Comprehensive monitoring method and device for multi-degree-of-freedom damping platform and medium

Abstract

This application discloses a comprehensive monitoring method, equipment, and medium for a multi-degree-of-freedom vibration reduction platform, including: acquiring load monitoring variables and state monitoring variables of the multi-degree-of-freedom vibration reduction platform; inputting the load monitoring variables into a knowledge-data dual-drive model for analysis to obtain baseline state indicators; fusing the state monitoring variables to generate actual state indicators corresponding to the baseline state indicators; performing a quantitative assessment of the difference between the baseline state indicators and the actual state indicators to obtain an assessment result of the severity of the difference; calculating and assigning weights to each individual assessment result to obtain a comprehensive diagnostic result for the multi-degree-of-freedom vibration reduction platform. This method can effectively realize the comprehensive monitoring of the multi-degree-of-freedom vibration reduction platform, avoid platform instability and structural component damage caused by abnormal conditions, and provide a guarantee for the continuous normal operation of the multi-degree-of-freedom vibration reduction platform and the vibration reduction object.

Description

Technical Field

[0001] This application relates to the field of integrated monitoring technology for multi-degree-of-freedom vibration reduction platforms, and in particular to an integrated monitoring method, equipment and medium for multi-degree-of-freedom vibration reduction platforms. Background Technology

[0002] With the rapid development of science and technology, multi-degree-of-freedom vibration damping platforms are widely used in various training simulators such as flight simulators, ship simulators, naval helicopter take-off and landing simulation platforms, tank simulators, car driving simulators, train driving simulators, earthquake simulators, as well as motion-sensing movies, entertainment equipment, and even in the docking of spacecraft and the refueling docking of aerial refueling tankers. Currently, research on multi-degree-of-freedom vibration damping platforms mainly focuses on improving the platform's vibration damping performance.

[0003] However, in practical applications, all components of a multi-degree-of-freedom vibration damping platform are in motion, and energy is mainly absorbed, transferred, and dissipated through the movement of internal components of the outriggers. It is a vulnerable structure, and there is a lack of research on comprehensive monitoring of multi-degree-of-freedom vibration damping platforms.

[0004] Therefore, it is necessary to propose a comprehensive monitoring scheme for multi-degree-of-freedom vibration reduction platforms to avoid platform instability and damage to platform structural components due to abnormal conditions, and to ensure the continuous normal operation of multi-degree-of-freedom vibration reduction platforms and vibration reduction objects. Summary of the Invention

[0005] The main purpose of this application is to provide a comprehensive monitoring method, equipment and medium for a multi-degree-of-freedom vibration reduction platform, which aims to effectively realize the comprehensive monitoring of the multi-degree-of-freedom vibration reduction platform, avoid platform instability and damage to platform structural components due to abnormal conditions, and provide a guarantee for the continuous normal operation of the multi-degree-of-freedom vibration reduction platform and the vibration reduction object.

[0006] To achieve the above objectives, this application provides a comprehensive monitoring method for a multi-degree-of-freedom vibration reduction platform, the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform comprising:

[0007] Obtain the load monitoring variables and state monitoring variables of the multi-degree-of-freedom vibration reduction platform;

[0008] The load monitoring variables are input into a pre-trained knowledge-data dual-drive model for analysis to obtain the corresponding baseline state indicators; wherein, the knowledge-data dual-drive model is trained based on a knowledge graph and a data model, and the knowledge graph serves as the prior knowledge of the data model;

[0009] By integrating the aforementioned state monitoring variables, the actual state index corresponding to the benchmark state index is generated.

[0010] Based on the baseline state index and the actual state index, a quantitative assessment of the difference is performed to obtain the assessment results of the severity of the difference of several individual state indices of the multi-degree-of-freedom vibration reduction platform.

[0011] The severity of differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform are evaluated and weighted respectively to obtain the comprehensive diagnostic results of the multi-degree-of-freedom vibration reduction platform.

[0012] Optionally, the training process of the knowledge data dual-drive model includes:

[0013] Obtain the load monitoring variable dataset and the state monitoring variable dataset;

[0014] The data in the state monitoring variable dataset are fused to obtain the actual state index of the sample;

[0015] The data model is constructed, and the knowledge graph is constructed based on the physical characteristics of the multi-degree-of-freedom vibration reduction platform and their correlations.

[0016] The knowledge graph is used as prior knowledge for the Bayesian method, and the Bayesian method is used in combination with the knowledge graph and the data model to obtain an initial dual-drive model.

[0017] The data from the loaded monitoring variable dataset is input into the initial dual-drive model for prediction to obtain the training results;

[0018] Based on the training results and the actual state indicators of the samples, the parameters of the initial dual-drive model are adjusted to obtain the parameter-adjusted initial dual-drive model.

[0019] Return to the step of inputting the data in the monitored variable dataset into the initial dual-drive model for prediction to obtain the training result; and so on, perform parameter iteration until the training result and the actual state index of the sample reach the same level, and obtain the trained knowledge data dual-drive model.

