A vibration response prediction method, device, system and medium of a precision equipment

Abstract

The application discloses a kind of precision equipment vibration response prediction method, device, system and computer readable storage medium, applied to superconducting quantum technical field.The method comprises: the benchmark point of precision equipment is tested for vibration response, and the vibration response data of benchmark point is obtained;Based on the vibration response data of benchmark point, the vibration response data of benchmark point is analyzed using the neural network model related to benchmark point and target test point established in advance, and the vibration response data of target test point corresponding to benchmark point is obtained;Wherein, benchmark point is located on the surface of precision equipment, and target test point is located at the internal preset area of precision equipment;The vibration response data of target prediction point in extremely low temperature position can be predicted by vibration response test to benchmark point at normal temperature, combined with trained neural network model, to realize the vibration measurement of precision equipment under extremely low temperature condition.

Description

Technical Field

[0001] This invention relates to the field of superconducting quantum technology, and in particular to a method, apparatus, system, and computer-readable storage medium for predicting the vibration response of precision equipment. Background Technology

[0002] In recent years, superconducting technology has been increasingly widely used in scientific research and industry, such as in superconducting quantum computers, nuclear fusion experimental devices, high-energy particle accelerators, superconducting magnet energy storage systems, magnetohydrodynamics (MHD), strong magnetic field research, chemical analysis (Nuclear Magnetic Resonance Spectroscopy), medicine (MRI), transportation systems (maglev trains), superconducting power transmission, and superconducting motors. It has developed into a practical technology with a certain scale and great prospects.

[0003] The development and scientific research of the aforementioned technologies all require an extremely low temperature cooling system environment, typically down to the K level, or even the mK level. Further research indicates that the level needs to be down to the nK or even pK level. The temperature stability of this extremely low temperature cooling system directly affects the normal operation of the entire device. This extremely low temperature cooling system is a precision device with very high requirements for processing and manufacturing processes, and it is extremely susceptible to the influence of external environmental vibrations. It usually needs to be kept at a very low vibration level (below μm). Therefore, various vibration reduction and isolation measures need to be taken and real-time monitoring is required. However, direct vibration testing at extremely low temperatures is currently a challenge.

[0004] Taking superconducting quantum computers as an example, their chip systems and microwave components require operation at extremely low temperatures. The establishment of their cryogenic cooling systems typically employs dilution refrigerators, which are precision devices. Through staged cooling by the dilution refrigerator, the extremely low temperature of the experimental area (usually around 10 mK) is ultimately achieved. Because the chip systems and microwave devices have extremely high requirements for temperature stability, even small vibrations can cause temperature drift. Furthermore, long-term vibration can easily cause resonance in various components, leading to loosening of screws or nuts, ultimately affecting the normal operation of the precision equipment. Currently, there are no suitable vibration sensors to monitor vibration in this extremely low-temperature environment. Therefore, during experimental testing, it is impossible to promptly identify whether the performance degradation of vibration-sensitive components is caused by vibration or other factors, seriously affecting the continuous long-term operation of the entire experimental system and the identification of effective data during experiments.

[0005] In existing technologies, to reduce the impact of vibration and noise on experimental test devices, vibration sources are kept as far away from the main body of the precision instrument as possible. Alternatively, damping layers are added to reduce the transmission of environmental vibration between different components of the precision instrument. However, since the test specimen or actual research object is installed in the lowest temperature environment cavity of the precision equipment, existing vibration sensors cannot be used for testing at such extreme temperatures. Therefore, it is impossible to determine the magnitude of the vibration level ultimately transmitted from the internal vibration source or environmental vibration to the vicinity of the experimental area during the experiment. Real-time vibration response monitoring is not feasible, nor can the impact of vibration on the entire experiment be predicted to provide timely warnings or adjustments.

[0006] Therefore, how to provide a method, apparatus, system, and computer-readable storage medium capable of predicting the vibration response of precision equipment has become a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a method, device, system, and computer-readable storage medium for predicting the vibration response of precision equipment, which can realize vibration measurement of precision equipment under extremely low temperature conditions during use.

[0008] To address the aforementioned technical problems, embodiments of the present invention provide a method for predicting the vibration response of precision equipment, comprising:

[0009] Vibration response tests were performed on the reference points of the precision equipment to obtain the vibration response data of the reference points.

