Bridge deck slab damage detection method and system

Abstract

The invention discloses a bridge deck slab damage detection method and system, and the method comprises the steps: obtaining dynamic strain response signals of a plurality of measurement points disposed on a bridge deck slab, and processing the dynamic strain response signals to obtain strain time history data; performing modal decomposition on data of each measuring point, extracting a main strain modal shape, and calculating a shape curvature value reflecting local bending rigidity; constructing a damage collaborative analysis group by taking each measuring point as a reference point and combining a spatial proximity and structure homology principle; calculating a correlation coefficient between the reference point in the group and the vibration mode curvature value of each collaborative analysis point, and judging the collaboration based on a preset threshold value; counting an effective collaborative group proportion, performing fusion calculation to obtain a structure damage consistency index, and judging whether damage exists or not according to the structure damage consistency index; if it is judged that damage exists, the center range of the damage area is positioned based on the spatial distribution sudden change characteristics of the vibration mode curvature values. The early damage of the bridge deck slab can be efficiently and sensitively detected, accurate positioning is achieved, and reliable support is provided for preventive maintenance of a bridge.

Description

Technical Field

[0001] This invention belongs to the field of bridge structural health detection technology, and in particular relates to a method and system for detecting bridge deck damage. Background Technology

[0002] As a critical component of bridges that directly bears vehicle wheel loads and environmental effects, the early detection and precise location of bridge deck damage (such as cracking, spalling, and steel reinforcement corrosion) are crucial for ensuring bridge operational safety and guiding preventative maintenance. Currently, bridge deck damage detection methods mainly fall into two categories: one is surface inspection, such as manual inspection and high-definition image recognition, which relies on personnel experience, struggles to detect internal damage, and is inefficient; the other is detection based on dynamic response, such as vibration modal analysis and strain modal analysis, which indirectly identifies damage by deploying sensors to monitor changes in the structure's dynamic characteristics.

[0003] Existing dynamic response detection methods face significant bottlenecks in practice. First, most methods rely on establishing high-precision finite element models or signal models independently for each measurement point, resulting in high computational costs and making them unsuitable for large bridge deck structures with dense sensor deployments and numerous measurement points. Second, traditional methods often analyze data from individual measurement points in isolation, failing to fully utilize the crucial physical mechanism that bridge decks, as continuous structures, exhibit spatial correlation and propagation in their structural dynamic characteristics (such as mode shapes and curvature) changes caused by damage. For example, when damage occurs at a certain point, the decrease in local stiffness not only affects the dynamic response at that point but also influences the response modes of measurement points within a certain surrounding area. Existing point-to-point independent analysis or simple overall modal comparison methods struggle to accurately capture and quantify this "cooperative anomaly" caused by local damage and exhibiting specific spatial correlations, leading to insensitivity to early, minor damage, insufficient positioning accuracy, and susceptibility to environmental noise interference. Summary of the Invention

[0004] This invention provides a bridge deck damage detection method and system, aiming to solve the technical problems in the prior art of low sensitivity and inaccurate localization of early bridge deck damage detection, as well as low efficiency of multi-point analysis and calculation and failure to effectively utilize the spatial correlation features of damage.

[0005] In a first aspect, the present invention provides a method for detecting bridge deck damage, comprising:

[0006] Acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck, and process the dynamic strain response signals from each measuring point into strain time history data;

[0007] Modal decomposition is performed on the strain time history data of each measuring point to extract the principal strain mode shape of each measuring point. Based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points of a certain measuring point in the preset structural direction, the curvature value of a certain mode shape reflecting the local bending deformation characteristics of the measuring point is calculated.

[0008] Using each measuring point as a reference point, a damage collaborative analysis group is established between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from measuring points other than a certain reference point based on the distance between measuring points and structural features.

[0009] For a certain damage co-analysis group, based on the mode curvature values ​​of the certain reference point and all co-analysis points corresponding to the reference point, the correlation coefficient between the mode curvature value of the reference point in the group and the mode curvature values ​​of each co-analysis point is calculated, and it is determined whether each correlation coefficient is greater than a preset consistency threshold.

[0010] Damage co-analysis groups in which all correlation coefficients within a group are greater than the consensus threshold are marked as valid co-analysis groups;

[0011] The effective number of all effective collaborative groups is counted. Based on the ratio of the effective number to the total number of damage collaborative analysis groups, the structural damage consistency index is calculated, and it is determined whether the structural damage consistency index is less than a preset damage judgment threshold.

[0012] If the structural damage consistency index is not less than the damage judgment threshold, it is determined that there is structural damage to the bridge deck, and the center range of the damage area is determined based on the abrupt change location of the mode curvature distribution of all effective coordinating groups.

[0013] In a second aspect, the present invention provides a bridge deck damage detection system, comprising:

[0014] The acquisition module is configured to acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck and process the dynamic strain response signals from each measuring point into strain time history data.

[0015] The calculation module is configured to perform modal decomposition on the strain time history data of each measuring point, extract the principal strain mode shape of each measuring point, and calculate the curvature value of a mode shape that reflects the local bending deformation characteristics of the measuring point based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points of a certain measuring point in a preset structural direction.

[0016] The construction module is configured to take each measuring point as a reference point and establish a damage collaborative analysis group between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from the measuring points other than the certain reference point based on the distance between the measuring points and the structural features.

[0017] The first judgment module is configured to, for a certain damage co-analysis group, calculate the correlation coefficient between the mode shape curvature value of the certain reference point and the mode shape curvature value of each co-analysis point within the group based on the mode shape curvature value of the certain reference point and all co-analysis points corresponding to the certain reference point, and determine whether each correlation coefficient is greater than a preset consistency threshold.

[0018] The marking module is configured to mark damage co-analysis groups in which all correlation coefficients within a group are greater than the consistency threshold as valid co-analysis groups.

[0019] The second judgment module is configured to count the effective number of all effective collaborative groups, calculate the structural damage consistency index based on the ratio of the effective number to the total number of damage collaborative analysis groups, and determine whether the structural damage consistency index is less than a preset damage judgment threshold.