[0020] Optionally, the step of performing a quantitative assessment of the difference between the baseline state index and the actual state index to obtain the assessment results of the severity of the difference of each of the individual state indices of the multi-degree-of-freedom vibration reduction platform includes:

[0021] By comparing the baseline state index with the actual state index, residual data is obtained, and a damage hypothesis test is performed on the residual data to obtain the test results.

[0022] The residual data is confirmed to meet the preset threshold based on the test results; the residual data characterizes the damage state and / or overload state and / or the stable state of the top plate of the multi-degree-of-freedom vibration reduction platform.

[0023] If the residual data meets the threshold, then the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are normal.

[0024] If the residual data does not meet the preset threshold, the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are abnormal.

[0025] Optionally, after the step of determining that the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are abnormal if the residual data does not meet the preset threshold, the method further includes:

[0026] The severity of the residual data was classified using hypothesis testing, and the classification results were obtained.

[0027] A graded alarm will be triggered based on the grading results.

[0028] Optionally, the step of calculating and assigning weights to the severity assessment results of the differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform to obtain the comprehensive diagnostic result of the multi-degree-of-freedom vibration reduction platform includes:

[0029] The evaluation results are analyzed using a pre-defined analytic hierarchy process (AHP) to obtain subjective weights.

[0030] Obtain the reciprocal of the L2 norm or the reciprocal of the sum of the absolute values ​​of the residual data, and the state anomaly probability of the error data, as an objective weight;

[0031] Based on the subjective weights and the objective weights, a weight optimization function with weight parameters as variables is established;

[0032] The comprehensive diagnostic result is obtained by combining the severity assessment result of the difference with the weight optimization function.

[0033] Optionally, the step of combining the severity assessment result of the difference with the weight optimization function to obtain the comprehensive diagnostic result includes:

[0034] The comprehensive diagnostic result is obtained by combining the severity assessment result of the difference with the weight optimization function using any one of the Bayesian weighted fusion method, DS evidence theory, or linear weighted average method.

[0035] Optionally, after the step of calculating and assigning weights to the assessment results of the severity of differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform to obtain the comprehensive diagnostic result of the multi-degree-of-freedom vibration reduction platform, the method further includes:

[0036] The comprehensive diagnostic results are fed back into the knowledge graph to expand the knowledge graph; wherein the knowledge graph contains mechanistic information, state features and diagnostic results.

[0037] Optionally, the multi-degree-of-freedom vibration reduction platform includes at least two of the following individual state indicators: support, plate, outrigger, and center position overload-proof spring damping support; the plate includes an upper plate and a lower plate.

[0038] The load monitoring variables include the stress at the joint position of the multi-degree-of-freedom vibration reduction platform, the acceleration-tilt angle of the upper plate and the acceleration-tilt angle of the lower plate, the acceleration and tilt angle of the lower plate, the internal force between the plate and the outrigger, and the overload pressure of the lower plate.

[0039] The status monitoring variables include the relative displacement of the upper plate and the lower plate, the acceleration and tilt angle of the upper plate, the relative velocity between the telescopic part and the fixed part of the outrigger, and the relative displacement between the telescopic part and the fixed part of the outrigger.

[0040] A smart skin sensor is provided in the area of ​​the joint between the support and the plate, and the smart skin sensor is used to obtain the stress at the joint.

[0041] The plate is equipped with an integrated acceleration-tilt sensor and a relative displacement sensor. The integrated acceleration-tilt sensor is used to obtain the acceleration-tilt angle of the upper plate and the acceleration-tilt angle of the lower plate, and the relative displacement sensor is used to obtain the relative displacement of the upper plate and the lower plate.

[0042] The outrigger is equipped with an outrigger speed sensor, an outrigger displacement sensor, and an outrigger force sensor; the outrigger speed sensor is used to obtain the relative speed between the telescopic part and the fixed part of the outrigger; the outrigger displacement sensor is used to obtain the relative displacement between the telescopic part and the fixed part of the outrigger; the outrigger force sensor is used to obtain the internal force between the plate and the outrigger.

[0043] The overload-resistant spring damping support at the center position is equipped with a pressure sensor; the pressure sensor is used to obtain the overload pressure of the lower plate.

[0044] This application also proposes a terminal device, which includes a memory, a processor, and a comprehensive monitoring program for a multi-degree-of-freedom vibration reduction platform stored in the memory and executable on the processor. When the comprehensive monitoring program for the multi-degree-of-freedom vibration reduction platform is executed by the processor, it implements the steps of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform as described above.

[0045] This application also proposes a computer-readable storage medium storing a comprehensive monitoring program for a multi-degree-of-freedom vibration damping platform. When executed by a processor, the comprehensive monitoring program for the multi-degree-of-freedom vibration damping platform implements the steps of the comprehensive monitoring method for the multi-degree-of-freedom vibration damping platform as described above.

