Bridge health monitoring method and system based on distributed bridge disease data summarization

Abstract

The invention relates to the technical field of big data processing bridge health monitoring, in particular to a distributed bridge disease data summarization bridge health monitoring method and system, and the method comprises the steps: obtaining vibration data and steel beam stress data of a bridge at a plurality of positions, and wind speed data near the bridge; obtaining the validity coefficient of each modal component of the vibration data to extract the de-noised vibration data in each time period of the current position, calculating the significant value of the dominant frequency down-shift characteristic of the de-noised vibration data in each time period of the current position, and further obtaining the abnormal coefficient of the bridge floor vibration in each time period of the current position; and obtaining the abnormal coefficient of the bridge deck steel beam stress at the current position in each time period by combining the random fluctuation degree of the de-noised steel beam stress data in each time period, and obtaining the significant value of the bridge load influence by combining the average distribution level of the wind speed data for monitoring the health state of the bridge. The bridge health monitoring precision can be improved.

Description

Technical Field

[0001] This application relates to the field of big data processing for bridge health monitoring technology, specifically to a bridge health monitoring method and system for distributed bridge defect data aggregation. Background Technology

[0002] Because bridge engineering projects are characterized by their dispersed distribution and long spans, they are susceptible to damage from environmental changes and human activities throughout their lifespan. Therefore, it is necessary to monitor various parameters of bridge operation and comprehensively assess their service condition. Bridge health monitoring generates a large amount of data, such as stress, displacement, and vibration information. By analyzing and mining this data, information reflecting the health status of the bridge can be extracted.

[0003] However, the raw data collected is often affected by clutter noise, leading to anomalies and impacting health monitoring results. Furthermore, bridges are subjected to various loads and random environmental vibrations during their service life, making it difficult to accurately identify bridge fault characteristics, resulting in low accuracy in bridge health monitoring. Publication number CN117648734A describes a smart bridge health monitoring method and assessment system that uses strain reconstruction at monitoring points to obtain strain data at test points and uses strain deviation and stability coefficient to determine the degree of damage. However, this method fails to adequately consider the effects of clutter noise and various loads during the monitoring process, resulting in insufficient accuracy in bridge health monitoring. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a method and system for bridge health monitoring that aggregates distributed bridge defect data. The specific technical solution adopted is as follows: This application provides a bridge health monitoring method based on distributed bridge defect data aggregation, including the following steps: Obtain vibration data, steel strand stress data, and wind speed data near the bridge at multiple locations; By analyzing the distribution characteristics of the modal components of the vibration data at the current location and the corresponding center frequencies, the effectiveness coefficients of each modal component of the vibration data are obtained to extract the denoised vibration data at the current location. Based on the frequency change trend characteristics of the denoised vibration data in the frequency domain at each time period, the significant values ​​of the main frequency shift characteristics of the denoised vibration data at the current location are obtained. Furthermore, the local peak fluctuation characteristics and the distribution symmetry characteristics of the denoised vibration data at each time period are analyzed to obtain the abnormal coefficients of the bridge deck vibration at the current location at each time period. Denoising steel strand stress data is obtained to determine the significance of the downward shift of the dominant frequency of the denoising steel strand stress data in each time period. Combined with the degree of random fluctuation of the denoising steel strand stress data in each time period, the anomaly coefficient of the bridge deck steel strand stress in each time period at the current location is obtained. By the difference between the anomaly coefficient of bridge deck vibration and the anomaly coefficient of steel strand stress at all locations of the bridge in each time period, as well as the average distribution level of wind speed data, the significance of the bridge load influence is obtained, which is used to monitor the health status of the bridge.

[0005] Preferably, the effectiveness coefficients of each modal component of the vibration data are obtained as follows: In the formula, This represents the validity coefficient of the j-th modal component of the vibration data at the current location within the current time period. This represents the normalized result of the kurtosis value corresponding to the j-th modal component of the vibration data at the current location during the current time period. This represents the normalized result of the center frequency value corresponding to the j-th modal component of the vibration data at the current location during the current time period. To avoid constants with numerators or denominators equal to 0, modal decomposition is performed on the vibration data within the current time period to obtain each modal component.

[0006] Preferably, the modal components corresponding to the minimum effectiveness coefficient of the current location in each time period are removed, and the remaining modal components are reconstructed to obtain the denoised vibration data of the current location in each time period.