[0010] Based on the vibration response data of the reference point, the vibration response data of the reference point is analyzed using a pre-established neural network model related to the reference point and the target test point to obtain the vibration response data of the target test point corresponding to the reference point; wherein, the reference point is located on the surface of the precision equipment, and the target test point is located in a preset area inside the precision equipment.

[0011] Optionally, the process of establishing the neural network model related to the benchmark point and the target test point includes:

[0012] An initial finite element model was established based on the 3D model of the precision equipment.

[0013] Based on the physical model of the precision equipment, experimental modal tests were conducted to obtain the experimental modal test results;

[0014] The initial finite element model is corrected using the test modal results to obtain the corrected calibrated finite element model.

[0015] Based on the pre-acquired environmental vibration acceleration data corresponding to the precision equipment and the calibration finite element model, a neural network training is performed to obtain a neural network model related to the reference point and the target test point.

[0016] Optionally, the modal testing based on the physical model of the precision equipment to obtain the modal test results includes:

[0017] A test modal wireframe model was constructed based on the physical model of the precision equipment.

[0018] Data was collected from the experimental modal wireframe model to obtain acceleration response data for each measuring point under different input excitations;

[0019] The test modal results are obtained based on all the acceleration response data and input excitation data. The test modal results include modal frequencies, damping and mode shapes under multiple modes.

[0020] Optionally, the input excitation is a hammer excitation or a vibrator excitation.

[0021] Optionally, the step of using the experimental modal test results to correct the initial finite element model to obtain a corrected calibrated finite element model includes:

[0022] The finite element calculation results are obtained by using an initial finite element model. The finite element calculation results include multiple sets of modal frequencies and corresponding mode shapes.

[0023] The finite element calculation results and the experimental modal test results are subjected to condensation processing to obtain condensed finite element calculation results and experimental modal test results;

[0024] Error analysis and sensitivity analysis are performed on the frequencies and mode shapes in the finite element calculation results after polycondensation and the experimental modal test results to determine the various parameters to be corrected in the finite element model;

[0025] The parameters to be corrected are iteratively corrected by a correction iteration until the iteration converges and the corrected parameters are obtained.

[0026] The corrected finite element model is obtained based on the corrected parameters;

[0027] The modified finite element model is subjected to error verification of mode shape and frequency. If the verification is successful, the modified finite element model is used as the calibration finite element model.

[0028] Optionally, the step of training a neural network based on pre-acquired environmental vibration acceleration data corresponding to the precision equipment and the calibration finite element model to obtain a neural network model related to the reference point and the target test point includes:

[0029] Vibration tests were conducted on the laboratory where the precision equipment was located to obtain environmental vibration acceleration data;

[0030] The environmental vibration acceleration data is used as the input excitation of the calibrated finite element model to obtain vibration response sample data for each reference point and each target test point.

[0031] A training set is constructed based on the vibration response sample data of each of the reference points and the vibration response sample data of each of the target test points;

[0032] The neural network is trained using the training set to obtain a neural network model that reflects the correspondence between each benchmark point and the corresponding target test point.

[0033] This invention also provides a vibration response prediction device for precision equipment, comprising:

[0034] The testing module is used to perform vibration response tests on the reference points of precision equipment and obtain the vibration response data of the reference points.

[0035] The analysis module is used to analyze the vibration response data of the reference point based on the vibration response data of the reference point, using a neural network model pre-established by the establishment module related to the reference point and the target test point, to obtain the vibration response data of the target test point corresponding to the reference point; wherein, the reference point is located on the surface of the precision equipment, and the target test point is located in a preset area inside the precision equipment.

[0036] Optionally, the establishment module includes:

[0037] The first establishment unit is used to establish an initial finite element model based on the 3D model of the precision equipment;

[0038] The testing unit is used to perform experimental modal tests based on the physical model of the precision equipment and obtain the experimental modal test results.

[0039] The correction unit is used to correct the initial finite element model using the test modal test results to obtain the corrected calibrated finite element model;

[0040] The training unit is used to train a neural network based on the pre-acquired environmental vibration acceleration data corresponding to the precision equipment and the calibration finite element model, so as to obtain a neural network model related to the reference point and the target test point.

[0041] Optionally, the training unit includes:

[0042] The testing subunit is used to conduct vibration tests on the laboratory where the precision equipment is located to obtain environmental vibration acceleration data.

[0043] The calculation subunit is used to take the environmental vibration acceleration data as the input excitation of the calibrated finite element model to obtain vibration response sample data for each reference point and each target test point.

