Efficient analysis method for the security of a BIM model assembly

Abstract

This application relates to the field of component safety analysis technology, specifically to an efficient method for analyzing the safety of components in a BIM model. The method includes: acquiring the vibration frequency sequence, temperature sequence, and tension sequence of each component at each acquisition time; clustering the vibration frequency sequences of each component at the current acquisition time to obtain anomalous clusters; obtaining the coupling field strength risk level of each component at the current acquisition time based on the randomness of the vibration frequency sequences, the local density of all anomalous clusters, and the degree of joint interaction between vibration frequency, temperature, and tension at all acquisition times, thereby obtaining the resource weight of each component at the current acquisition time; classifying the components at the current acquisition time and selecting corresponding safety analysis models for them. This application improves the efficiency of component safety analysis by conducting risk assessments on each component and selecting appropriate safety analysis models for them.

Description

Technical Field

[0001] This application relates to the field of component safety analysis technology, specifically to an efficient method for analyzing the safety of components in a BIM model. Background Technology

[0002] In traditional construction engineering, the safety analysis of components relies heavily on two-dimensional drawings, which is inefficient and prone to errors. Complex structural collisions and stress states are difficult to verify intuitively, and safety hazards are often revealed only after the fact. With the rapid development of BIM technology and computer capabilities, the safety analysis of components has undergone a revolution. Early BIM primarily achieved 3D visualization and foundation collision detection. Subsequently, parametric modeling and Industry Foundation Classes (IFC) standards deepened the integration of component attributes, enabling the early detection of design conflicts and structural risks, significantly reducing on-site changes and accidents, and ensuring the safety of life and property. It also promotes collaborative decision-making and refined safety management across the entire chain of design, construction, and operation, laying the foundation for intelligent review and automated compliance checks.

[0003] In the safety analysis of components using BIM models, the computational complexity increases exponentially due to the need for real-time updates to model geometry and boundary conditions in response to changes in construction status, while the safety analysis of components relies on high-precision physical simulation. Current technologies, while employing high-precision safety analysis models to analyze components ensures accuracy, the complex calculation process impacts efficiency, leading to delays in results. Conversely, using simplified models ignores the implicit correlations between high-risk components and their effects on different environmental fields, resulting in insufficient accuracy in risk assessment and consequently, low accuracy in early warnings for high-risk components. Therefore, a safety analysis method that balances accuracy and speed is urgently needed. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides an efficient analysis method for the safety of components in a BIM model, thereby resolving existing issues.

[0005] The efficient analysis method for the safety of components in a BIM model proposed in this application adopts the following technical solution: One embodiment of this application provides an efficient method for analyzing the safety of components in a BIM model, the method comprising the following steps: Obtain the vibration frequency sequence, temperature sequence, and tension sequence of each component at each acquisition time; The data of each component in the vibration frequency sequence at the current acquisition time are clustered. The local density of each cluster is obtained based on the total number of data in each cluster and its corresponding area. This density is then compared with a preset density threshold to obtain the abnormal clusters of each component at the current acquisition time. Based on the randomness of the vibration frequency sequence of each component at the current acquisition time and the average level of the local density of all abnormal clusters of each component, the latent risk index of each component at the current acquisition time is obtained. Based on the interdependence between the vibration frequency sequence and temperature sequence of each component at the current acquisition time and all previous acquisition times, as well as the correlation between the vibration frequency sequence and tension sequence, the coupling strength factor of each component at the current acquisition time is obtained. Combined with the implicit risk index of each component at the current acquisition time, the coupling field strength risk degree of each component at the current acquisition time is obtained. By combining the sum of the coupling field strength risk degrees of all components at the current acquisition time, the resource weight of each component at the current acquisition time is obtained, and then the components at the current acquisition time are classified. The corresponding safety analysis model is selected according to the category of each component at the current acquisition time.

[0006] Preferably, the process of obtaining the local density of each cluster is as follows: obtain the cluster area of ​​each cluster, and take the ratio of the total number of data points in each cluster to the cluster area as the local density of each cluster.