[0007] Preferably, the significance value of the downward shift of the dominant frequency of the denoised vibration data in each time period is the trend test result of all major frequencies in the nearest neighboring time periods of each time period. Specifically, the frequency domain transformation of the denoised vibration data in each time period is performed and the spectral centroid of the spectrum is extracted. The frequency corresponding to the position of the spectral centroid is taken as the dominant frequency, and each time period and the previous multiple time periods are taken as the nearest neighboring time periods of each time period.

[0008] Preferably, the abnormal coefficient of bridge deck vibration in each time period is obtained as follows: In the formula, In the formula, Let be the anomaly coefficient of the bridge deck vibration during the i-th time period at the current location. This represents the significant value of the vibration data at the current location within the i-th time period that exhibits local spike characteristics. This represents the skewness value of the denoised vibration data within the i-th time period at the current location. Let exp() be the significant value of the downward shift of the dominant frequency of the denoised vibration data in the i-th time period at the current position, and let e represent the exponential function with base e.

[0009] Preferably, the mean of the peak points of all data with values ​​greater than 0 and the root square amplitude of all data with values ​​greater than 0 in the denoised vibration data of the current location in each time period are extracted, and the ratio of the mean to the root square amplitude is calculated and used as a significant value for the presence of local peak features in the vibration data of the current location in each time period.

[0010] Preferably, the abnormality coefficient of the bridge deck steel strand stress at the current location within each time period is obtained as follows: In the formula, Let be the anomaly coefficient of the bridge deck steel strand stress during the i-th time period at the current location. This represents the significant value of the downward shift characteristic of the dominant frequency of the denoised steel strand stress data in the i-th time period at the current position. Let be the Shannon entropy of the denoised steel strand stress data in the i-th time period, and exp() represent an exponential function with base e.

[0011] Preferably, the abnormal coefficients of bridge deck vibration and steel strand stress at all locations of the bridge during the same time period are arranged in order of position, and the distance between the two sequences is taken as the abnormal synchronization value of bridge vibration and steel strand stress during the same time period.

[0012] Preferably, the average wind speed for each time period is calculated separately, and the normalized result of the correlation coefficient between the average wind speed for all time periods and the abnormal synchronization value is taken as the significant value of the bridge load impact. If the significant value of the bridge load impact is greater than or equal to the preset health risk threshold, then the bridge has a health risk; otherwise, the bridge does not have a health risk.

[0013] This application also provides a bridge health monitoring system for distributed bridge defect data aggregation, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the bridge health monitoring method for distributed bridge defect data aggregation described above.

[0014] As can be seen from the above, the bridge health monitoring method and system for distributed bridge defect data aggregation provided in this application has at least the following beneficial effects: This application analyzes the frequency distribution differences of monitoring data and the sharpness characteristics of the corresponding component signals at each frequency to analyze the impact of noise and remove it. Its advantage lies in reducing the interference of different mixed noises in the environment on the measured data. Furthermore, it obtains the abnormal characteristics of the vibration and steel strand stress at various locations of the bridge under the influence of structural damage, and considers the interaction between bridge deck vibration and steel strand stress as well as the influence characteristics of wind load. In this way, it analyzes the degree of structural damage to the bridge and conducts a health status assessment, which helps to improve the accuracy of bridge health monitoring. Attached Figure Description

[0015] 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.

[0016] Figure 1 A flowchart illustrating the steps of a distributed bridge defect data aggregation method for bridge health monitoring provided in this application. Detailed Implementation

[0017] 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 a bridge health monitoring method and system for distributed bridge defect data aggregation proposed in 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.

[0018] Unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. 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.

[0019] The following description, in conjunction with the accompanying drawings, details the specific scheme of the bridge health monitoring method and system for distributed bridge defect data aggregation provided in this application.

[0020] Please see Figure 1 The diagram illustrates a flowchart of a bridge health monitoring method for distributed bridge defect data aggregation according to an embodiment of this application, including the following steps: Step 1: Obtain vibration data, steel strand stress data, and wind speed data near the bridge at multiple locations.