[0044] Construct sub-units to build training sets based on vibration response sample data of each of the reference points and vibration response sample data of each of the target test points;

[0045] The training subunit is used to train the neural network using the training set to obtain a neural network model that reflects the correspondence between each benchmark point and the corresponding target test point.

[0046] This invention also provides a vibration response prediction system for precision equipment, comprising:

[0047] Memory, used to store computer programs;

[0048] A processor is used to execute the computer program to implement the steps of the vibration response prediction method for precision equipment as described above.

[0049] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the vibration response prediction method for precision equipment as described above.

[0050] This invention also provides a method, apparatus, system, and computer-readable storage medium for predicting the vibration response of precision equipment, comprising: performing vibration response tests on a reference point of the precision equipment to obtain vibration response data of the reference point; and analyzing the vibration response data of the reference point using a pre-established neural network model related to the reference point and a target test point to obtain vibration response data of the target test point corresponding to the reference point; wherein the reference point is located on the surface of the precision equipment, and the target test point is located in a preset area inside the precision equipment.

[0051] As can be seen, in this embodiment of the invention, a neural network model related to the reference point and the target test point is pre-established. Then, vibration response data of the reference point is obtained by performing vibration response tests on the reference point of the precision equipment. This vibration response data is then input into the trained neural network model. The neural network model analyzes the vibration response data of the reference point and outputs the vibration response data of the target test point corresponding to the reference point. In other words, this invention can predict the vibration response data of the target test point at an extremely low temperature by performing vibration response tests on the reference point at room temperature and combining the trained neural network model, thus realizing the vibration measurement of precision equipment under extremely low temperature conditions. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the prior art and embodiments 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.

[0053] Figure 1 A flowchart illustrating a vibration response prediction method for precision equipment provided in an embodiment of the present invention;

[0054] Figure 2 A schematic diagram illustrating the calibration process of a finite element model provided in an embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram illustrating the training process of a neural network model provided in an embodiment of the present invention;

[0056] Figure 4 A schematic diagram of the structure of a vibration response prediction device for precision equipment provided in an embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram of the structure of a vibration response prediction system for precision equipment provided in an embodiment of the present invention. Detailed Implementation

[0058] This invention provides a method, apparatus, system, and computer-readable storage medium for predicting the vibration response of precision equipment, which enables vibration measurement of precision equipment under extremely low temperature conditions during use.

[0059] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0060] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a vibration response prediction method for precision equipment provided in an embodiment of the present invention. The method includes:

[0061] S110: Perform vibration response tests on the reference points of precision equipment to obtain vibration response data of the reference points;

[0062] It should be noted that, in the embodiments of the present invention, each reference point and each target test point on the precision equipment can be predetermined. Each reference point is located on the surface of the precision equipment, that is, a position that can be measured at room temperature. Each target test point is located in a preset area inside the precision equipment. Specifically, it can be based on the low-temperature position when using diluted refrigerant in actual applications, and it is the position that needs to be measured. These positions determine each target test point. Then, a neural network model that can reflect the vibration response relationship between each reference point and the corresponding target test point is established. When it is necessary to perform vibration response testing on each target test point on the precision equipment, the vibration response data of the reference point can be obtained by performing vibration response testing on the reference point of the precision equipment.

[0063] S120: Based on the vibration response data of the reference point, the vibration response data of the reference point is analyzed using a pre-established neural network model related to the reference point and the target test point to obtain the vibration response data of the target test point corresponding to the reference point; wherein, the reference point is located on the surface of the precision equipment, and the target test point is located in a preset area inside the precision equipment.

[0064] Specifically, after obtaining the vibration response data of the reference point, the vibration response data of the reference point is used as the input of the trained neural network. The neural network analyzes the vibration response data of the reference point and outputs the vibration response data of the target test point corresponding to the reference point, thereby realizing the vibration response test of the target test point.

[0065] Furthermore, the process of establishing the aforementioned neural network model related to the benchmark and target test points may specifically include:

[0066] An initial finite element model is established based on the 3D model of the precision equipment;

[0067] Experimental modal tests were conducted based on the physical model of the precision equipment to obtain the experimental modal test results;

[0068] The initial finite element model was corrected using the experimental modal test results to obtain the corrected calibrated finite element model;

[0069] Based on the pre-acquired environmental vibration acceleration data corresponding to the precision equipment and the calibrated finite element model, a neural network model related to the benchmark point and the target test point is obtained through neural network training.