[0020] The determination module is configured to determine that the bridge deck has structural damage if the structural damage consistency index is not less than the damage determination threshold, and to determine the center range of the damage area based on the abrupt change location of the mode curvature distribution of all effective coordinating groups.

[0021] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the bridge deck damage detection method according to any embodiment of the present invention.

[0022] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of the bridge deck damage detection method of any embodiment of the present invention.

[0023] The bridge deck damage detection method and system of this application calculates the mode curvature value representing local stiffness and intelligently constructs an analysis group reflecting the spatial continuity of the structure. Focusing on the core feature of "disruption of local cooperative relationships" that inevitably occurs when damage occurs, it examines the consistency of the mode curvature change patterns at each point within the group and statistically analyzes the proportion of "effective cooperative groups" where this consistency is maintained in the overall structure. This allows it to extremely sensitively capture abnormal patterns that spread spatially caused by local damage. This method significantly reduces the computational burden of independently analyzing massive amounts of data and improves detection efficiency. More importantly, by mining the spatial correlation characteristics of damage, it enhances the ability to identify and locate early and minor damage, achieving a leap from "point analysis" to "correlation network analysis." This provides an efficient, sensitive, and reliable intelligent detection method for the structural health monitoring of bridge decks. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart of a bridge deck damage detection method provided in an embodiment of the present invention;

[0026] Figure 2 This is a structural block diagram of a bridge deck damage detection system provided in an embodiment of the present invention;

[0027] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Please see Figure 1 The diagram shows a flowchart of a bridge deck damage detection method according to this application.

[0030] like Figure 1 As shown, the bridge deck damage detection method specifically includes the following steps;

[0031] Step S101: Obtain the dynamic strain response signals of at least two measuring points deployed on the bridge deck, and process the dynamic strain response signals of each measuring point into strain time history data.

[0032] In this step, the dynamic strain response signal is received and bandpass filtered to remove high-frequency noise and low-frequency drift, resulting in a first filtered signal; the first filtered signal is then subjected to wavelet threshold denoising to obtain a second filtered signal; the second filtered signal is resampled according to a preset sampling frequency and arranged in chronological order to obtain strain time history data corresponding to each measurement point.

[0033] In one specific embodiment, the goal of data processing is to extract clean strain time histories that reflect the true dynamic behavior of the structure from the raw, noisy sensor signals, laying a high-quality data foundation for subsequent modal analysis and feature extraction. The specific implementation includes three progressive sub-steps:

[0034] First, bandpass filtering is performed. The system receives the raw dynamic strain response signal x_raw(t) from sensors (such as fiber Bragg grating strain sensors) at various measuring points. This signal typically contains high-frequency electronic noise, electromagnetic interference, and low-frequency drift caused by factors such as temperature changes. A bandpass frequency range [f_low, f_high] (e.g., 0.5 Hz to 50 Hz) matching the dynamic characteristics of the bridge deck structure is set to cover the dominant frequency component caused by vehicle excitation. x_raw(t) is filtered using a digital filter (such as a Butterworth filter) to effectively suppress noise and drift outside the frequency band, resulting in the first filtered signal x_bp(t). This step initially purifies the signal, preserving the structural response within the frequency band of interest.

[0035] Secondly, wavelet thresholding denoising is performed. Despite bandpass filtering, random noise still exists in the passband of x_bp(t). Discrete wavelet transform is used to decompose x_bp(t) at multiple scales, obtaining wavelet coefficients at different scales. Among these coefficients, the large coefficients representing the true structural response have larger amplitudes, while the small coefficients representing noise have smaller amplitudes and are widely distributed. By presetting a threshold (such as using a general threshold or an adaptive threshold based on signal characteristics), wavelet coefficients smaller than the threshold at each scale are set to zero or shrunk, and then wavelet reconstruction is performed using the processed coefficients. This process yields the second filtered signal x_denoised(t), which significantly reduces random noise while better preserving the local features and abrupt changes of the signal, which is crucial for the subsequent accurate extraction of modal parameters.

[0036] Finally, resampling and timing alignment are performed. Since the initial sampling frequency of the sensor may be too high (leading to data redundancy) or uneven, and there may be slight clock deviations between channels, x_denoised(t) needs to be standardized. A suitable preset sampling frequency f_s that satisfies the Nyquist sampling theorem is selected. Using anti-aliasing filtering and interpolation methods, x_denoised(t) at each measurement point is resampled to a uniform f_s. Subsequently, based on a unified timestamp or trigger signal, the resampled signals from all measurement points are precisely aligned and arranged on the time axis, ultimately generating a time-continuous, formatted strain time history for each measurement point. This data constitutes the direct input for all subsequent analysis work.

[0037] Step S102: Modal decomposition is performed on the strain time history data of each measuring point to extract the principal strain mode shape of each measuring point. Based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points in the preset structural direction, the curvature value of a certain mode shape reflecting the local bending deformation characteristics of the measuring point is calculated.

[0038] In this step, strain time history data at a certain measuring point is used as the input signal; the input signal is processed using the empirical mode decomposition method to obtain multiple intrinsic mode functions (EMFs); the energy proportion of each EMF is calculated, and the EMF with the largest energy proportion is determined as the principal EMF; Hilbert transform is performed on the principal EMF to obtain the corresponding analytical signal, and the instantaneous amplitude and instantaneous phase are extracted from the analytical signal; based on the change of the instantaneous phase over time, the dominant vibration period in the signal is identified, and the principal strain mode shape corresponding to the dominant vibration period is reconstructed.

[0039] Furthermore, a local analysis set is defined for a certain measuring point. The local analysis set includes the measuring point and at least two measuring points that are directly adjacent to the measuring point along one of the two preset structural directions, either longitudinal or transverse, of the bridge deck.

[0040] Obtain the principal strain mode shape values ​​of all measuring points in the local analysis set;

[0041] The measurement points in the local analysis set are arranged in spatial order, and the principal strain mode values ​​of all measurement points constitute a discrete mode sequence.