[0046] The comprehensive monitoring method, device, terminal equipment, and storage medium for a multi-degree-of-freedom vibration reduction platform proposed in this application acquire load monitoring variables and state monitoring variables of several individual state indicators of the multi-degree-of-freedom vibration reduction platform; input the load monitoring variables into a knowledge-data dual-driven model for analysis to obtain baseline state indicators; fuse the state monitoring variables to generate actual state indicators corresponding to the baseline state indicators; perform a quantitative assessment of the differences between the baseline state indicators and the actual state indicators to obtain an assessment result of the severity of the differences; calculate and assign weights to each assessment result to obtain a comprehensive diagnostic result for the multi-degree-of-freedom vibration reduction platform. This can effectively realize the comprehensive monitoring of the multi-degree-of-freedom vibration reduction platform, avoid platform instability and structural component damage caused by abnormal conditions, and provide a guarantee for the continuous normal operation of the multi-degree-of-freedom vibration reduction platform and the vibration reduction object. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating a first exemplary embodiment of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application;

[0048] Figure 2 This is a schematic diagram of a comprehensive diagnosis based on the fusion technology of knowledge-data dual-drive and optimized weights;

[0049] Figure 3 This is an architecture diagram of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application;

[0050] Figure 4 This is a flowchart illustrating a third exemplary embodiment of the integrated monitoring method for the multi-degree-of-freedom vibration reduction platform of this application;

[0051] Figure 5 This is a flowchart illustrating the fourth exemplary embodiment of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application.

[0052] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0053] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application.

[0054] The main solution of this application embodiment is as follows: Obtain the load monitoring variables and state monitoring variables of the multi-degree-of-freedom vibration reduction platform; input the load monitoring variables into a knowledge-data dual-drive model for analysis to obtain baseline state indicators; fuse the state monitoring variables to generate actual state indicators corresponding to the baseline state indicators; perform a quantitative assessment of the differences between the baseline state indicators and the actual state indicators to obtain an assessment result of the severity of the differences; calculate and assign weights to each assessment result to obtain a comprehensive diagnostic result for the multi-degree-of-freedom vibration reduction platform. This effectively achieves comprehensive monitoring of the multi-degree-of-freedom vibration reduction platform, avoids platform instability and structural component damage caused by abnormal states, and provides a guarantee for the continuous normal operation of the multi-degree-of-freedom vibration reduction platform and the vibration reduction object.

[0055] Specifically, refer to Figure 1 , Figure 1 This is a flowchart illustrating a first exemplary embodiment of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application. The comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform includes:

[0056] Step S110: Obtain the load monitoring variables and state monitoring variables corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform;

[0057] Specifically, the load monitoring variables may include the measured values ​​of physical loads such as force and pressure on each component on the platform. For example, several individual state indicators of the multi-degree-of-freedom vibration damping platform include at least two of the following: support, plate, outrigger, and overload-resistant spring-damped support at the center position. The plate includes the upper plate and the lower plate. The load monitoring variables include the stress at the joint position of the multi-degree-of-freedom vibration damping platform, the acceleration-tilt angle of the upper plate and the acceleration-tilt angle of the lower plate, the acceleration and tilt angle of the lower plate, the internal force between the plate and the outrigger, and the overload pressure of the lower plate.

[0058] The condition monitoring variables include the relative displacement of the upper and lower plates, the acceleration and tilt angle of the upper plate, the relative velocity between the telescopic and fixed parts of the outriggers, and the relative displacement between the telescopic and fixed parts of the outriggers.

[0059] Step S120: Input the load monitoring variables into the pre-trained knowledge data dual-drive model for analysis to obtain the corresponding baseline state index; wherein, the knowledge data dual-drive model is trained based on a knowledge graph and a data model, and the knowledge graph serves as the prior knowledge of the data model;

[0060] Specifically, refer to Figure 2 , Figure 2This diagram illustrates a comprehensive diagnostic approach based on a knowledge-data dual-drive and optimized weight fusion technology. The knowledge-data dual-drive model combines a knowledge graph and a data model; the knowledge graph provides prior knowledge, while the data model is trained using actual measurement data. This can be viewed as a method of integrating theoretical knowledge and practical data. By analyzing the knowledge-data dual-drive model, baseline state indicators corresponding to the input load monitoring variables are obtained. These baseline state indicators can be used to assess the normal operating status of the multi-degree-of-freedom vibration reduction platform.

[0061] Step S130: Integrate the state monitoring variables to generate the actual state index corresponding to the benchmark state index;

[0062] Specifically, by integrating status monitoring variables with baseline status indicators, actual status indicators are obtained, which can be used to reflect the current actual status of the platform, and further take into account various changes and impacts that may exist in the actual working environment.

[0063] Step S140: Based on the benchmark state index and the actual state index, perform a quantitative evaluation of the difference to obtain the evaluation results of the severity of the difference of several individual state indices of the multi-degree-of-freedom vibration reduction platform.

[0064] Specifically, by conducting a quantitative assessment based on the differences between baseline and actual state indicators, the performance deviations or problems of each component of the platform can be identified, thereby obtaining an assessment result of the severity of the differences among the components of the multi-degree-of-freedom vibration damping platform. For example, the assessment result of the severity of the differences can include information such as the health status and performance problems of each component.

[0065] Difference analysis can cover aspects such as damage, overload, and excessive motion of various components in a multi-degree-of-freedom vibration damping platform, thereby providing a more comprehensive assessment of the platform's health status. For example, the smaller the difference, the more normal the platform's condition.