[0007] Preferably, the abnormal clusters of each component at the current acquisition time refer to the clusters among all clusters of each component at the current acquisition time whose local density is greater than a preset density threshold.

[0008] Preferably, the process for obtaining the latent risk index of each component at the current data collection time is as follows: Based on the randomness of the vibration frequency sequence of each component at the current acquisition time, the marginal spectral entropy value of each component at the current acquisition time is obtained. Calculate the latent risk index of each component at the current data acquisition moment: In the formula, Let be the implicit risk index of the i-th component at the current data collection moment. Let be the marginal spectral entropy value of the i-th component at the current acquisition time. Let be the local density of the j-th anomalous cluster of the i-th component at the current acquisition time. This represents the total number of abnormal clusters for the i-th component at the current acquisition time.

[0009] Preferably, the process of obtaining the marginal spectral entropy value of each component at the current acquisition time is as follows: the vibration frequency sequence of each component at the current acquisition time is used as the input of the Hilbert-Huang transform marginal spectral entropy analysis algorithm, and the empirical mode decomposition stopping criterion is used to output the marginal spectral entropy value of each component at the current acquisition time.

[0010] Preferably, the process for obtaining the coupling strength factor of each component at the current acquisition time is as follows: The vibration frequency sequence and temperature sequence of each component at each acquisition time are used as input to the mutual information algorithm, and the mutual information value of each component at each acquisition time is output; the vibration frequency sequence and tension sequence of each component at each acquisition time are used as input to the covariance analysis algorithm, and the covariance value of each component at each acquisition time is output. The sequence of mutual information values ​​of each component at the current acquisition time and all previous acquisition times, arranged in ascending order of time, is denoted as the mutual information sequence of each component at the current acquisition time; the sequence of covariance values ​​of each component at the current acquisition time and all previous acquisition times, arranged in ascending order of time, is denoted as the covariance sequence of each component at the current acquisition time. The mutual information sequence and covariance sequence of each component at the current acquisition time are used as inputs to the weighted moving average algorithm, and the output fusion value is recorded as the coupling strength factor of each component at the current acquisition time.

[0011] Preferably, the formula for calculating the coupling field strength risk of each component at the current acquisition time is: In the formula, Let represent the coupling field strength risk level of the i-th component at the current acquisition moment. Let be the implicit risk index of the i-th component at the current data collection moment. Let be the coupling strength factor of the i-th component at the current acquisition time.

[0012] Preferably, the formula for calculating the resource weight of each component at the current acquisition time is: In the formula, Let i be the resource weight of the i-th component at the current acquisition time. Let represent the coupling field strength risk level of the i-th component at the current acquisition moment. This represents the sum of the coupling field strength risk of all components at the current acquisition moment.

[0013] Preferably, the specific process of classifying the components at the current acquisition time is as follows: when the resource weight of any component at the current acquisition time is greater than the preset monitoring threshold, the component at the current acquisition time is regarded as a high-risk component; otherwise, the component at the current acquisition time is regarded as a low-risk component.

[0014] Preferably, the specific process of selecting the corresponding security analysis model according to the category of each component at the current acquisition time is as follows: if any component at the current acquisition time is a high-risk component, then the high-precision full-order coupled simulation model is used as the security analysis model of the component at the current acquisition time; otherwise, the reduced-order proxy model is used as the security analysis model of the component at the current acquisition time.

[0015] This application has at least the following beneficial effects: 1. By analyzing the marginal spectral entropy value and stress concentration risk characteristics of the vibration frequency sequence, a latent risk index for each component at the current acquisition time is constructed, which directly characterizes the degree of synergistic harm of stress concentration and vibration randomness, making the subsequent risk assessment of each component more accurate.

[0016] 2. To address the limitation of implicit risk indices in capturing time-varying cumulative effects, this paper uses vibration, temperature, and tension sequences as inputs, employs a mutual information algorithm to quantify the nonlinear effects of temperature and vibration, evaluates the correlation between tension and vibration diffusion through covariance analysis, and further uses weighted moving average fusion to determine the coupling strength. This reflects the combined effect strength of the tension-temperature-vibration field in real time, eliminates random interference in the dynamic interaction of environmental loads, and assesses the failure risk of each component from the overall trend of risk evolution, thereby improving the accuracy of risk assessment.