[0021] In this embodiment, the distributed bridge monitoring system connects the hardware facilities of various monitoring areas, forming a vertically integrated data transmission network. The monitoring center platform serves as the regional data aggregation center, achieving efficient application of massive amounts of data through big data processing technology. Each area is equipped with an independent hardware platform and monitoring platform to monitor the status of the bridges under its jurisdiction. The hardware platform connects to the bridge's on-site sensor array via a bridgehead switch, collecting vibration data and steel strand stress data at multiple locations on the bridge, as well as wind speed data near the bridge. The monitoring platform handles data preprocessing to ensure data quality meets upload requirements. The sampling frequency for vibration data and steel strand stress is set to 100Hz, and the sampling frequency for wind speed data is set to 1Hz. The bridge monitoring data is ultimately uploaded to the monitoring center via the detection platform, where it undergoes big data processing to achieve bridge health monitoring.

[0022] Step 2: By analyzing the distribution characteristics of the modal components of the vibration data in each time period and their corresponding center frequencies, the effectiveness coefficients of each modal component of the vibration data are obtained to extract the denoised vibration data in each time period. Based on the frequency change trend characteristics of the denoised vibration data in the frequency domain in each time period, the significance value of the downward shift of the main frequency of the denoised vibration data in each time period is obtained. The local peak fluctuation characteristics and the distribution symmetry characteristics of the denoised vibration data in each time period are analyzed to obtain the abnormal coefficient of the bridge deck vibration in each time period.

[0023] During health monitoring, the collected data is often affected by noise, and the combined effect of multiple load factors makes bridge monitoring data complex and variable, making it difficult to obtain accurate fault characteristics. Therefore, this embodiment will describe in detail the processing of monitoring data from any one of the bridges.

[0024] To improve the data reliability of bridge health monitoring systems, data cleaning is essential and crucial for ensuring the system's effective operation. Sensor malfunctions, data transmission errors, or other unexpected events can cause outliers. This application uses vibration data as an example for data cleaning. Bridges have a large mass, and wind and vehicle loads excite them relatively slowly. Furthermore, due to material damping and frictional dissipation, the vibration information generated by the overall stress and deformation of the bridge primarily appears in low-frequency form. In contrast, vibration information generated by environmental noise is unrelated to the overall structural deformation of the bridge and easily causes high-frequency vibrations. Therefore, information reflecting the structural state of a bridge is typically concentrated in the low-frequency range, while environmental noise and other interference are located in the high-frequency range.

[0025] Therefore, in this embodiment, variational mode decomposition (VMD) is used to process the vibration data to obtain different frequency components. Taking vibration data from any time period at the current location as an example, the time period is set to 5 minutes. The vibration data within this time period is used as the input to the variational mode decomposition algorithm, and the number of modal components output by the VMD algorithm is set to 9. Each modal component corresponds to a specific center frequency, which can be obtained by the VMD algorithm. Furthermore, the effective modal components have a higher degree of sharpness, while the components corresponding to noise should have strong randomness and lower sharpness. The kurtosis value corresponding to each modal component is then calculated. The larger the kurtosis, the sharper the modal component. All center frequency values ​​and kurtosis values ​​are subjected to max-min normalization. The normalization result has a value range of [0,1]. The effectiveness coefficient of each modal component of the vibration data at the current location within the current time period is then calculated. In this embodiment, the specific calculation formula is as follows: In the formula, This represents the validity coefficient of the j-th modal component of the vibration data at the current location within the current time period. This represents the normalized result of the kurtosis value corresponding to the j-th modal component of the vibration data at the current location during the current time period. This represents the normalized result of the center frequency value corresponding to the j-th modal component of the vibration data at the current location during the current time period. To avoid constants with numerators or denominators of 0, the value range is 0 to 0.1; in this embodiment, it is set to 0.001. The obtained... The larger the value, the more effective information the modal function component contains. The modal component corresponding to the minimum effectiveness coefficient is discarded as a noisy modal component, and then all the remaining modal components are reconstructed to obtain the denoised vibration data for the current location and time period, which helps to reduce the interference of mixed noise.