[0066] Step S150 involves calculating and assigning weights to the severity assessment results of the differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform, thereby obtaining the comprehensive diagnostic results of the multi-degree-of-freedom vibration reduction platform.

[0067] Specifically, the severity assessment results of the differences among the components are weighted and assigned based on the importance and impact of each component within the overall system. This results in a comprehensive diagnostic result for the multi-degree-of-freedom vibration reduction platform, taking into account both the severity of the differences among the components and their relative importance within the system.

[0068] This embodiment, through the above-described scheme, specifically acquires the load monitoring variables and state monitoring variables of several individual state indicators of the multi-degree-of-freedom vibration reduction platform; inputs the load monitoring variables into a knowledge-data dual-driven model for analysis to obtain the baseline state indicators; integrates the state monitoring variables to generate the actual state indicators corresponding to the baseline state indicators; performs a quantitative assessment of the differences between the baseline state indicators and the actual state indicators to obtain the assessment results of the severity of the differences; calculates and assigns weights to each assessment result to obtain the comprehensive diagnostic results of the multi-degree-of-freedom vibration reduction platform. This can effectively realize the comprehensive monitoring of the multi-degree-of-freedom vibration reduction platform, avoid platform instability and structural component damage caused by abnormal conditions, and provide a guarantee for the continuous normal operation of the multi-degree-of-freedom vibration reduction platform and the vibration reduction object.

[0069] Based on the first embodiment described above, a second embodiment of this application is proposed. (Refer to...) Figure 3 , Figure 3 This is an architecture diagram of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application. The multi-degree-of-freedom vibration reduction platform includes at least two of the following individual state indicators: support, plate, legs, and center position overload-resistant spring-damped support; the plate includes an upper plate and a lower plate.

[0070] The multi-degree-of-freedom vibration reduction platform includes at least two of the following components: a support, a plate, outriggers, and a central overload-resistant spring-damped support. The plate includes an upper plate and a lower plate.

[0071] The load monitoring variables include the surface stress at the joint position of the multi-degree-of-freedom vibration damping platform, the acceleration and tilt angle of the lower plate, the internal force between the plate and the outrigger, and the overload pressure of the upper plate.

[0072] The status monitoring variables include the relative displacement of the upper plate and the lower plate, the acceleration and tilt angle of the upper plate, the relative velocity between the telescopic part and the fixed part of the outrigger, and the relative displacement between the telescopic part and the fixed part of the outrigger.

[0073] A smart skin sensor is provided in the vicinity of the joint position between the support and the plate, and the smart skin sensor is used to obtain the stress at the joint position.

[0074] The plate is equipped with an integrated acceleration-tilt sensor and a relative displacement sensor. The integrated acceleration-tilt sensor is used to obtain the acceleration-tilt angle of the upper plate and the acceleration-tilt angle of the lower plate, and the relative displacement sensor is used to obtain the relative displacement of the upper plate and the lower plate.

[0075] The outrigger is equipped with an outrigger speed sensor, an outrigger displacement sensor, and an outrigger force sensor; the outrigger speed sensor is used to obtain the relative speed between the telescopic part and the fixed part of the outrigger; the outrigger displacement sensor is used to obtain the relative displacement between the telescopic part and the fixed part of the outrigger; the outrigger force sensor is used to obtain the internal force between the lower plate and the outrigger.

[0076] The overload-resistant spring damping support at the center position is equipped with a pressure sensor; the pressure sensor is used to obtain the overload pressure of the upper plate.

[0077] In addition, the intelligent skin sensor, accelerator-tilt integrated sensor, relative displacement sensor, velocity sensor, displacement sensor, force sensor, and pressure sensor in the multi-degree-of-freedom mechanical vibration reduction platform are all wirelessly transmitted to the monitoring center with diagnostic and early warning functions via 4G / 5G. This allows the data receiving module and server of the monitoring center to execute steps S110 to S150 based on the knowledge graph library, diagnostic analysis module, and early warning and information release module. The data acquired by each sensor and the final comprehensive diagnostic results are then sent to the human-machine interface so that users can operate based on this data.

[0078] Furthermore, referring to Figure 4 , Figure 4 This is a flowchart illustrating a third exemplary embodiment of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application. Based on the first and second embodiments described above, step S140 involves performing a quantitative assessment of the difference between the baseline state index and the actual state index to obtain assessment results of the severity of differences for several individual state indices of the multi-degree-of-freedom vibration reduction platform, including:

[0079] Step S410: Compare the baseline state index with the actual state index to obtain residual data, and perform a damage hypothesis test on the residual data to obtain the test results;

[0080] Specifically, residual data is the difference between baseline state indicators and actual state indicators, reflecting the deviation between the actual operating state and the expected state of the system.

[0081] Step S420: Confirm whether the residual data meets the preset threshold based on the test results; the residual data characterizes the damage state and / or overload state and / or the stable state of the upper plate of the multi-degree-of-freedom vibration reduction platform.