[0017] 3. Based on the real-time coupled field strength risk degree calculation of component resource weights, each component is classified, so that high-risk components are executed with high-precision full-order coupled simulation models, while low-risk components are executed with simplified models. This breaks through the zero-sum game of accuracy and efficiency in traditional lightweight models, focuses limited resources on high-risk components, avoids redundant calculations for non-critical components, and improves the efficiency of safety analysis while ensuring the accuracy of component safety analysis. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating the steps of an efficient analysis method for the safety of components in a BIM model provided in this application; Figure 2 A flowchart illustrating the process of obtaining the resource weights of each component provided in this application at the current acquisition time. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an efficient analysis method for the safety of components in a BIM model proposed according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0022] The following, in conjunction with the accompanying drawings, details the specific scheme of an efficient analysis method for the safety of components in a BIM model provided in this application.

[0023] This application provides an efficient method for analyzing the safety of components in a BIM model, specifically, the following efficient method for analyzing the safety of components in a BIM model is provided. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps: Step 1: Obtain the vibration frequency sequence, temperature sequence, and tension sequence of each component at each acquisition time.

[0024] In the BIM model, all components in the BIM model are obtained by parsing the hierarchical structure and attribute set of IFC standard components, and extracting all physical entity instances containing geometric topology and physical parameters through the hierarchical structure and attribute set of IFC standard components.

[0025] Virtual triaxial accelerometers are attached to the center-of-gravity coordinates of each component to extract vibration frequency data in real time from the structural dynamics simulation module, which is used to quantify the hidden fatigue risk caused by vortex-induced resonance. For each component, virtual infrared temperature sensors are embedded in the thermally coupled mesh nodes on its surface to synchronously acquire surface temperature data of the component, which is used to monitor the nonlinear effect of thermal stress on material properties. Virtual tension monitoring points are loaded at the finite element nodes of each component to acquire conductor tension data in real time, which is used to characterize the dynamic mechanical load under wind vibration and icing conditions.

[0026] The data acquisition frequency of the three types of virtual sensors was set to 10Hz, and the three types of data were processed for noise reduction using a moving average filtering algorithm. The vibration frequency, temperature, and tension data of each component were normalized separately. Specifically, this embodiment uses a minimum-maximum normalization method to normalize each type of data to eliminate dimensional differences. Both the moving average filtering algorithm and the minimum-maximum normalization method are well-known techniques, and their specific processes will not be described in detail.

[0027] For the denoised and normalized data finally acquired for each component, the various types of data of each component from all acquisition times from the start acquisition time to the current acquisition time are sorted in ascending order of time to obtain the vibration frequency sequence, temperature sequence and tension sequence of each component at the current acquisition time, providing a high signal-to-noise ratio input for subsequent multi-field coupling analysis.

[0028] Similarly, the vibration frequency sequence, temperature sequence, and tension sequence of each component at each acquisition time are obtained.

[0029] Step 2: Cluster the data of each component in the vibration frequency sequence at the current acquisition time. Obtain the local density of each cluster based on the total number of data in each cluster and its corresponding area, and compare it with the preset density threshold to obtain the abnormal clusters of each component at the current acquisition time. Based on the randomness of the vibration frequency sequence of each component at the current acquisition time and the average level of the local density of all abnormal clusters of each component, obtain the latent risk index of each component at the current acquisition time.

[0030] Building components are prone to implicit stress concentration and nonlinear deformation under complex environmental loads, such as wind vibration, temperature changes, and the coupling of multiple physical fields of structural stress. Traditional simplified models often ignore or weaken the nonlinear correlation mechanisms between these physical fields, making it difficult to effectively identify and warn of potential safety hazards, thus increasing the risk of structural failure.