[0070] It should be noted that a 3D model of a precision device can be obtained, and then a corresponding initial finite element model can be established based on the 3D model of the precision device, including determining the initial values ​​of each parameter of the initial finite element model.

[0071] Then, based on the physical model of the precision equipment, experimental modal tests are performed to obtain the experimental modal test results. This process may specifically include:

[0072] Construct experimental modal wireframe models based on the physical models of precision equipment;

[0073] Data was collected from the experimental modal wireframe model to obtain the acceleration response data of each measuring point under different input excitations;

[0074] The test modal results are obtained based on all acceleration response data and input excitation data. The test modal results include modal frequencies, damping and mode shapes under multiple modes.

[0075] It should be noted that in practical applications, a corresponding experimental modal control-by-wire model can be established based on the physical model of the precision equipment. Specifically, the boundary conditions of the physical model of the precision equipment can be determined first, including the modal type, such as free model or elastic mode. The boundary conditions of the physical model of the precision equipment can be used as the boundary conditions for realizing the modal control-by-wire model. The overall coordinate system used for the test can also be identified, and the input excitation method (such as hammer excitation or vibrator excitation) can be further determined. According to the actual test requirements, the appropriate sensor type and installation method can be selected. During vibration testing, the vibration acceleration data of the installation point (i.e., the measuring point) can be obtained through the sensor. Then, the strategy degrees of freedom can be determined according to the actual test requirements, including the position, number, and direction of the measuring points. The geometric model of the experimental modal control-by-wire model can be generated based on the determined degrees of freedom. Connect the sensors to the data acquisition system to collect data from the experimental modal control-by-wire model. Pre-acquisition can be performed first to determine reasonable parameters (e.g., sampling rate, acquisition instrument range setting, sampling duration, etc.). Then, after determining the parameters, formal acquisition is conducted, which involves collecting acceleration response data at various measurement points under different input excitations. The modal frequencies, damping, and mode shapes under multiple modes are extracted from all the acceleration response data as the experimental modal test results. After obtaining the modal frequencies, damping, and mode shapes under multiple modes, the extracted information can be verified. For example, it can be verified using modal judgment criteria (MAC), modal participation (MP), modal phase collinearity, and modal confidence factors. If the verification is successful, for example, if the correlation between different modes reaches a preset value, it indicates that there is no overlap between the different modes. The data obtained at this point, combined with the corresponding input excitation data, can be used as the final experimental modal test results.

[0076] After obtaining the test results of the experimental modalities, in order to improve the accuracy of the final trained neural network, the initial finite element model can be corrected using the test results of the experimental modalities in this embodiment of the invention to obtain the corrected calibrated finite element model, and then the subsequent neural network training can be carried out based on the corrected calibrated finite element model.

[0077] Furthermore, the process of correcting the initial finite element model using experimental modal test results to obtain the corrected calibrated finite element model can specifically include:

[0078] The finite element calculation results are obtained by using the initial finite element model. The finite element calculation results include multiple sets of modal frequencies and corresponding mode shapes.

[0079] The finite element calculation results and experimental modal test results are subjected to condensation processing to obtain condensed finite element calculation results and experimental modal test results;

[0080] Error analysis and sensitivity analysis were performed on the frequencies and mode shapes in the finite element calculation results and experimental modal test results after polycondensation to determine the various parameters to be corrected in the finite element model.

[0081] The parameters to be corrected are iteratively corrected by a correction iteration until the iteration converges and the corrected parameters are obtained.

[0082] The corrected finite element model is obtained based on the corrected parameters;

[0083] The modified finite element model is subjected to error verification of mode shape and frequency. If the verification is successful, the modified finite element model is used as the calibration finite element model.