[0082] Specifically, damage hypothesis testing is a statistical method used to verify whether residual data conforms to the expected damage hypothesis, i.e., whether the system is damaged or overloaded. Damage hypothesis testing may involve using statistical tests (such as t-tests, analysis of variance, or Bayesian hypothesis testing) to determine whether there are significant differences in the residual data associated with damage or overload.

[0083] A preset threshold is a pre-determined limit value used to define the normal or abnormal state of residual data. The threshold setting can be based on actual conditions; it can be done by selecting an existing threshold or by calibrating it according to specific circumstances, to ensure sensitivity and accuracy in detecting abnormal states.

[0084] Residual data is not only used to determine whether there is an anomaly, but also to provide information about the damage status, overload status and stability of the top plate, such as the health of the outriggers, platform stability, platform load-bearing capacity and other aspects.

[0085] Step S431: If the residual data meets the threshold, the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are normal.

[0086] Step S432: If the residual data does not meet the preset threshold, the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are abnormal.

[0087] Specifically, based on whether the residual data meets the threshold, the components of the multi-degree-of-freedom vibration damping platform are evaluated to determine whether their status is normal or abnormal. A normal status indicates that the system is within the expected operating range, while an abnormal status may indicate problems requiring further inspection and maintenance. The feedback of the evaluation results (normal or abnormal) will guide subsequent maintenance and repair work. This feedback mechanism helps to monitor and manage the status of the multi-degree-of-freedom vibration damping platform in real time, improving the system's reliability and safety.

[0088] Further, after step S432, which states that if the residual data does not meet a preset threshold, the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are abnormal, the method further includes:

[0089] Step A10: Using hypothesis testing, the severity of the residual data is classified to obtain the classification results;

[0090] Specifically, in hypothesis testing, the analysis of residual data involves not only determining whether anomalies exist, but also quantifying the degree of anomalies to gain a more comprehensive understanding of the system's health. Severity grading can be achieved using various methods, such as classifying residual data into mild, moderate, and severe levels, or grading based on the magnitude of statistical measures.

[0091] The grading result is an evaluation obtained by grading the residual data, reflecting the degree of anomaly of each component or system in hypothesis testing. The grading result can be presented in an intuitive way, such as through color coding or numbers, so that users can quickly understand the health status of each component.

[0092] For example, the specific calculation formula can be referred to as follows, where Br is the grading result:

[0093]

[0094] Where H p1 As a lossy assumption, H p0 Assuming no loss, E represents the residual data. If the difference between the forecast baseline and current observations is within the normal range, then B... r <1, if B r A value greater than 1 indicates that the current data supports the lossy hypothesis, and its magnitude reflects the degree of damage. Bayesian factors with different value ranges can be used for the quantitative grading of damage severity, i.e., 1 < B. r <3 indicates the damage is negligible; 3 < B r <10 indicates substantial damage, l0 < B r <30 indicates a very high degree of damage, 30 < B r A value <100 indicates a decisive injury.

[0095] Step A20: Perform a graded alarm based on the graded results.

[0096] Specifically, tiered alarms are an alert mechanism based on a tiered classification system, designed to promptly notify relevant personnel or system operators of the severity of any anomalies. Tiered alarms can employ different alarm levels, such as low, medium, and high, to allow for varying emergency measures based on the severity. This helps improve sensitivity to changes in system status and mitigate potential risks.

[0097] Further, in step S150, weights are calculated and assigned to the severity assessment results of the differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform to obtain the comprehensive diagnostic results of the multi-degree-of-freedom vibration reduction platform, including:

[0098] Step S440: The evaluation results are analyzed using a preset analytic hierarchy process to obtain subjective weights;

[0099] Step S450: Obtain the reciprocal of the L2 norm or the reciprocal of the sum of absolute values ​​of the residual data, and the state anomaly probability of the error data, as an objective weight;

[0100] Step S460: Based on the subjective weights and the objective weights, establish a weight optimization function with weight parameters as variables;

[0101] Step S470: Combine the severity assessment result of the difference with the weight optimization function to obtain the comprehensive diagnostic result.

[0102] Specifically, the analytic hierarchy process (AHP) can be used to determine the subjective weights of the model output, and the weights based on the inverse of the model error and the probability of state anomalies can be used as objective weights. A weight optimization function w = f(w_z, w_k_e, w_k_p) can be established with weight parameters as variables, outputting weights from 1 to N. Here, underscores represent subscripts, expressed in formula form. For example, "_z" is the subscript z, representing the subjective weight; "_k_e" is the subscript ke, representing the objective weight based on the inverse of the model error; and "_k_p" represents the objective weight based on the probability of state anomalies.

[0103] By combining subjective and objective weights appropriately, and designing a weight optimization function, a comprehensive weight allocation can be achieved for the evaluation results of each component of a multi-degree-of-freedom vibration reduction platform. This method helps to more comprehensively consider the importance of different factors and adjust the weights based on the results of the difference analysis, making the comprehensive diagnostic results more accurate and reliable.

[0104] Further, in step S470, the severity assessment result of the difference is combined with the weight optimization function to obtain the comprehensive diagnostic result, including:

[0105] Step S471: Using any one of the Bayesian weighted fusion method, DS evidence theory, or linear weighted average method, the severity assessment result of the difference is combined with the weight optimization function to obtain the comprehensive diagnostic result.