[0031] Because vibration causes dynamic stress within components, abnormal vibration often leads to stress concentration, increasing the risk of component failure. Therefore, taking the i-th component as an example, the vibration frequency sequence of the i-th component at the current acquisition time is used as input to the DBSCAN clustering algorithm. A neighborhood radius of 15 and a minimum sample size of 5 are set to ensure that outliers are not noise. The output is all clusters after clustering. The DBSCAN clustering algorithm is a well-known technique, and its specific process will not be elaborated further. Subsequently, the local density of all clusters is calculated. Specifically, the local density is calculated as follows: the cluster area of ​​each cluster is calculated using the convex hull algorithm, and the ratio of the total number of data points within each cluster to the cluster area is taken as the local density of each cluster. Clusters with high local density represent that the vibration frequency data of the cluster is highly concentrated in spatial distribution, indicating a risk hotspot area for stress resonance in the component. Therefore, a preset density threshold is set. In one embodiment of this application, the preset density threshold is the average of the local densities of all clusters of the i-th component at the current acquisition time; clusters with local densities greater than the preset density threshold among all clusters of the i-th component at the current acquisition time are taken as abnormal clusters of the i-th component at the current acquisition time, representing the stress over-limit risk area of ​​the i-th component due to abnormal vibration.

[0032] The vibration frequency sequence of the i-th component at the current acquisition time is used as input to the Hilbert-Huang Transform (HHT) marginal spectral entropy analysis algorithm. Using the empirical mode decomposition stopping criterion, the standard deviation threshold is set to 0.2 to avoid over-decomposition, and the marginal spectral frequency resolution is set to 0.5 Hz. The output is the marginal spectral entropy value of the i-th component at the current acquisition time, thus characterizing the randomness of the vibration energy distribution in the frequency domain. The Hilbert-Huang Transform (HHT) marginal spectral entropy analysis algorithm is a well-known technique, and its specific process will not be elaborated further.

[0033] As a preferred implementation, based on the randomness of the vibration frequency sequence of each component at the current acquisition time and the average level of the local density of all abnormal clusters of each component, the latent risk index of each component at the current acquisition time is obtained, which is used to characterize the degree of synergistic harm of stress concentration and vibration energy randomness of each component under multi-field coupling.

[0034] In this embodiment, the latent risk index of the i-th component at the current data collection time is denoted as . The specific calculation formula is as follows: In the formula, Let be the implicit risk index of the i-th component at the current data collection moment. Let be the marginal spectral entropy value of the i-th component at the current acquisition time. This value directly reflects the structural response state of the component under dynamic load by capturing the nonlinear characteristics of vibration behavior. The larger the value, the stronger the randomness of the vibration energy distribution, the more likely it is to be caused by complex multi-field coupling, and the more likely the structure is to be in an unstable state, which can easily lead to hidden fatigue or stress diffusion. The local density of the j-th anomalous cluster of the i-th component at the current acquisition time quantifies the severity of the stress concentration area of ​​the component. The larger the value, the higher the stress concentration and the greater the risk of the structure. This represents the total number of abnormal clusters for the i-th component at the current acquisition time.

[0035] The latent risk index is used to quantify the comprehensive degree of latent failure risk of each component. It characterizes the degree of synergistic harm caused by stress concentration and randomness of vibration energy under multi-field coupling. Its value maps the failure risk level of the component in real time. The larger the value, the higher the latent failure risk of the component.

[0036] Step 3: Based on the interdependence between the vibration frequency sequence and temperature sequence of each component at the current acquisition time and all previous acquisition times, as well as the correlation between the vibration frequency sequence and tension sequence, obtain the coupling strength factor of each component at the current acquisition time. Combined with the implicit risk index of each component at the current acquisition time, obtain the coupling field strength risk degree of each component at the current acquisition time. Combined with the sum of the coupling field strength risk degrees of all components at the current acquisition time, obtain the resource weight of each component at the current acquisition time, and then classify the components at the current acquisition time. Select the corresponding safety analysis model according to the category of each component at the current acquisition time.