[0026] Furthermore, vibration and steel strand stress states can effectively reflect the load effects on the bridge structure. Taking vibration data collected at the current location of the bridge as an example, specifically, when the bridge has structural damage, under the action of multiple loads, its vibration data exhibits a downward shift in the dominant frequency, and stiffness loss easily leads to local spikes and waveform distortion in the vibration data. Therefore, the fast discrete Fourier transform is used to obtain the spectrum of the denoised vibration data at each time period at the current location, and then the spectral centroid of each spectrum is calculated. The spectral centroid corresponds to the region where the frequency energy distribution is most concentrated, and the frequency corresponding to the position of the spectral centroid is taken as the dominant frequency. The current position's i-th time period and its preceding N time periods are considered as the i-th time period's nearest neighbor time periods, where N ranges from [12, 15], and in this embodiment, it is set to 15. The Thiel-Sen estimation method is used to calculate the trend characteristics of all major frequencies within the nearest neighbor time periods. In this embodiment, the Mann-Kendall trend test algorithm is used to analyze the trends of all major frequencies within the i-th time period's nearest neighbor time periods. The trend test result is taken as the significance value of the downward shift characteristic of the dominant frequency in the denoised vibration data of the current position's i-th time period, denoted as... The result The smaller the value, the more significant the downward shift in the dominant frequency of the vibration data within that neighboring time period. It should be noted that the specific application process of the Mann-Kendall trend test algorithm is a well-known technique, and this embodiment does not impose any special restrictions, so it will not be elaborated upon here.

[0027] Furthermore, to analyze the local peaks and waveform distortion characteristics of the denoised vibration data, we again take the denoised vibration data of the i-th time period at the current position as an example. Since the collected vibration data typically exhibits an up-and-down reciprocating pattern, this embodiment uses an automatic multi-scale peak finding algorithm to obtain the peak points of all data points greater than 0 in the denoised vibration data of the current position within this time period. Then, the mean of all these peaks is calculated and denoted as... And the square root magnitude of all data points greater than 0 within that period, denoted as The result This reflects the degree of protrusion of local peaks, and the resulting... This reflects the average vibration energy of the denoised vibration data within that time period. Under different loads, the obtained average vibration energy used to reflect the denoised vibration data... The smaller the value, the smaller the overall fluctuation of the denoised vibration data, and the better the result used to reflect the degree of local peak protrusion. The larger the proportion of the value in the overall fluctuation, the more significant the corresponding local peak feature. Therefore, the significance value of the vibration data with local peak features in the i-th time period at the current position is calculated by the following formula: The result This reflects the local spike characteristics of the vibration data during this period.

[0028] Meanwhile, when the bridge structure is in good condition, the vibration data exhibits a high degree of vertical symmetry. However, if structural damage causes severe distortion of the vibration waveform, the symmetry characteristics will significantly decrease. Therefore, in this embodiment, the skewness value of the denoised vibration data in the i-th time period is calculated and denoted as . The calculation of the skewness value is a well-known technique, and the specific process will not be described in detail here. The larger the value, the more asymmetric the denoised vibration data is within that time period.

[0029] Therefore, in this embodiment, based on the significant values ​​of local peak features in the vibration data at the current location for each time period and the skewness value of the denoised vibration data, combined with the significant values ​​of the downward shift of the dominant frequency of the denoised vibration data for each time period, the anomaly coefficient of the bridge deck vibration at the current location for each time period is calculated. The formula is as follows: In the formula, Let be the anomaly coefficient of the bridge deck vibration during the i-th time period at the current location. This represents the significant value of the vibration data at the current location within the i-th time period that exhibits local spike characteristics. This represents the skewness value of the denoised vibration data within the i-th time period at the current location. Let be the significance value of the downward shift of the dominant frequency in the denoised vibration data for the i-th time period at the current position. exp() represents an exponential function with base e, ensuring the denominator is always positive. The resulting... This reflects the main frequency changes and waveform anomalies in the denoised vibration data during this period.

[0030] Step 3: Obtain denoised steel strand stress data to obtain significant values ​​of the downward shift characteristics of the dominant frequency of denoised steel strand stress data in each time period. Combined with the degree of random fluctuation of denoised steel strand stress data in each time period, obtain the anomaly coefficient of bridge deck steel strand stress in each time period at the current location. By the difference between the anomaly coefficient of bridge deck vibration and the anomaly coefficient of steel strand stress at all locations of the bridge in each time period, as well as the average distribution level of wind speed data, obtain significant values ​​of the bridge load influence for monitoring the bridge health status.