[0106] Specifically, the Bayesian weighted fusion method, based on Bayes' theorem, calculates the posterior probabilities of different factors by considering prior information and observed data. Then, it combines these posterior probabilities using weights to obtain the final comprehensive probability. This effectively combines information from different factors, making it suitable for multi-source information fusion while also considering uncertainty.

[0107] Dempster-Shafer Evidence Theory (DS) is based on the theory of uncertainty. It represents evidence as basic probability assignment functions, and by merging these functions, it obtains overall evidence for decision-making. DS is particularly suitable for handling evidence from different sources, especially in cases of conflict or inconsistency, providing a more flexible approach to information fusion.

[0108] The linear weighted average method is simple and intuitive. It involves linearly weighting and summing the results of the severity assessment of differences. The weights are calculated using the previous analytic hierarchy process and objective weights, making it easy to understand and implement.

[0109] By combining the severity assessment results of the differences with a weighted optimization function, the assessment results of each component in the multi-degree-of-freedom vibration reduction platform can be weighted and integrated to obtain a comprehensive diagnostic result, which reflects the different levels of attention to the severity of differences under different situations.

[0110] This embodiment, through the above-described scheme, specifically by employing hypothesis testing and severity grading, allows system operators or maintenance personnel to more comprehensively understand the abnormal state grading alarm mechanism of each component of the multi-degree-of-freedom vibration damping platform. This provides crucial information for timely action, ensuring rapid maintenance and repair when problems occur. It can compensate for the lack of comprehensive state monitoring and early warning functions in traditional six-degree-of-freedom platforms, automatically issuing warnings when the platform experiences abnormal states, preventing platform instability and structural component damage due to abnormal states, and ensuring the continuous normal operation of the multi-degree-of-freedom platform and its vibration damping targets.

[0111] Reference Figure 5 , Figure 5 This is a flowchart illustrating a fourth exemplary embodiment of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform of this application. Based on the above... Figure 2 The training process of the knowledge data dual-drive model, as shown in the embodiment, includes:

[0112] Step S510: Obtain the load monitoring variable dataset and the state monitoring variable dataset;

[0113] Step S520: The data in the state monitoring variable dataset are fused to obtain the actual state index of the sample;

[0114] Step S530: Construct the data model and build the knowledge graph based on the physical characteristics of the multi-degree-of-freedom vibration reduction platform and their correlations;

[0115] Step S540: Use the knowledge graph as prior knowledge for the Bayesian method, and use the Bayesian method to combine the knowledge graph and the data model to obtain an initial dual-drive model;

[0116] Step S550: Input the data from the load monitoring variable dataset into the initial dual-drive model for prediction to obtain the training results;

[0117] Step S560: Combine the training results and the actual state indicators of the samples to adjust the parameters of the initial dual-drive model, obtain the parameter-adjusted initial dual-drive model, and return to step S550: input the data in the load monitoring variable dataset into the initial dual-drive model for prediction to obtain the training results.

[0118] Step S570, and so on, performs parameter iteration until the training result and the actual state index of the sample are consistent, thus obtaining the trained knowledge data dual-drive model.

[0119] Specifically, the load monitoring variable dataset and the state monitoring variable dataset can include time series data of key variables such as pressure, displacement, and velocity, which are used to reflect the operating status of the system.

[0120] Data from different state monitoring variables are fused to obtain more comprehensive indicators that reflect the actual state of the system. This fusion process may include steps such as data normalization and matrix transformation with dimensionality reduction capabilities to ensure that the contribution weights of each local response to the overall health indicators are reasonable, and that the obtained state indicators accurately reflect the overall state of the system.

[0121] A knowledge graph consists of structural attributes, structural features, and state features related to the platform's state, as well as the relationships between them. This information includes the platform's physical characteristics and key parameters of its operational state.

[0122] A single-item data model describes a data model of a single type of state change, while a dual-drive model is a combination of a knowledge graph and a data model. By constraining the data model with a knowledge graph, it makes it more accurately reflect the actual state.

[0123] The construction of data models can utilize machine learning methods, such as models based on long short-term networks, to capture patterns and trends in the loaded data. Knowledge graphs provide prior knowledge, such as relationships between physical features and structural attributes, which helps improve the interpretability and accuracy of data models.

[0124] The knowledge graph can include the physical characteristics of the multi-degree-of-freedom vibration reduction platform, the relationships between structural properties, and features related to healthy and damaged states, effectively enabling a deep understanding of the state of the multi-degree-of-freedom vibration reduction platform.

[0125] Using a dataset with loaded monitoring variables, predictions are made through an initial dual-drive model to obtain preliminary training results. These results can then be compared with actual state indicators to evaluate the model's accuracy.

[0126] By combining training results with actual health indicators of the samples, the parameters of the initial dual-drive model are adjusted through repeated iterative processes of input, analysis, and adjustment. In this way, by continuously adjusting the model parameters iteratively, the model's predictive ability gradually approaches the actual situation, improving its accuracy. Through multiple iterations and continuous adjustment of model parameters, until the training results and actual state indicators are sufficiently consistent, it can be ensured that the model can accurately reflect the system's operating state.