[0042] Using the central difference method, the mode curvature value at a measuring point located in the middle of the mode sequence is calculated. The calculation formula is as follows:

[0043] ,

[0044] Where κ is the mode curvature value at a certain measuring point. The principal strain mode shape value at a certain measuring point. and Δd represents the principal strain mode vibration values ​​of the two adjacent measuring points along the preset structural direction of a certain measuring point, and Δd is the known spatial distance between the adjacent measuring points.

[0045] In one specific embodiment, strain time history data s(t) at a certain measuring point is used as the input signal. First, it is processed using Empirical Mode Decomposition (EMD), an adaptive time-frequency analysis method. Based on the local time-scale characteristics of the signal itself, EMD decomposes s(t) into a series of intrinsic mode functions (IMFs) arranged from high frequency to low frequency and a residual term representing the trend through an iterative sieving process. Each IMF component is a narrowband signal, representing a vibration mode at a specific time scale in the signal.

[0046] Next, the energy of each IMF component is calculated as E_i=∫c_i(t)²dt, where c_i(t) is the i-th intrinsic mode function. The IMF component with the largest energy proportion usually contains the most significant vibration response information of the structure excited by the main excitation source such as the vehicle, and is therefore determined as the Dominant IMF, denoted as c_d(t).

[0047] Subsequently, a Hilbert transform is applied to c_d(t) to construct its analytic signal z(t) = c_d(t) + j·H[c_d(t)], where H[·] represents the Hilbert transform. The instantaneous amplitude A(t) and instantaneous phase θ(t) can be accurately extracted from the analytic signal. The rate of change of the instantaneous phase θ(t) with time is the instantaneous frequency. By analyzing the evolution of θ(t), the dominant and stable vibration period T_dominant in the signal can be identified. Finally, using this dominant period as a reference, c_d(t) is periodically segmented and averaged to eliminate random fluctuations, reconstructing the smooth and normalized principal strain mode shape value φ corresponding to the measuring point. This value φ is essentially the steady-state strain response amplitude of the measuring point under the dominant vibration mode of the structure relative to a standard excitation, and is a concentrated manifestation of the local dynamic characteristics of the structure.

[0048] Calculation of mode curvature value:

[0049] After obtaining the principal strain mode shapes at all measuring points, it is necessary to further calculate the physical quantity that more sensitively reflects changes in local bending stiffness—the mode curvature. For a target measuring point... Define a local analysis set for it. This set contains It itself, and along one of the two preset structural directions of the bridge deck—longitudinal (driving direction) or transverse (perpendicular to the driving direction)—with... At least one measuring point directly adjacent to each other before and after (e.g.) and Obtain the principal strain mode shape values ​​for all measuring points within this set. , , .

[0050] Arrange these measuring points in order of their spatial location (e.g.) , , Arranged in a symmetrical pattern, the mode shapes constitute a discrete sequence of mode shapes distributed along the structural direction. The curvature of the structure is mathematically defined as the second derivative of the mode shape with respect to spatial coordinates. Here, the central difference method is used to numerically differentiate this discrete sequence to approximate the target measurement point located in the middle of the sequence with high accuracy. The modal curvature value κ is calculated using the following formula:

[0051] ,

[0052] The calculated κ is a scalar with a clear physical meaning. Its absolute value directly reflects the severity of local bending deformation at that point under the dominant vibration mode. For a uniform and undamaged beam-slab structure, the curvature value should change gradually along the span. However, when damage occurs at a point, causing a decrease in the local bending stiffness EI, according to the beam bending theory (curvature κ = M / (EI), where M is the bending moment), under the same internal force M, the curvature value κ at that point will increase abnormally, thus becoming an extremely sensitive indicator of damage.

[0053] In summary, the modal curvature value κ is calculated based on the central difference method. This operation transforms the single-point modal amplitude φ into a spatial differential characteristic that directly and quantitatively reflects the local bending stiffness. Theoretical analysis and numerous experiments have shown that the most direct manifestation of structural damage (such as cracking and material degradation) is a decrease in local stiffness, and the modal curvature value κ is extremely sensitive to changes in stiffness. By constructing a "local analysis set" and applying the central difference formula, this method cleverly utilizes a sparsely deployed sensor network to achieve high-precision discretization estimation of the curvature field of the structural continuum. The resulting modal curvature value κ is no longer an isolated response amplitude, but a core damage indicator containing rich information about the local structural state. This provides an irreplaceable and physically clear quantitative foundation for subsequent damage detection and localization by analyzing its spatial correlation patterns, and is the core of this method's ability to achieve high damage detection sensitivity and accuracy.

[0054] Step S103: Using each measuring point as a reference point, a damage collaborative analysis group is established between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from the measuring points other than the certain reference point based on the distance between the measuring points and the structural features.

[0055] In this step, obtain the spatial position coordinates of all measurement points; calculate the Euclidean distances between a certain reference point and all other measurement points, where the other measurement points are any measurement point except a certain reference point among all measurement points; screen out the measurement points whose distances from a certain reference point are less than the spatial neighborhood radius threshold as spatial proximity candidate points; combine with the CAD design drawings of the bridge deck to identify the structural construction partitions of the bridge deck, and the structural construction partitions include main girder units, diaphragm units, and continuous bridge deck blocks; screen out the measurement points that are located in the same structural construction partition as the certain reference point from the spatial proximity candidate points; define the finally screened set of measurement points as the collaborative analysis point set of a certain reference point, thereby forming a certain damage collaborative analysis group.

[0056] In a specific embodiment, first, obtain the spatial information of the measurement points. The system reads the accurate spatial position coordinates of all measurement points, which are usually obtained by total station or GPS measurement during the sensor layout stage and stored in the database. The coordinate form is three-dimensional Cartesian coordinates (x, y, z). For the plane analysis of the bridge deck, the horizontal projection coordinates (x, y) are mainly used.

[0057] Second, calculate spatial proximity. Select the current reference point (with coordinates (x_i, y_i)), and calculate its Euclidean distance d_ij = sqrt((x_i - x_j) +(y_i - y_j) 2 ) with all other measurement points 2 (j≠i) in the database. Set a key parameter - the spatial neighborhood radius threshold R. This threshold is determined based on structural dynamics principles and engineering experience, usually taking a part of the wavelength corresponding to the fundamental frequency of the structure, or set according to the sensor layout density (such as 2 - 3 times the average measurement point spacing), for example, R = 5 meters. Screen out all measurement points that satisfy d_ij < R to form a preliminary "spatial proximity candidate point" set C_i. This step is based on the "localized damage effect" hypothesis to ensure that the analysis focuses on the area around the reference point that may be affected by the same damage.