[0084] It should be noted that, please refer to Figure 2In this embodiment of the invention, finite element calculation results are first obtained based on an initial finite element model. Specifically, the boundary conditions of the initial finite element model are consistent with the boundary conditions of the experimental modal wireframe model, and the material properties, component connection relationships, etc., are consistent with the physical model of the precision equipment. The obtained finite element calculation results can include the frequencies and mode shapes of each order, that is, multiple sets of frequencies and mode shapes are obtained. Then, the obtained finite element calculation results and experimental modal test results are subjected to condensation processing. Specifically, the condensation function built into the finite element software or the experimental testing software can be used to condense the finite element calculation results and experimental modal test results respectively, to obtain condensed finite element calculation results and experimental modal test results. For example, the condensed data includes data of a preset number of orders (e.g., frequency, stiffness, damping, mode shape, etc.). These data can well express the physical model of the precision equipment. The frequencies and mode shapes in the finite element calculation results and experimental modal test results after polycondensation are sorted from smallest to largest frequency, and a model matching (MAC) analysis is performed. Specifically, the similarity between the mode shapes in the finite element calculation results and the mode shapes in the experimental modal test results at each frequency is calculated through MAC analysis. Then, by reading the material parameters, spring stiffness, bolt prestress, and other data of the initial finite element model, parameters that need to be calibrated are initially screened, and upper and lower limits for each parameter are further set. The equivalent target order and weight coefficients are selected, and the sensitivity matrix of the design parameters to the equivalent target is calculated and displayed, thereby realizing the sensitivity analysis of each parameter. Based on the sensitivity analysis results, the parameters to be corrected are determined. Positive parameters (i.e., design parameters to be corrected); after determining each parameter to be corrected, iterative correction is performed on each parameter. Specifically, the maximum number of iterations and iteration compensation can be set, and then iterative calculations are performed. During the iteration process, iterative process diagrams are displayed, including parameter convergence diagrams, optimization target convergence diagrams, MAC iteration diagrams, sensitivity matrix diagrams for each step, etc. Then, based on the iterative convergence diagrams, it can be further determined whether each parameter to be corrected has converged, that is, whether it has converged to a corresponding constant value. If so, the corrected parameters are obtained, and the corrected finite element model is further determined based on the corrected parameters. If it has not converged, the process returns to the finite element calculation results and starts the correction process again. In addition, after obtaining the corrected parameters and the corrected finite element model, a quality inspection can be performed to further ensure the accuracy of the finite element model. Specifically, the corrected finite element model can be re-performed with MAC calculation analysis, and the mode shapes and frequencies can be checked for errors. If the inspection passes, it means that the consistency between the corrected finite element model and the experimental modal wireframe model can reach the required accuracy, and the corrected finite element model is used as the final calibration finite element model. Otherwise, return to the step of obtaining the finite element calculation results from the initial finite element model and start the calibration again.

[0085] Furthermore, the process of training a neural network based on pre-acquired environmental vibration acceleration data corresponding to the precision equipment and a calibrated finite element model to obtain a neural network model related to the reference point and the target test point can specifically include:

[0086] Vibration tests were conducted on the laboratory where the precision equipment was located to obtain environmental vibration acceleration data;

[0087] Using environmental vibration acceleration data as input excitation for calibrating the finite element model, vibration response sample data for each reference point and each target test point are obtained.

[0088] A training set is constructed based on the vibration response sample data of each benchmark point and the vibration response sample data of each target test point;

[0089] The neural network is trained using a training set to obtain a neural network model that reflects the correspondence between each benchmark point and the corresponding target test point.

[0090] It should be noted that in this embodiment of the invention, multiple reference points and multiple target test points of the precision equipment can be predetermined. Each reference point is located on the surface of the precision equipment, i.e., a position measurable at room temperature. Each target test point is located within the precision equipment at an extremely low temperature, and is the position to be measured. These positions determine each target test point. Specifically, vibration tests are performed on the laboratory containing the precision equipment to obtain multiple sets of environmental vibration acceleration data. This environmental vibration acceleration data is then used as the input excitation for calibrating the finite element model, thereby obtaining vibration response sample data for each reference point and each target test point under each environmental vibration acceleration. Based on the vibration response sample data of each reference point and each target test point, a training set (xi, yi) is constructed, where xi represents the vibration response sample data of the i-th reference point (i.e., the input data), and yi represents the desired output, specifically the vibration response sample data of the i-th target test point, which is the theoretical value of the target prediction point. Please refer to... Figure 3The system takes a training set (xi, yi) as input, predicts values ​​using a backpropagation (BP) neural network, compares these predicted values ​​with the corresponding theoretical values ​​to obtain the mean squared error (MSE), and updates the BP neural network parameters based on this MSE. It then checks if the training set iteration is complete. If not, it returns to the process of obtaining predicted values ​​using the BP neural network for the next set of data in the training set and performs another round of training and updates until the training set iteration is complete. Finally, it checks if the final MSE meets a preset requirement. If it does, it obtains the final BP neural network parameters and a trained neural network model reflecting the correspondence between each benchmark point and its corresponding target test point. If it does not meet the preset requirement, it returns to the training set and retrains until the preset requirement is met, resulting in the final neural network model. Finally, it analyzes the vibration response data of the benchmark points using this trained neural network model and outputs the vibration response data of the target test points corresponding to those benchmark points.