[0037] Because the safety of components in the BIM model is affected by the complex coupling of multiple physical fields such as structural load, temperature, and vibration, and its risks are time-varying, cumulative, and spatially correlated, it is difficult to accurately predict the risk evolution trend and achieve real-time early warning of high-risk scenarios by relying solely on the constructed implicit risk index, which poses a risk of delayed or misjudgment of early warning.

[0038] The greater the interdependence between temperature data and vibration frequency data, the greater the influence of temperature on vibration frequency, and thus the greater the likelihood of component failure. Therefore, using the vibration frequency and temperature sequences of the i-th component at each acquisition time as input, a mutual information algorithm is used. A dynamic sliding window of 120 seconds is set to cover typical wind-induced vibration cycles, and the histogram is divided into 20 bins to balance accuracy and real-time performance. The final output is the mutual information value of the i-th component at each acquisition time, thereby quantifying the interdependence between temperature and vibration frequency data.

[0039] To assess the impact of tension changes on vibration propagation, the correlation between vibration and tension data can be analyzed. A higher correlation indicates a greater impact of tension changes on vibration propagation, thus indicating a higher risk for the component. Based on this, using the vibration frequency sequence and tension sequence of the i-th component at each acquisition time as input, a covariance analysis algorithm is used. In this embodiment, the attenuation factor is set to 0.9. The output is the covariance value of the i-th component at each acquisition time, which characterizes the correlation between the tension and vibration frequency of the component at each acquisition time.

[0040] The sequence of mutual information values ​​of the i-th component at the current acquisition time and all previous acquisition times, arranged in ascending order of time, is denoted as the mutual information sequence of the i-th component at the current acquisition time; the sequence of covariance values ​​of the i-th component at the current acquisition time and all previous acquisition times, arranged in ascending order of time, is denoted as the covariance sequence of the i-th component at the current acquisition time.

[0041] Subsequently, for each component at each acquisition time, the mutual information sequence and covariance sequence of the i-th component at the current acquisition time are used as input. A weighted moving average (WMA) algorithm is used to smoothly fuse the mutual information sequence and covariance sequence, with a time window of n seconds. In this embodiment, the weights of both parameters are 0.5. The output fused value is recorded as the coupling strength factor of the i-th component at the current acquisition time, used to characterize the strength of the multi-field coupling synergy effect at the current acquisition time. It analyzes the dynamic interaction of environmental loads; the larger the value, the stronger the multi-field coupling synergy effect, and the greater the potential for accelerated risk evolution. In this embodiment, n is set to 60.

[0042] It should be noted that the mutual information algorithm, covariance analysis algorithm, and weighted moving average algorithm are all well-known techniques in the field, and the specific processes will not be elaborated here.

[0043] As a preferred implementation, the coupling field strength risk of each component at the current acquisition time is obtained based on the coupling strength factor and implicit risk index of each component at the current acquisition time, which is used to characterize the comprehensive failure risk of each component at the current acquisition time.

[0044] In this embodiment, the coupling field strength risk of the i-th component at the current acquisition time is denoted as . Its specific expression is: In the formula, Let represent the coupling field strength risk level of the i-th component at the current acquisition moment. The implicit risk index of the i-th component at the current acquisition time captures the quasi-static risk characteristics of implicit failure of the component under dynamic load. The larger the value, the stronger the synergistic harm of stress concentration and vibration randomness in the structural response of the component, corresponding to a higher probability of fatigue failure or stress diffusion. Let be the coupling strength factor of the i-th component at the current acquisition time.

[0045] The coupled field strength risk level represents the comprehensive failure risk intensity of components under complex multi-field coupling. It quantifies the combined effect of the implicit structural risk and the dynamic coupling strength of the environment in real time, thereby solving the nonlinear, time-varying cumulative, and spatially correlated risks that are difficult to capture by traditional models. The larger the value, the greater the risk of failure of the i-th component at the current acquisition time, corresponding to a vicious cycle of multiple interaction reinforcements.