[0031] Large bridges commonly use cable-stayed or suspension structures with steel strands. Bridge deck loads are transferred to the main cables via the steel strands, resulting in an interaction between bridge deck vibration and the stress state of the steel strands. When structural damage occurs, the collected steel strand stress also exhibits a corresponding downward shift in the dominant frequency and significant fluctuations in stress magnitude. Taking the steel strand stress data collected at the current location as an example, the same steps described above are used to first denoise the data, obtaining denoised steel strand stress data. Then, the significant value of the downward shift in the dominant frequency of the denoised steel strand stress data at the current location during the i-th time period is obtained from the denoised steel strand stress data, denoted as [value missing]. The stress data of the steel strands differs from the vibration data. To obtain the amplitude fluctuation anomaly characteristics, this embodiment calculates the Shannon entropy of the denoised steel strand stress data in the i-th time period, denoted as . The result The larger the value, the more significant the stress fluctuation in the steel strands during that period. The anomaly coefficient of the bridge deck steel strand stress in the i-th period at the current location is then calculated using the following formula: In the formula, Let be the anomaly coefficient of the bridge deck steel strand stress during the i-th time period at the current location. Let exp() represent the significant value of the downward shift characteristic of the dominant frequency of the denoised steel strand stress data in the i-th time period at the current position. To ensure the denominator remains positive, an exponential function is used for mapping, resulting in... This reflects the main frequency variation and abnormal fluctuation characteristics of the denoised steel strand stress data during this period.

[0032] In large bridges, each location on the bridge deck has corresponding steel strand connections, and there is an interaction between bridge deck vibration and steel strand stress. This results in similar changes in vibration anomalies and steel strand stress anomalies when bridge structural damage occurs. Therefore, the anomaly coefficients of bridge deck vibration and steel strand stress at all locations in the bridge at each time period are arranged in positional order. Furthermore, in this embodiment, the DTW distance between the two sequences is calculated as the anomaly synchronization value of bridge vibration and steel strand stress at each time period. The obtained anomaly synchronization value reflects the correlation characteristics between the changes in vibration anomalies and steel strand stress anomalies. It should be noted that in this embodiment, the above positional order is only the order of the positions arranged from left to right in the same direction on the bridge. In actual application scenarios, implementers can set the positional order themselves; this embodiment does not impose any special limitations on this.

[0033] Furthermore, under wind loads, the aforementioned abnormal characteristics of bridge structural damage may become more pronounced as wind speed increases. For example, strong winds can even cause bridge deck swaying, further exacerbating the opening and closing of cracks and leading to a significant increase in abnormal synchronization values. Therefore, to consider the impact of wind loads on the bridge structural condition, this embodiment statistically analyzes the average wind speed for each time period. Preferably, the normalized Spearman correlation coefficient between the average wind speed for all time periods and the abnormal synchronization values ​​is used as the significant value of the bridge load impact. The larger the significant value of the load impact, the more severe the impact of wind loads on bridge operation, and the greater the potential structural damage.

[0034] In this embodiment, by deeply analyzing the corresponding abnormal characteristics of vibration and steel strand stress at various locations of the bridge under the influence of structural damage, and further considering the interaction between bridge deck vibration and steel strand stress as well as the influence characteristics of wind load, the degree of structural damage of the bridge is analyzed, thereby conducting a health status assessment of the bridge.

[0035] Specifically, in this embodiment, a health risk threshold is preset. If the significant value of the bridge load's impact is greater than or equal to the health risk threshold, the corresponding bridge has a health risk; otherwise, the bridge does not have a health risk and its operating status is normal. In this embodiment, the health risk threshold is set to 0.45. In actual application scenarios, implementers can set it according to the actual situation; this embodiment does not impose any special restrictions on it. The above process according to this embodiment can achieve bridge health monitoring and improve the accuracy of bridge health monitoring.

[0036] Based on the same inventive concept as the above method, this application embodiment also provides a bridge health monitoring system for distributed bridge defect data aggregation, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the bridge health monitoring method for distributed bridge defect data aggregation described above.

[0037] It is understood 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, the above description focuses on specific embodiments of this specification. 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.

[0038] 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.

[0039] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the protection scope of this application.