[0091] As can be seen, in this embodiment of the invention, a neural network model related to the reference point and the target test point is pre-established. Then, vibration response data of the reference point is obtained by performing vibration response tests on the reference point of the precision equipment. This vibration response data is then input into the trained neural network model. The neural network model analyzes the vibration response data of the reference point and outputs the vibration response data of the target test point corresponding to the reference point. That is, by performing vibration response tests on the reference point at room temperature and combining the trained neural network model, this invention can predict the vibration response data of the target prediction point at an extremely low temperature location, thereby realizing vibration measurement of any position of precision equipment under extremely low temperature conditions.

[0092] Based on the above embodiments, this invention also provides a vibration response prediction device for precision equipment, please refer to [link / reference needed]. Figure 4 The device includes:

[0093] Test module 11 is used to perform vibration response tests on the reference points of precision equipment and obtain vibration response data of the reference points;

[0094] Analysis module 12 is used to analyze the vibration response data of the reference point based on the vibration response data of the reference point and to obtain the vibration response data of the target test point corresponding to the reference point by using the neural network model related to the reference point and the target test point pre-established by the establishment module 13. The reference point is located on the surface of the precision equipment, and the target test point is located in a preset area inside the precision equipment.

[0095] Furthermore, the aforementioned module 13 includes:

[0096] The first establishment unit is used to establish an initial finite element model based on the 3D model of the precision equipment;

[0097] The test unit is used to perform experimental modal tests based on the physical model of precision equipment and obtain experimental modal test results;

[0098] The correction element is used to correct the initial finite element model using the experimental modal test results, so as to obtain the corrected calibrated finite element model.

[0099] The training unit is used to train a neural network based on pre-acquired environmental vibration acceleration data corresponding to the precision equipment and a calibrated finite element model, so as to obtain a neural network model related to the reference point and the target test point.

[0100] Furthermore, the training units include:

[0101] The testing subunit is used to conduct vibration tests on the laboratory where the precision equipment is located, and to obtain environmental vibration acceleration data.

[0102] The computational sub-unit is used to take environmental vibration acceleration data as input excitation for calibrating the finite element model and obtain vibration response sample data for each reference point and each target test point.

[0103] Construct sub-units to build training sets based on vibration response sample data of each benchmark point and vibration response sample data of each target test point;

[0104] The training subunit is used to train the neural network using the training set to obtain a neural network model that reflects the correspondence between each benchmark point and the corresponding target test point.

[0105] It should be noted that the vibration response prediction device for precision equipment provided in the embodiments of the present invention has the same beneficial effects as the vibration response prediction method for precision equipment provided in the above embodiments. For a detailed description of the vibration response prediction method for precision equipment involved in the embodiments of the present invention, please refer to the above embodiments. The present invention will not repeat the details here.

[0106] like Figure 5 As shown, based on the above embodiments, this invention also provides a vibration response prediction system for precision equipment, comprising:

[0107] Memory 20 is used to store computer programs;

[0108] The processor 21 is used to execute a computer program to implement the steps of the vibration response prediction method for precision equipment as described above.

[0109] The electronic devices provided in this embodiment may include, but are not limited to, smartphones, tablets, laptops, or desktop computers.

[0110] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0111] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 20 is used to store at least the following computer program 201, which, after being loaded and executed by the processor 21, is capable of implementing the relevant steps of the vibration response prediction method for precision equipment disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, and the storage method may be temporary or permanent storage. The operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, set offsets.

[0112] In some embodiments, the electronic device may further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.

[0113] Those skilled in the art will understand that Figure 5 The structures shown do not constitute a limitation on electronic devices and may include more or fewer components than those shown.

[0114] It is understood that if the vibration response prediction method for precision equipment in the above embodiments is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes: USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, magnetic disk, or optical disk, and other media capable of storing program code.

[0115] Based on this, and building upon the above embodiments, this invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the vibration response prediction method for precision equipment as described above.

