative frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate of the target frequency band. Determine whether the stability index meets the stability requirements. If it does, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable, and the test boundary conditions are set for re-examination.
2. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, The test conditions include at least the following: boundary state, excitation point location, coordinates of N response measurement points, sensor installation method, and data acquisition channel configuration; wherein, N ≥ 2.
3. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, The validity screening of data from m trials includes: The raw test data for each experiment is pre-checked. If there are two consecutive hits, the corresponding test data is deemed invalid and discarded; otherwise, the pre-check is passed. The coherence function and frequency response function of the target frequency band are calculated for the test data that pass the pre-inspection. If both the coherence function and the frequency response function meet the preset requirements, the corresponding test data is determined to be valid; otherwise, the corresponding test data is determined to be invalid and discarded.
4. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, The normalization process for the mode shape vectors specifically includes: The mode shape vector is normalized using amplitude normalization: ,in, This represents the r-th mode shape vector of the k-th valid test. It represents the i-th component of the r-th mode shape vector of the k-th valid test.
5. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, The relative frequency deviation is calculated based on the r-th modal frequency from m valid tests, specifically including: Calculate the r-th modal frequency of m valid trials average ; Calculate the relative frequency deviation based on the average value. ,in, This is the error term.
6. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, Based on the normalized results of the r-th order mode shape vectors from all valid experiments, the mode shape deviation is calculated, specifically including: Based on the normalized results of the r-th order mode vectors from all valid experiments, the mode correlation coefficient between any two valid experiments is calculated. ,in, This represents the r-th mode shape vector of the p-th valid test. The normalization result, This represents the r-th mode shape vector of the q-th valid test. The normalization result; The mean value is obtained by averaging the modal correlation coefficients of all pairwise combinations of the r-th order. ; Based on the mean value, the modal shape deviation is calculated as follows: .
7. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, Identify the nodal positions of the r-th mode shape in all valid tests and calculate the nodal position deviations, specifically including: Based on the normalized result of the r-th order mode shape vector obtained from each test, the mode shape node positions are identified; the node positions are the axial positions where the mode shape curve crosses zero or has the minimum amplitude. Calculate the node position deviation based on the node position. Where L represents the range of axial measuring points on the rotor. and These represent the mode node positions of the p-th and q-th valid tests, respectively.
8. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, The stability index is expressed as: ,in, , , , These represent the weights of the corresponding indicators. For normalization function, Indicates the relative frequency deviation. Indicates mode shape deviation. Indicates the deviation of the node position. This represents the pass rate of the coherence function in the target frequency band.
9. The modal parameter stability evaluation method based on single-point excitation and multi-point response according to claim 1, characterized in that, Before calculating the stability index, the coherence function qualification rate of the target frequency band is also calculated, which specifically includes: Calculate the number of frequency points within the target frequency band that satisfy the coherence function requirements; The proportion of frequency points that meet the coherence function requirements to the total number of frequency points is calculated to obtain the coherence function qualification rate of the target frequency band.
10. A modal parameter stability evaluation system based on single-point excitation and multi-point response, characterized in that, include: The setting module is used to set the test boundary conditions, fix the test conditions under the boundary conditions, and perform m repeated single-point excitation multi-point response modal tests based on the fixed test conditions, where m ≥ 3; The filtering module is used to filter the validity of all experimental data. After each invalid experiment is removed, a single-point excitation multi-point response modal experiment is added, and the validity of the added experimental data is filtered until m valid experimental data are obtained. The extraction module is used to extract the r-th modal frequency and mode shape vector from m valid tests, and to normalize the mode shape vector. The calculation module is used to calculate the relative frequency deviation based on the r-th modal frequency of m valid tests; it is also used to calculate the mode shape deviation based on the normalized result of the r-th mode shape vector of all valid tests. The identification module is used to identify the node positions of the r-th mode shape in all valid tests and to calculate the node position deviations. The calculation module is also used to calculate the stability index based on the relative frequency deviation, mode shape deviation, node position deviation, and coherence function qualification rate of the target frequency band. The judgment module is used to determine whether the stability index meets the stability requirements. If it does, the r-th modal parameter is determined to be stable and can be used for dynamic model verification; otherwise, the r-th modal parameter is determined to be unstable and the system returns to set experimental boundary conditions for re-examination.