[0127] This embodiment, through the above-described scheme, specifically constructs a data model and knowledge graph by fusing state monitoring variable data and combining it with the system's physical characteristics and structural attributes. It utilizes a Bayesian method to combine prior knowledge provided by the knowledge graph with the data model, providing an initial dual-drive model. The initial model is used for prediction to obtain training results. Combined with the training results and actual state indicators, parameters are adjusted to improve the model's fitting ability. Through repeated iterations, the model gradually becomes more consistent with the actual state, achieving a well-trained knowledge-data dual-drive model. Thus, the ability to accurately predict the state of a multi-degree-of-freedom vibration reduction platform can be continuously improved during the learning process.

[0128] Furthermore, this application also proposes a terminal device, which includes a memory, a processor, and a comprehensive monitoring program for a multi-degree-of-freedom vibration damping platform stored in the memory and executable on the processor. When the comprehensive monitoring program for the multi-degree-of-freedom vibration damping platform is executed by the processor, it implements the steps of the comprehensive monitoring method for the multi-degree-of-freedom vibration damping platform as described above.

[0129] Since the comprehensive monitoring program of this multi-degree-of-freedom vibration reduction platform adopts all the technical solutions of all the aforementioned embodiments when it is executed by the processor, it has at least all the beneficial effects brought about by all the technical solutions of all the aforementioned embodiments, which will not be repeated here.

[0130] Furthermore, this application also proposes a computer-readable storage medium storing a comprehensive monitoring program for a multi-degree-of-freedom vibration damping platform. When the comprehensive monitoring program for the multi-degree-of-freedom vibration damping platform is executed by a processor, it implements the steps of the comprehensive monitoring method for the multi-degree-of-freedom vibration damping platform as described above.

[0131] Since the comprehensive monitoring program of this multi-degree-of-freedom vibration reduction platform adopts all the technical solutions of all the aforementioned embodiments when it is executed by the processor, it has at least all the beneficial effects brought about by all the technical solutions of all the aforementioned embodiments, which will not be repeated here.

[0132] Compared to existing technologies, the comprehensive monitoring method, device, terminal equipment, and storage medium for multi-degree-of-freedom vibration reduction platforms proposed in this application acquire the load monitoring variables and state monitoring variables of the multi-degree-of-freedom vibration reduction platform; input the load monitoring variables into a knowledge-data dual-drive model for analysis to obtain baseline state indicators; fuse the state monitoring variables to generate actual state indicators corresponding to the baseline state indicators; perform a quantitative assessment of the differences between the baseline state indicators and the actual state indicators to obtain an assessment result of the severity of the differences; calculate and assign weights to each assessment result to obtain a comprehensive diagnostic result for the multi-degree-of-freedom vibration reduction platform. This can effectively achieve comprehensive monitoring of the multi-degree-of-freedom vibration reduction platform, avoid platform instability and structural component damage caused by abnormal states, and provide a guarantee for the continuous normal operation of the multi-degree-of-freedom vibration reduction platform and the vibration reduction object.

[0133] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system 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 system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0134] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0135] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, controlled terminal, or network device, etc.) to execute the methods of each embodiment of this application.

[0136] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A comprehensive monitoring method of a multi-degree-of-freedom vibration reduction platform, characterized in that, The comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform includes the following steps: Obtain the load monitoring variables and state monitoring variables of the multi-degree-of-freedom vibration reduction platform; The load monitoring variables are input into a pre-trained knowledge-data dual-drive model for analysis to obtain the corresponding baseline state indicators; wherein, the knowledge-data dual-drive model is trained based on a knowledge graph and a data model, and the knowledge graph serves as the prior knowledge of the data model; By integrating the aforementioned state monitoring variables, the actual state index corresponding to the benchmark state index is generated. Based on the baseline state index and the actual state index, a difference quantification assessment is performed to obtain the severity assessment results of the differences in several individual state indexes of the multi-degree-of-freedom vibration damping platform. The specific steps include: comparing the baseline state index and the actual state index to obtain residual data, and performing a damage hypothesis test on the residual data to obtain the test results; confirming whether the residual data meets a preset threshold based on the test results; the residual data characterizes the damage state and / or overload state and / or the stable state of the upper plate of the multi-degree-of-freedom vibration damping platform; if the residual data meets the threshold, the assessment results corresponding to the several individual state indexes of the multi-degree-of-freedom vibration damping platform are normal; if the residual data does not meet the threshold, the assessment results corresponding to the several individual state indexes of the multi-degree-of-freedom vibration damping platform are abnormal. The steps include: calculating and assigning weights to the severity assessment results of the differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform, and obtaining the comprehensive diagnostic result of the multi-degree-of-freedom vibration reduction platform; specifically, the steps include: The evaluation results are analyzed using a pre-defined analytic hierarchy process (AHP) to obtain subjective weights. Obtain the reciprocal of the L2 norm or the reciprocal of the sum of the absolute values ​​of the residual data, and use it as an objective weight along with the state anomaly probability of the error data. Based on the subjective weights and the objective weights, a weight optimization function with weight parameters as variables is established; The comprehensive diagnostic result is obtained by combining the severity assessment result of the difference with the weight optimization function.