[0058] Third, incorporate structural construction information, call or associate the CAD design drawings, BIM models or construction partition records of the bridge deck. Through coordinate mapping, automatically identify or manually predefine the structural construction partitions of the bridge deck. Typical partitions include: main girder units (such as the top plate areas corresponding to the No. 1 T-beam and the No. 2 box girder web), diaphragm units (the diaphragm and the directly connected bridge deck area), and continuous bridge deck blocks (the continuously poured areas separated by expansion joints or construction joints). The system obtains the structural construction partition ID to which each measurement point belongs.

[0059] Fourth, screen continuous structural measurement points. For each measurement point in the spatially adjacent candidate point set C_i obtained in the previous step, the system determines whether its structural construction partition ID is the same as the ID of the reference point. Only those measurement points located within the same structural construction partition as the reference point are retained. For example, if the reference point is located in "Main Beam Element No. 3", only adjacent measurement points also located in that main beam element are retained. This step is crucial because it ensures that the measurement points within the analysis group are not only spatially close but also within the same stress-continuous "structural subsystem". This avoids forcibly associating measurement points belonging to different, independent stress-bearing components (such as two independent main beams), thus guaranteeing the physical premise of highly coordinated dynamic responses of measurement points within the group under healthy conditions.

[0060] Finally, a damage co-analysis group is formed. The set of measurement points retained after the above dual screening is formally defined as the current benchmark point. Collaborative analysis point set Thus, a complete damage co-analysis group G_i is constructed, denoted as G_i= ∪ This process is repeated for each measuring point on the bridge deck, eventually forming multiple damage co-analysis groups equal to the number of measuring points. These groups together constitute an analysis network covering the entire monitoring area.

[0061] In summary, by strictly limiting the collaborative analysis to the same structural unit (such as a single main beam), this method ensures that, under healthy conditions, all measuring points within the group, due to being under the same load transfer path and boundary conditions, should exhibit extremely high inherent consistency in their dynamic responses (especially the modal curvature characterizing local stiffness). When damage occurs at some point within this unit, this consistency will be locally disrupted. Therefore, by monitoring whether this pre-set high consistency is broken, damage can be detected with extreme sensitivity and reliability, much like an "immune system recognizing its own abnormalities," effectively avoiding misjudgments caused by normal response differences between different structural units.

[0062] The constructed damage synergy analysis groups essentially define "local structural health communities." When a group is determined to be "synergistically failed" (i.e., an invalid group), its clear physical boundaries (specific main beam or bridge deck block) immediately provide strong prior constraints for damage localization, rapidly narrowing the suspected area from the entire bridge deck to a specific structural sub-region. This provides crucial spatial context and search focus for subsequent precise localization steps, greatly improving localization efficiency and accuracy.

[0063] This method defines its analysis logic directly based on actual bridge design drawings and construction zones, rather than relying on abstract mathematical clustering. This allows it to naturally adapt to various complex bridge deck structures (T-beams, small box girders, hollow slabs, etc.). Simultaneously, through structural zoning filtering, it can automatically eliminate normal response variations caused by structural discontinuities (such as expansion joints and hinge joints) or differences in stiffness between different components. This enhances the method's robustness in real-world complex environments and reduces false alarms caused by normal structural non-uniformity.

[0064] This lays the foundation for multi-scale damage assessment. By establishing this physically based local analysis group network, the system can assess damage at two levels: each structural unit (microscale) and the entire bridge deck (macroscale). It can diagnose independent damage from the synergistic changes of a specific main girder, and assess the overall health of the bridge deck by statistically analyzing the proportion of effective synergistic groups across all main girder units, thus achieving multi-scale, hierarchical damage status assessment from local to global perspectives.

[0065] Step S104: For a certain damage co-analysis group, based on the mode curvature values ​​of the certain reference point and all co-analysis points corresponding to the certain reference point, calculate the correlation coefficient between the mode curvature value of the certain reference point in the group and the mode curvature values ​​of each co-analysis point, and determine whether each correlation coefficient is greater than a preset consistency threshold.

[0066] In this step, for a pre-constructed damage co-analysis group G_k={P_b, P_{c1},P_{c2},..., P_{cm}}, where P_b is the reference point and P_{c1}...P_{cm} are m co-analysis points, assuming that the mode curvature values ​​of all points have been calculated through the aforementioned steps, let the curvature value of the reference point be κ_b, and the curvature values ​​of each co-analysis point be κ_{c1},κ_{c2},...,κ_{cm}.

[0067] Correlation coefficient calculation: For each co-analysis point P_{ci} (i from 1 to m) within the group, calculate the Pearson correlation coefficient r_i between its mode shape curvature value κ_{ci} and the reference point mode shape curvature value κ_b. The Pearson correlation coefficient is used to measure the degree of linear correlation between two variables, and its calculation formula is as follows:

[0068] r_i=Cov(κ_b,κ_{ci}) / (σ_{κ_b}*σ_{κ_{ci}}),

[0069] Where Cov represents covariance and σ represents standard deviation. In practical calculations, the correlation coefficient is usually estimated based on a sufficient number of sample data collected under different working conditions or at different time points. For example, using multiple sets of data collected at the same time each day over the past week, the observations of κ_b and κ_{ci} on different days are combined into two sample sequences, and then the correlation coefficient r_i of these two sequences is calculated. The value of r_i ranges from [-1, 1]. Under healthy conditions, since the force and deformation of the same set of measuring points are highly coordinated, their curvature values ​​should rise and fall together, and r_i should be very close to 1.