[0116] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0117] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0118] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0119] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0120] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting the vibration response of precision equipment, characterized in that, The method comprises the following steps: vibration response test is performed on a reference point of a precision device to obtain vibration response data of the reference point; based on the vibration response data of the reference point, a neural network model previously established in relation to the reference point and a target test point is used to analyze the vibration response data of the reference point to obtain vibration response data of the target test point corresponding to the reference point; wherein the reference point is located on the surface of the precision device, the target test point is located at a preset region inside the precision device, and the target test point is a low-temperature position that needs to be measured when dilute refrigerant is used; wherein: the process of establishing the neural network model in relation to the reference point and the target test point comprises: an initial finite element model is established based on a 3D model of the precision device; experimental modal testing is performed based on a physical model of the precision device to obtain experimental modal testing results; the initial finite element model is corrected using the experimental modal testing results to obtain a corrected calibration finite element model; a neural network is trained based on previously obtained environmental vibration acceleration data corresponding to the precision device and the calibration finite element model to obtain a neural network model in relation to the reference point and the target test point.

2. The vibration response prediction method of a precision apparatus according to claim 1, characterized by, the experimental modal testing based on the physical model of the precision device to obtain the experimental modal testing results comprises: an experimental modal wireframe model is constructed based on the physical model of the precision device; data acquisition is performed on the experimental modal wireframe model to obtain acceleration response data of each measuring point under different input excitations; experimental modal testing results are obtained based on all the acceleration response data and input excitation data, and the experimental modal testing results include modal frequencies, dampings and vibration modes under multiple orders of modes.

3. The vibration response prediction method of a precision apparatus according to claim 2, characterized by, The input excitation is a force hammer excitation or a shaker excitation.

4. The precision equipment vibration response prediction method according to claim 2, characterized by, the correction of the initial finite element model using the experimental modal testing results to obtain the corrected calibration finite element model comprises: finite element calculation results are calculated using the initial finite element model, and the finite element calculation results include multiple sets of modal frequencies and corresponding vibration modes; condensed finite element calculation results and experimental modal testing results are obtained by condensing the finite element calculation results and the experimental modal testing results; error analysis and sensitivity analysis are performed on the frequencies and vibration modes in the condensed finite element calculation results and the experimental modal testing results to determine the to-be-corrected parameters of the finite element model; the to-be-corrected parameters are iteratively corrected through correction iteration until the corrected parameters are obtained when the iteration converges; a corrected finite element model is obtained based on the corrected parameters; error verification of the vibration modes and the frequencies of the corrected finite element model is performed, and the corrected finite element model is taken as the calibration finite element model if the verification is passed.

5. The precision equipment vibration response prediction method according to claim 1, characterized by, the neural network training based on the previously obtained environmental vibration acceleration data corresponding to the precision device and the calibration finite element model to obtain the neural network model in relation to the reference point and the target test point comprises: vibration testing is performed on a laboratory where the precision device is located to obtain environmental vibration acceleration data; input excitation of the calibration finite element model, to obtain vibration response sample data of each reference point and each target test point respectively; constructing a training set based on the vibration response sample data of each reference point and the vibration response sample data of each target test point; training a neural network using the training set to obtain a neural network model reflecting the corresponding relationship between each reference point and the corresponding target test point.

6. A vibration response prediction device for precision equipment, characterized in that, Comprising: a test module for testing the vibration response of the reference point of the precision equipment to obtain the vibration response data of the reference point; an analysis module for analyzing the vibration response data of the reference point based on the vibration response data of the reference point and the neural network model related to the reference point and the target test point previously established by the establishing module, to obtain the vibration response data of the target test point corresponding to the reference point; wherein the reference point is located on the surface of the precision equipment, the target test point is located at a pre-set region inside the precision equipment, and the target test point is a low-temperature position that needs to be measured when using dilute refrigerant; the establishing module comprises: a first establishing unit for establishing an initial finite element model based on the 3D model of the precision equipment; a test unit for performing experimental modal testing based on the physical model of the precision equipment to obtain experimental modal testing results; a correction unit for correcting the initial finite element model using the experimental modal testing results to obtain a corrected calibration finite element model; a training unit for performing neural network training based on the previously obtained environmental vibration acceleration data corresponding to the precision equipment and the calibration finite element model to obtain a neural network model related to the reference point and the target test point.

7. A system for predicting the vibrational response of a precision device, the system comprising: Comprising: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the vibration response prediction method of the precision equipment according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the vibration response prediction method of the precision equipment according to any one of claims 1 to 5.

Images