[0046] Because BIM models need to respond to changes in construction status in real time and simultaneously handle multi-field coupled simulation of structural thermal vibration in component safety analysis, the demand for computing resources grows exponentially. The incremental update strategy adopted by traditional lightweight models is prone to error accumulation in complex interaction chains, and resource allocation is limited by the trade-off between high-precision simulation of key areas and simplification of non-key areas, which cannot simultaneously guarantee the timeliness of early warning for high-risk scenarios and the accuracy of global calculation.

[0047] Based on the real-time calculated coupling field strength risk, the resource weights of each component at the current acquisition time are obtained. In this embodiment, the resource weight of the i-th component at the current acquisition time is denoted as... Its specific expression is: In the formula, Let i be the resource weight of the i-th component at the current acquisition time. Let represent the coupling field strength risk level of the i-th component at the current acquisition moment. This represents the sum of the coupling field strength risk of all components at the current acquisition moment. The larger the value, the greater the risk of failure of the i-th component relative to other components at the current acquisition time. The process for obtaining the resource weight of each component at the current acquisition time is as follows: Figure 2 As shown.

[0048] Furthermore, by calculating the resource weight of each component in real time, the components at the current data collection moment are classified. Specifically, preset monitoring thresholds are set. In this embodiment, the preset monitoring threshold is 0.1. When, the i-th component at the current acquisition time is designated as a high-risk component; when At the current acquisition time, the i-th component is designated as a low-risk component. Furthermore, based on the category of each component, the safety analysis model for each component is switched at millisecond levels. When the i-th component at the current acquisition time is a high-risk component, the high-precision full-order coupled simulation model is used as the safety analysis model for that component at the current acquisition time; when the i-th component at the current acquisition time is a low-risk component, the reduced-order surrogate model is used as the safety analysis model for that component at the current acquisition time. During the model switching process, incremental variable transfer is used to ensure the continuity of the physical field.

[0049] It should be noted that the high-precision full-order coupled simulation model employs full-order multiphysics coupled simulation for high-risk components. Based on real-time updated boundary conditions, it performs transient finite element analysis with millisecond-level time steps, fully solving the structural dynamics equations, heat conduction equations, and nonlinear material constitutive models. It simultaneously iteratively calculates the thermal stress distribution and vibration response spectrum, thereby accurately capturing the effects of latent crack propagation and fatigue accumulation. The reduced-order surrogate model is a simplified version of the high-precision full-order coupled simulation model, differing from it in the following ways: the finite element node degrees of freedom are compressed to 5% of the original model, retaining only key modal parameters. Non-master degrees of freedom are eliminated using the static condensation method, and nonlinear calculations are replaced by a linear equivalent stiffness matrix. Simultaneously, non-critical physics fields are frozen, reducing the computational load. The construction of both the high-precision full-order coupled simulation model and the reduced-order surrogate model are techniques well-known to those skilled in the art and will not be elaborated upon here.

[0050] By dynamically quantifying the failure risk intensity of each component through real-time calculation of the coupling field strength risk, and classifying each component accordingly, computational resources are tilted towards high-risk components. Priority is given to ensuring the simulation accuracy and early warning timeliness of components in high stress concentration and multi-field strong coupling regions, while avoiding redundant calculations of low-risk components. This effectively suppresses error accumulation and optimizes the global model analysis efficiency under limited resources, overcoming the shortcomings of traditional lightweight models in high-risk scenarios such as lag in response or imbalance in resource allocation.