[0070] Consistency Threshold Determination: A preset consistency threshold T_cons is set. This threshold is a key parameter, determined based on statistical analysis of long-term monitoring data of healthy structures. For example, historical correlation coefficient data within all similar synergistic analysis groups can be calculated during the long-term monitoring period after bridge completion and confirmation of no damage. A high percentile of this distribution (such as the 95th percentile or the mean minus two standard deviations) can be used as T_cons, for example, T_cons = 0.85. This threshold represents the minimum acceptable level of synergy within normal fluctuations.

[0071] Subsequently, the m correlation coefficients r_1, r_2, ..., r_m calculated within group G_k are judged one by one: whether r_i > T_cons is true. Only when all m correlation coefficients within the group meet the condition of being greater than T_cons is the damage synergy analysis group considered to have maintained good synergy consistency in the current state.

[0072] In one specific embodiment, if any correlation coefficient is not greater than a preset consistency threshold, the damage synergy analysis group is not marked as a valid synergy group.

[0073] Step S105: Mark the damage co-analysis group where all correlation coefficients within the group are greater than the consistency threshold as a valid co-analysis group.

[0074] Step S106: Count the effective number of all effective collaborative groups, calculate the structural damage consistency index based on the ratio of the effective number to the total number of damage collaborative analysis groups, and determine whether the structural damage consistency index is less than the preset damage judgment threshold.

[0075] In this step, the total number of damage synergy analysis groups N_total and the number of valid synergy groups marked as valid are obtained;

[0076] The preliminary consistency ratio R_0 = N_valid / N_total is calculated based on the total number of damage co-analysis groups and the number of valid groups.

[0077] The average of the correlation coefficients calculated within all valid synergy groups is defined as the overall synergy degree S_mean;

[0078] Based on the overall synergy S_mean and the preset reference synergy S_ref, the structural damage consistency index CI is calculated, and its expression is:

[0079] CI=R_0×[1+λ×(S_mean-S_ref)],

[0080] Wherein, λ is the preset sensitivity coefficient.

[0081] Furthermore, if the structural damage consistency index is less than a preset damage judgment threshold, a status report indicating that the current bridge deck is in a healthy or normal state is directly generated.

[0082] In one specific embodiment, all damage co-analysis groups established on the entire bridge deck structure are traversed and counted. Assuming there are M measurement points, theoretically, M damage co-analysis groups can be established with each measurement point as a reference point (some boundary measurement points may not be able to construct valid groups, and their N_valid is counted as 0). The total number of damage co-analysis groups, N_total, is obtained. Simultaneously, from the results of the previous step (step S105), the number of groups marked as "valid co-analysis groups," i.e., the number of valid groups, N_valid, is counted. An "valid co-analysis group" refers to a group in which all correlation coefficients are greater than a preset consistency threshold T_cons.

[0083] Secondly, the preliminary consistency ratio is calculated. Based on the two statistics mentioned above, the preliminary consistency ratio R_0 is calculated. R_0 directly reflects the proportion of units that still maintain good synergy among all the pre-defined analysis units in the entire structure. When the structure is perfectly healthy, R_0 should be close to 1; as damage occurs and spreads, more and more local synergies are broken, leading to a decrease in N_valid and a drop in R_0.

[0084] To more precisely measure the "quality" of synergy rather than just its "presence," the average correlation coefficients within all valid synergy groups are calculated. Specifically, all calculated correlation coefficient values ​​within each valid synergy group (each valid group has m values, where m is the number of synergy analysis points in that group) are collected, and the arithmetic mean of these values ​​is calculated, defined as the overall synergy score S_mean. S_mean reflects the tightness of synergy within those local regions that are still considered "valid." Even if a group is not judged as invalid, its correlation coefficient may decline from its historical high, and S_mean can capture this subtle quality decay.

[0085] Call the pre-stored reference synergy degree S_ref. S_ref is a statistical value (such as the mean or a specific quantile) of S_mean calculated based on a large amount of historical monitoring data during the recognized healthy operation period of the bridge (such as the load test stage after completion or the stage where long-term monitoring confirms no abnormalities). At the same time, set a sensitivity coefficient λ (λ > 0), which is used to adjust the influence weight of the deviation of S_mean relative to S_ref on the final index, and can be determined through historical data regression analysis or engineering experience according to the sensitivity requirements of the structure to damage.

[0086] Finally, calculate the structural damage consistency index CI using the following formula:

[0087] CI = R_0 × [1 + λ × (S_mean - S_ref)]

[0088] This formula combines the "breadth" (R_0) and "depth" (S_mean) of synergy. When S_mean is higher than S_ref, it has a positive gain on CI; when S_mean is lower than S_ref, even if R_0 remains unchanged temporarily, CI will be lowered due to the quality decline, which makes the index more sensitive to early and minor synergy degradation.

[0089] Compare the calculated CI value with a preset damage judgment threshold T_damage. T_damage is a critical value determined according to safety requirements, historical data statistical analysis, or model simulation. The judgment logic is: if CI < T_damage, it is determined that there is structural damage to the bridge deck; otherwise, it is determined that the overall synergy state of the structure is normal or within an acceptable range. This judgment result provides a decision basis for whether to trigger the damage location and warning process subsequently.

[0090] In step S107, if the structural damage consistency index is not less than the damage judgment threshold, it is determined that there is structural damage to the bridge deck, and the central range of the damage area is determined based on the mutation positions of the modal curvature distributions of all effective synergy groups.

[0091] In this step, the modal curvature values ​​and spatial coordinates of all measuring points in each effective coordination group are extracted; the modal curvature values ​​of all measuring points extracted from all effective coordination groups are correlated with the spatial coordinates of all measuring points on the two-dimensional plane of the bridge deck; on the two-dimensional plane, a spatial interpolation method is used to generate a spatial distribution cloud map reflecting the modal curvature values ​​of the entire bridge deck; in the spatial distribution cloud map, regions with modal curvature values ​​higher than a preset high threshold are identified; the identified regions are merged and their boundaries are optimized to determine one or more suspected damage regions; for each suspected damage region, the mean coordinate value of all measuring points within the suspected damage region is calculated, and a minimum circumscribed circle is fitted based on the region shape with the mean coordinate value as the center, which is taken as the center range of the damage region.