2. The comprehensive monitoring method for a multi-degree-of-freedom vibration reduction platform as described in claim 1, characterized in that, The training process of the knowledge data dual-drive model includes: Obtain the load monitoring variable dataset and the state monitoring variable dataset; The data in the state monitoring variable dataset are fused to obtain the actual state index of the sample; The data model is constructed, and the knowledge graph is constructed based on the physical characteristics of the multi-degree-of-freedom vibration reduction platform and their correlations. The knowledge graph is used as prior knowledge for the Bayesian method, and the Bayesian method is used in combination with the knowledge graph and the data model to obtain an initial dual-drive model. The data from the loaded monitoring variable dataset is input into the initial dual-drive model for prediction to obtain the training results; Based on the training results and the actual state indicators of the samples, the parameters of the initial dual-drive model are adjusted to obtain the parameter-adjusted initial dual-drive model. Return to the step of inputting the data in the monitored variable dataset into the initial dual-drive model for prediction to obtain the training result; and so on, perform parameter iteration until the training result and the actual state index of the sample reach the same level, and obtain the trained knowledge data dual-drive model.

3. The comprehensive monitoring method for a multi-degree-of-freedom vibration reduction platform as described in claim 1, characterized in that, Following the step of determining that if the residual data does not meet the threshold, the evaluation results corresponding to several individual state indicators of the multi-degree-of-freedom vibration reduction platform are abnormal, the method further includes: The severity of the residual data was classified using hypothesis testing, and the classification results were obtained. A graded alarm will be triggered based on the grading results.

4. The comprehensive monitoring method for a multi-degree-of-freedom vibration reduction platform as described in claim 1, characterized in that, The step of combining the severity assessment result of the difference with the weight optimization function to obtain the comprehensive diagnostic result includes: By employing any one of the Bayesian weighted fusion method, DS evidence theory, or linear weighted average method, the severity assessment results of the differences are combined with the weight optimization function to obtain several individual state indicators of the comprehensive diagnostic results.

5. The comprehensive monitoring method for a multi-degree-of-freedom vibration reduction platform as described in claim 1, characterized in that, After the steps of calculating and assigning weights to the assessment results of the severity of differences in several individual state indicators of the multi-degree-of-freedom vibration reduction platform to obtain the comprehensive diagnostic result of the multi-degree-of-freedom vibration reduction platform, the method further includes: The comprehensive diagnostic results are fed back into the knowledge graph to expand the knowledge graph; wherein the knowledge graph contains mechanistic information, state features and diagnostic results.

6. The comprehensive monitoring method for a multi-degree-of-freedom vibration reduction platform as described in claim 1, characterized in that, The multi-degree-of-freedom vibration reduction platform includes at least two of the following components: a support, a plate, outriggers, and a central overload-resistant spring-damped support. The plate includes an upper plate and a lower plate. The load monitoring variables include the surface stress at the joint position of the multi-degree-of-freedom vibration damping platform, the acceleration and tilt angle of the lower plate, the internal force between the plate and the outrigger, and the overload pressure of the upper plate. The status monitoring variables include the relative displacement of the upper plate and the lower plate, the acceleration and tilt angle of the upper plate, the relative velocity between the telescopic part and the fixed part of the outrigger, and the relative displacement between the telescopic part and the fixed part of the outrigger. A smart skin sensor is provided in the vicinity of the joint position between the support and the plate, and the smart skin sensor is used to obtain the stress at the joint position. The plate is equipped with an integrated acceleration-tilt sensor and a relative displacement sensor. The integrated acceleration-tilt sensor is used to obtain the acceleration-tilt angle of the upper plate and the acceleration-tilt angle of the lower plate, and the relative displacement sensor is used to obtain the relative displacement of the upper plate and the lower plate. The outrigger is equipped with an outrigger speed sensor, an outrigger displacement sensor, and an outrigger force sensor; the outrigger speed sensor is used to obtain the relative speed between the telescopic part and the fixed part of the outrigger; the outrigger displacement sensor is used to obtain the relative displacement between the telescopic part and the fixed part of the outrigger; the outrigger force sensor is used to obtain the internal force between the lower plate and the outrigger. The overload-resistant spring damping support at the center position is equipped with a pressure sensor; the pressure sensor is used to obtain the overload pressure of the upper plate.

7. A terminal device, characterized in that, The terminal device includes a memory, a processor, and a comprehensive monitoring program for a multi-degree-of-freedom vibration reduction platform stored in the memory and executable on the processor. When the comprehensive monitoring program for the multi-degree-of-freedom vibration reduction platform is executed by the processor, it implements the steps of the comprehensive monitoring method for the multi-degree-of-freedom vibration reduction platform as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a comprehensive monitoring program for a multi-degree-of-freedom vibration damping platform. When the comprehensive monitoring program for the multi-degree-of-freedom vibration damping platform is executed by a processor, it implements the steps of the comprehensive monitoring method for the multi-degree-of-freedom vibration damping platform as described in any one of claims 1-6.

Images