Claims

1. A bridge health monitoring method based on distributed bridge defect data aggregation, characterized in that, Includes the following steps: Obtain vibration data, steel strand stress data, and wind speed data near the bridge at multiple locations; By analyzing the distribution characteristics of the modal components of the vibration data at the current location and the corresponding center frequencies, the effectiveness coefficients of each modal component of the vibration data are obtained to extract the denoised vibration data at the current location. Based on the frequency change trend characteristics of the denoised vibration data in the frequency domain at each time period, the significant values ​​of the main frequency shift characteristics of the denoised vibration data at the current location are obtained. Furthermore, the local peak fluctuation characteristics and the distribution symmetry characteristics of the denoised vibration data at each time period are analyzed to obtain the abnormal coefficients of the bridge deck vibration at the current location at each time period. Denoising steel strand stress data is obtained to determine the significance of the downward shift of the dominant frequency of the denoising steel strand stress data in each time period. Combined with the degree of random fluctuation of the denoising steel strand stress data in each time period, the anomaly coefficient of the bridge deck steel strand stress in each time period at the current location is obtained. By the difference between the anomaly coefficient of bridge deck vibration and the anomaly coefficient of steel strand stress at all locations of the bridge in each time period, as well as the average distribution level of wind speed data, the significance of the bridge load influence is obtained, which is used to monitor the health status of the bridge.

2. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 1, characterized in that, The method for obtaining the validity coefficients of each modal component of the vibration data is as follows: In the formula, This represents the validity coefficient of the j-th modal component of the vibration data at the current location within the current time period. This represents the normalized result of the kurtosis value corresponding to the j-th modal component of the vibration data at the current location during the current time period. This represents the normalized result of the center frequency value corresponding to the j-th modal component of the vibration data at the current location during the current time period. To avoid constants with numerators or denominators equal to 0, modal decomposition is performed on the vibration data within the current time period to obtain each modal component.

3. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 1, characterized in that, The modal components corresponding to the minimum effectiveness coefficient of the current location in each time period are removed, and the remaining modal components are reconstructed to obtain the denoised vibration data of the current location in each time period.

4. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 1, characterized in that, The significance value of the downward shift of the dominant frequency of the denoised vibration data in each time period is the trend test result of all major frequencies in the nearest neighboring time periods of each time period. Specifically, the frequency domain transformation of the denoised vibration data in each time period is performed and the spectral centroid of the spectrum is extracted. The frequency corresponding to the position of the spectral centroid is taken as the dominant frequency, and each time period and the previous multiple time periods are taken as the nearest neighboring time periods of each time period.

5. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 1, characterized in that, The abnormal coefficients of bridge deck vibration in each time period are obtained as follows: In the formula, In the formula, Let be the anomaly coefficient of the bridge deck vibration during the i-th time period at the current location. This represents the significant value of the vibration data at the current location within the i-th time period that exhibits local spike characteristics. This represents the skewness value of the denoised vibration data within the i-th time period at the current location. Let exp() be the significant value of the downward shift of the dominant frequency of the denoised vibration data in the i-th time period at the current position, and let e represent the exponential function with base e.

6. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 5, characterized in that, Extract the mean of all peak points of the greater than 0 data in the denoised vibration data of the current location for each time period, as well as the root square amplitude of all data in the greater than 0 data. Calculate the ratio of the mean to the root square amplitude, and use it as the significant value of the vibration data of the current location for each time period that has local peak characteristics.

7. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 1, characterized in that, The method for obtaining the anomaly coefficient of the bridge deck steel strand stress at the current location in each time period is as follows: In the formula, Let be the anomaly coefficient of the bridge deck steel strand stress during the i-th time period at the current location. This represents the significant value of the downward shift characteristic of the dominant frequency of the denoised steel strand stress data in the i-th time period at the current position. Let be the Shannon entropy of the denoised steel strand stress data in the i-th time period, and exp() represent an exponential function with base e.

8. The bridge health monitoring method for distributed bridge defect data aggregation as described in claim 1, characterized in that, The abnormal coefficients of bridge deck vibration and steel strand stress at all locations of the bridge during the same time period are arranged in order of location, and the distance between the two obtained sequences is taken as the abnormal synchronization value of bridge vibration and steel strand stress during the same time period.

9. A bridge health monitoring method for distributed bridge defect data aggregation as described in claim 8, characterized in that, The average wind speed for each time period is calculated separately. The normalized result of the correlation coefficient between the average wind speed for all time periods and the abnormal synchronization value is taken as the significant value of the bridge load influence. If the significant value of the bridge load influence is greater than or equal to the preset health risk threshold, the bridge has a health risk; otherwise, the bridge does not have a health risk.

10. A bridge health monitoring system for distributed bridge defect data aggregation, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the bridge health monitoring method for distributed bridge defect data aggregation as described in any one of claims 1-9.

Images