[0092] In one specific embodiment, firstly, the feature data of effective coordination groups are extracted. The system iterates through all groups G_valid that are marked as "effective coordination groups". For each effective coordination group, two key features of all measurement points (including the reference point and all coordination analysis points) are extracted: 1) the calculated modal curvature value κ; 2) the two-dimensional spatial coordinates (x, y) of the measurement point on the bridge deck plane. The data extracted from all effective coordination groups are merged to form a dataset D={(x_i,y_i,κ_i)}, where i is the index of all measurement points belonging to at least one effective coordination group. This dataset represents the curvature space distribution of the measurement point group considered to have good coordination from the current structural state.

[0093] Secondly, a spatial distribution cloud map of modal curvature is generated. The coordinates in dataset D are correlated with curvature values, and then processed using spatial interpolation methods (e.g., Kriging interpolation or inverse distance weighted interpolation) on the two-dimensional plane projection of the bridge deck. Based on the distribution of discrete measurement points (x_i, y_i, κ_i), the interpolation process estimates the modal curvature value κ_est(x, y) at any other location (x, y) on the plane, thus generating a continuous spatial distribution cloud map of modal curvature covering the entire bridge deck or target area. This cloud map visually displays the spatial variation of curvature values ​​using color depth or contour lines, where high-value areas (usually displayed in warm colors) indicate potential local stiffness degradation.

[0094] On the generated distribution cloud map, a preset high-value threshold κ_th ​​is set. This threshold can be determined based on the statistical distribution of curvature values ​​in historical health data (such as the mean plus several times the standard deviation), or based on engineering experience. Through image processing or numerical analysis algorithms, all continuous pixel regions where κ_est(x, y) > κ_th ​​are automatically identified in the cloud map. These regions are the preliminary suspected damage areas marked due to abnormally increased curvature.

[0095] Due to measurement errors, interpolation smoothing, or the complexity of the damage itself, the initially identified areas may present as multiple adjacent small patches or irregular boundaries. Morphological operations (such as dilation and erosion) and spatial clustering algorithms are applied to merge multiple small regions with a spatial distance less than a preset merging distance, and the boundaries of the regions are smoothed and optimized to form one or more clearly defined and connected final suspected damage regions A_1, A_2, ..., A_p.

[0096] Finally, the center range of the damage area is determined. For each final suspected damage area A_k, the spatial coordinate set of all original measurement points falling within the optimized boundary is obtained. The mean coordinate of all points in this set is calculated. This point is the geometric center of the suspected damaged area. Subsequently, using this geometric center... Starting from a point, calculate the smallest circumcircle that completely encompasses all measuring points within the region. The center of this circle is the final determined center point of the damaged area, and its radius r_k defines the core range of the damage's impact. The output is a series of center coordinates and radii {(Center_k, r_k)}, providing clear spatial positioning guidance for subsequent precise verification and repair.

[0097] In summary, by generating the background field using spatial interpolation based on "effective coordination group" data, this method essentially uses regions where the structure remains normal under its current state as a reference benchmark. The "abnormally high value" regions identified on this benchmark have clear physical meaning—they are local abrupt change zones relative to the overall "normal" deformation mode of the current structure. This approach, compared to directly comparing all original data or fixed historical baselines, is better able to eliminate systematic offset interference caused by overall environmental changes (such as temperature) or load variations, resulting in more stable and reliable positioning results.

[0098] Through a processing chain of "identification-merging-fitting the minimum circumcircle," the method can effectively handle various situations, including single concentrated damage, multiple scattered damage, and irregularly shaped damage. The final output "center range" (center and radius) is a highly simplified and engineered expression that provides both the core location of the damage (center) and quantifies the approximate scale of its spatial impact (radius). This format is very easy for field engineers to understand and conduct subsequent detailed inspections (such as radar scanning and percussion inspection), achieving a seamless integration of automated detection and manual verification.

[0099] The method presented in this application, by monitoring the consistency of the mode curvature variation patterns at various measuring points within a local region defined by both spatial proximity and structural homology, can extremely sensitively capture the disruption of local mechanical correlation caused by early damage, thus achieving highly sensitive early warning of minor, early-stage damage to bridge decks. Secondly, a fully automated intelligent analysis process is constructed, from high-quality signal preprocessing to the extraction of sensitive features (mode curvature) based on empirical mode decomposition and central difference methods, and then to collaborative quantitative assessment based on statistical correlation analysis, ultimately fusing together to generate macroscopic damage indicators and precise spatial positioning. This process significantly reduces the computational resource consumption of independently and complexly modeling massive numbers of measuring points, greatly improving the analysis efficiency of large-scale monitoring while ensuring high accuracy. Furthermore, this scheme has a solid physical foundation and strong engineering adaptability. The core characteristic, mode curvature, is directly related to local flexural stiffness, with a clear physical meaning. The analysis logic closely aligns with the actual bridge structure (e.g., main girder and bridge deck zoning), and utilizes the bridge's own health history data to establish a personalized assessment baseline. This effectively overcomes environmental interference and individual structural differences, significantly improving the method's anti-interference capability and reliability, and greatly reducing false alarms. Finally, the method's output is intuitive and practical, providing not only a clear determination of damage existence but also damage area location information expressed as "center coordinates and range." This achieves a seamless connection from intelligent detection to precise maintenance guidance, providing efficient, reliable, and easily implementable core technical support for preventative maintenance and intelligent management of bridges.

[0100] Please see Figure 2 The diagram shows a structural block diagram of a bridge deck damage detection system according to this application.

[0101] like Figure 2 As shown, the bridge deck damage detection system 200 includes an acquisition module 210, a calculation module 220, a construction module 230, a first judgment module 240, a marking module 250, a second judgment module 260, and a determination module 270.