[0051] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0052] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0053] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An efficient method for analyzing the safety of components in a BIM model, characterized in that, The method includes the following steps: Obtain the vibration frequency sequence, temperature sequence, and tension sequence of each component at each acquisition time; The data of each component in the vibration frequency sequence at the current acquisition time are clustered. The local density of each cluster is obtained based on the total number of data in each cluster and its corresponding area. This density is then compared with a preset density threshold to obtain the abnormal clusters of each component at the current acquisition time. Based on the randomness of the vibration frequency sequence of each component at the current acquisition time and the average level of the local density of all abnormal clusters of each component, the latent risk index of each component at the current acquisition time is obtained. Based on the interdependence between the vibration frequency sequence and temperature sequence of each component at the current acquisition time and all previous acquisition times, as well as the correlation between the vibration frequency sequence and tension sequence, the coupling strength factor of each component at the current acquisition time is obtained. Combined with the implicit risk index of each component at the current acquisition time, the coupling field strength risk degree of each component at the current acquisition time is obtained. By combining the sum of the coupling field strength risk degrees of all components at the current acquisition time, the resource weight of each component at the current acquisition time is obtained, and then the components at the current acquisition time are classified. The corresponding safety analysis model is selected according to the category of each component at the current acquisition time. The process for obtaining the latent risk index of each component at the current data collection time is as follows: Based on the randomness of the vibration frequency sequence of each component at the current acquisition time, the marginal spectral entropy value of each component at the current acquisition time is obtained. Calculate the latent risk index of each component at the current data acquisition moment: In the formula, Let be the implicit risk index of the i-th component at the current data collection moment. Let be the marginal spectral entropy value of the i-th component at the current acquisition time. Let be the local density of the j-th anomalous cluster of the i-th component at the current acquisition time. This represents the total number of abnormal clusters for the i-th component at the current acquisition time.

2. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The process of obtaining the local density of each cluster is as follows: obtain the cluster area of ​​each cluster, and take the ratio of the total number of data points in each cluster to the cluster area as the local density of each cluster.

3. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The abnormal clusters of each component at the current acquisition time refer to the clusters among all clusters of each component at the current acquisition time whose local density is greater than a preset density threshold.

4. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The process of obtaining the marginal spectral entropy value of each component at the current acquisition time is as follows: the vibration frequency sequence of each component at the current acquisition time is used as the input of the Hilbert-Huang transform marginal spectral entropy analysis algorithm, and the empirical mode decomposition stopping criterion is used to output the marginal spectral entropy value of each component at the current acquisition time.

5. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The process for obtaining the coupling strength factor of each component at the current acquisition time is as follows: The vibration frequency sequence and temperature sequence of each component at each acquisition time are used as input to the mutual information algorithm, and the mutual information value of each component at each acquisition time is output; the vibration frequency sequence and tension sequence of each component at each acquisition time are used as input to the covariance analysis algorithm, and the covariance value of each component at each acquisition time is output. The sequence of mutual information values ​​of each component at the current acquisition time and all previous acquisition times, arranged in ascending order of time, is denoted as the mutual information sequence of each component at the current acquisition time; the sequence of covariance values ​​of each component at the current acquisition time and all previous acquisition times, arranged in ascending order of time, is denoted as the covariance sequence of each component at the current acquisition time. The mutual information sequence and covariance sequence of each component at the current acquisition time are used as inputs to the weighted moving average algorithm, and the output fusion value is recorded as the coupling strength factor of each component at the current acquisition time.

6. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The formula for calculating the coupling field strength risk of each component at the current acquisition time is as follows: In the formula, Let represent the coupling field strength risk level of the i-th component at the current acquisition moment. Let be the implicit risk index of the i-th component at the current data collection moment. Let be the coupling strength factor of the i-th component at the current acquisition time.

7. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The formula for calculating the resource weight of each component at the current acquisition time is as follows: In the formula, Let i be the resource weight of the i-th component at the current acquisition time. Let represent the coupling field strength risk level of the i-th component at the current acquisition moment. This represents the sum of the coupling field strength risk of all components at the current acquisition moment.

8. The efficient analysis method for the safety of components in a BIM model as described in claim 1, characterized in that, The specific process of classifying components at the current acquisition time is as follows: when the resource weight of any component at the current acquisition time is greater than the preset monitoring threshold, the component at the current acquisition time is regarded as a high-risk component. Conversely, the component at the current acquisition time will be considered a low-risk component.

9. The efficient analysis method for the safety of components in a BIM model as described in claim 8, characterized in that, The specific process of selecting the corresponding safety analysis model based on the category of each component at the current acquisition time is as follows: if any component at the current acquisition time is a high-risk component, then the high-precision full-order coupled simulation model is used as the safety analysis model of that component at the current acquisition time. Otherwise, the downgraded proxy model will be used as the security analysis model for this component at the current acquisition time.

Images