[0102] The acquisition module 210 is configured to acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck and process the dynamic strain response signals from each measuring point into strain time history data. The calculation module 220 is configured to perform modal decomposition on the strain time history data of each measuring point, extract the principal strain mode shapes of each measuring point, and calculate the curvature value of a mode shape reflecting the local bending deformation characteristics of the measuring point based on the principal strain mode shape of a certain measuring point and the principal strain mode shapes of adjacent measuring points of a certain measuring point in a preset structural direction. The construction module 230 is configured to establish a damage co-analysis group between a certain measuring point and at least one co-analysis point corresponding to the certain measuring point, wherein the co-analysis point is selected from measuring points other than the certain measuring point according to the distance between measuring points and the structural features. The first judgment module 240 is configured to, for a certain damage co-analysis group, based on the certain... The system uses a reference point and the modal curvature values ​​of all corresponding collaborative analysis points. It calculates the correlation coefficient between the modal curvature values ​​of the reference point and each collaborative analysis point within the group, and determines whether each correlation coefficient is greater than a preset consistency threshold. A marking module 250 is configured to mark damage collaborative analysis groups where all correlation coefficients are greater than the consistency threshold as valid collaborative groups. A second judgment module 260 is configured to count the number of valid collaborative groups, calculate a structural damage consistency index based on the ratio of the number of valid collaborative groups to the total number of damage collaborative analysis groups, and determine whether the structural damage consistency index is less than a preset damage judgment threshold. A determination module 270 is configured to determine that the bridge deck has structural damage if the structural damage consistency index is not less than the damage judgment threshold, and determine the center range of the damage area based on the abrupt change locations of the modal curvature distribution of all valid collaborative groups.

[0103] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0104] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the bridge deck damage detection method in any of the above method embodiments.

[0105] In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows:

[0106] Acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck, and process the dynamic strain response signals from each measuring point into strain time history data;

[0107] Modal decomposition is performed on the strain time history data of each measuring point to extract the principal strain mode shape of each measuring point. Based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points of a certain measuring point in the preset structural direction, the curvature value of a certain mode shape reflecting the local bending deformation characteristics of the measuring point is calculated.

[0108] Using each measuring point as a reference point, a damage collaborative analysis group is established between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from measuring points other than a certain reference point based on the distance between measuring points and structural features.

[0109] For a certain damage co-analysis group, based on the mode curvature values ​​of the certain reference point and all co-analysis points corresponding to the reference point, the correlation coefficient between the mode curvature value of the reference point in the group and the mode curvature values ​​of each co-analysis point is calculated, and it is determined whether each correlation coefficient is greater than a preset consistency threshold.

[0110] Damage co-analysis groups in which all correlation coefficients within a group are greater than the consensus threshold are marked as valid co-analysis groups;

[0111] The effective number of all effective collaborative groups is counted. Based on the ratio of the effective number to the total number of damage collaborative analysis groups, the structural damage consistency index is calculated, and it is determined whether the structural damage consistency index is less than a preset damage judgment threshold.

[0112] If the structural damage consistency index is not less than the damage judgment threshold, it is determined that there is structural damage to the bridge deck, and the center range of the damage area is determined based on the abrupt change location of the mode curvature distribution of all effective coordinating groups.

[0113] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the bridge deck damage detection system, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, and these remote memories may be connected to the bridge deck damage detection system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0114] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the bridge deck damage detection method described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the bridge deck damage detection system. The output device 340 may include a display screen or other display device.

[0115] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0116] In one implementation, the above-described electronic device is applied to a bridge deck damage detection system for a client, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0117] Acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck, and process the dynamic strain response signals from each measuring point into strain time history data;

[0118] Modal decomposition is performed on the strain time history data of each measuring point to extract the principal strain mode shape of each measuring point. Based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points of a certain measuring point in the preset structural direction, the curvature value of a certain mode shape reflecting the local bending deformation characteristics of the measuring point is calculated.

[0119] Using each measuring point as a reference point, a damage collaborative analysis group is established between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from measuring points other than a certain reference point based on the distance between measuring points and structural features.

[0120] For a certain damage co-analysis group, based on the mode curvature values ​​of the certain reference point and all co-analysis points corresponding to the reference point, the correlation coefficient between the mode curvature value of the reference point in the group and the mode curvature values ​​of each co-analysis point is calculated, and it is determined whether each correlation coefficient is greater than a preset consistency threshold.

[0121] Damage co-analysis groups in which all correlation coefficients within a group are greater than the consensus threshold are marked as valid co-analysis groups;

[0122] The effective number of all effective collaborative groups is counted. Based on the ratio of the effective number to the total number of damage collaborative analysis groups, the structural damage consistency index is calculated, and it is determined whether the structural damage consistency index is less than a preset damage judgment threshold.

[0123] If the structural damage consistency index is not less than the damage judgment threshold, it is determined that there is structural damage to the bridge deck, and the center range of the damage area is determined based on the abrupt change location of the mode curvature distribution of all effective coordinating groups.

[0124] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, 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 can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting bridge deck damage, characterized in that, include: Acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck, and process the dynamic strain response signals from each measuring point into strain time history data; Modal decomposition is performed on the strain time history data of each measuring point to extract the principal strain mode shape of each measuring point. Based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points of a certain measuring point in the preset structural direction, the curvature value of a certain mode shape reflecting the local bending deformation characteristics of the measuring point is calculated. Using each measuring point as a reference point, a damage collaborative analysis group is established between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from measuring points other than a certain reference point based on the distance between measuring points and structural features. For a certain damage co-analysis group, based on the mode curvature values ​​of the certain reference point and all co-analysis points corresponding to the reference point, the correlation coefficient between the mode curvature value of the reference point in the group and the mode curvature values ​​of each co-analysis point is calculated, and it is determined whether each correlation coefficient is greater than a preset consistency threshold. Damage co-analysis groups in which all correlation coefficients within a group are greater than the consensus threshold are marked as valid co-analysis groups; The effective number of all effective collaborative groups is counted. Based on the ratio of the effective number to the total number of damage collaborative analysis groups, the structural damage consistency index is calculated, and it is determined whether the structural damage consistency index is less than a preset damage judgment threshold. If the structural damage consistency index is not less than the damage judgment threshold, it is determined that there is structural damage to the bridge deck, and the center range of the damage area is determined based on the abrupt change location of the mode curvature distribution of all effective coordinating groups.

2. The bridge deck damage detection method according to claim 1, characterized in that, The process of converting the dynamic strain response signals of each measuring point into strain time history data includes: The dynamic strain response signal is received, and the dynamic strain response signal is bandpass filtered to remove high-frequency noise and low-frequency drift, to obtain a first filtered signal; The first filtered signal is subjected to wavelet threshold denoising processing to obtain the second filtered signal; The second filtered signal is resampled at a preset sampling frequency and arranged in chronological order to obtain strain time history data corresponding to each measuring point.

3. The bridge deck damage detection method according to claim 1, characterized in that, The modal decomposition of strain time history data at each measuring point, and the extraction of principal strain mode shapes at each measuring point, includes: Use the strain time history data of a certain measuring point as the input signal; The input signal is processed using the empirical mode decomposition method to obtain multiple intrinsic mode functions; Calculate the energy percentage of each intrinsic mode function, and determine the intrinsic mode function with the largest energy percentage as the principal intrinsic mode function; The principal eigenmode functions are subjected to Hilbert transform to obtain the corresponding analytical signals, and the instantaneous amplitude and instantaneous phase are extracted from the analytical signals. Based on the change of the instantaneous phase over time, the dominant vibration period in the signal is identified, and the principal strain mode shape corresponding to the dominant vibration period is reconstructed.

4. The bridge deck damage detection method according to claim 1, characterized in that, The curvature value of a mode shape reflecting the local bending deformation characteristics of a measuring point, calculated based on the principal strain mode shape of a certain measuring point and the principal strain mode shapes of adjacent measuring points in a preset structural direction, includes: Define a local analysis set for a certain measuring point. The local analysis set includes the measuring point and at least two measuring points that are directly adjacent to the measuring point along one of the two preset structural directions, either longitudinal or transverse, of the bridge deck. Obtain the principal strain mode shape values ​​of all measurement points in the local analysis set; The measurement points in the local analysis set are arranged in spatial order, and the principal strain mode values ​​of all measurement points constitute a discrete mode sequence. Using the central difference method, the mode curvature value at a measuring point located in the middle of the mode sequence is calculated. The calculation formula is as follows: ， Where κ is the mode curvature value at a certain measuring point. The principal strain mode shape value at a certain measuring point. and Δd represents the principal strain mode vibration values ​​of the two adjacent measuring points along the preset structural direction of a certain measuring point, and Δd is the known spatial distance between the adjacent measuring points.

5. The bridge deck damage detection method according to claim 1, characterized in that, The establishment of a damage co-analysis group between a certain reference point and at least one corresponding co-analysis point includes: Obtain the spatial coordinates of all measuring points; Calculate the Euclidean distance between a given reference point and all other measuring points, where the other measuring points are any measuring points excluding the given reference point. Points whose distance from a certain reference point is less than the spatial neighborhood radius threshold are selected as spatial neighbor candidate points; Based on the CAD design drawings of the bridge deck, the structural construction zones of the bridge deck are identified, including main beam units, diaphragm units, and continuous bridge deck blocks. From the spatially adjacent candidate points, select the measurement points that are located within the same structural partition as the reference point; The final set of measurement points is defined as the set of collaborative analysis points for a certain benchmark point, thus forming a certain damage collaborative analysis group.

6. The bridge deck damage detection method according to claim 1, characterized in that, The structural damage consistency index, calculated based on the ratio of the effective quantity to the total number of damage synergy analysis groups, includes: Obtain the total number of damage synergy analysis groups N_total, and the number of valid synergy groups marked as valid N_valid; The preliminary consistency ratio R_0 = N_valid / N_total is calculated based on the total number of damage co-analysis groups and the number of valid groups. The average of the correlation coefficients calculated within all valid synergy groups is defined as the overall synergy degree S_mean; Based on the overall synergy S_mean and the preset reference synergy S_ref, the structural damage consistency index CI is calculated, and its expression is: CI=R_0×[1+λ×(S_mean-S_ref)], Wherein, λ is the preset sensitivity coefficient.

7. The bridge deck damage detection method according to claim 1, characterized in that, The determination of the center range of the damage region based on the abrupt changes in the modal curvature distribution of all effective coordinating groups includes: Extract the mode curvature values ​​and spatial coordinates of all measuring points in each effective coordination group; The mode curvature values ​​of all measuring points extracted from all effective coordinating groups are correlated with the spatial coordinates of all measuring points on the two-dimensional plane of the bridge deck. On the two-dimensional plane, a spatial interpolation method is used to generate a spatial distribution cloud map that reflects the modal curvature values ​​of the entire bridge deck. In the spatial distribution cloud map, regions with mode curvature values ​​higher than a preset high threshold are identified; The identified regions are merged and their boundaries are optimized to determine one or more suspected damage areas. For each suspected damage area, the mean coordinates of all measuring points within the suspected damage area are calculated, and a minimum circumscribed circle is fitted based on the shape of the area with the mean coordinates as the center, which is then used as the central range of the damage area.

8. A bridge deck damage detection system, characterized in that, include: The acquisition module is configured to acquire dynamic strain response signals from at least two measuring points deployed on the bridge deck and process the dynamic strain response signals from each measuring point into strain time history data. The calculation module is configured to perform modal decomposition on the strain time history data of each measuring point, extract the principal strain mode shape of each measuring point, and calculate the curvature value of a mode shape that reflects the local bending deformation characteristics of the measuring point based on the principal strain mode shape of a certain measuring point and the principal strain mode shape of adjacent measuring points of a certain measuring point in a preset structural direction. The construction module is configured to take each measuring point as a reference point and establish a damage collaborative analysis group between a certain reference point and at least one certain collaborative analysis point corresponding to the certain reference point. The certain collaborative analysis point is selected from the measuring points other than the certain reference point based on the distance between the measuring points and the structural features. The first judgment module is configured to, for a certain damage co-analysis group, calculate the correlation coefficient between the mode shape curvature value of the certain reference point and the mode shape curvature value of each co-analysis point within the group based on the mode shape curvature value of the certain reference point and all co-analysis points corresponding to the certain reference point, and determine whether each correlation coefficient is greater than a preset consistency threshold. The marking module is configured to mark damage co-analysis groups in which all correlation coefficients within a group are greater than the consistency threshold as valid co-analysis groups. The second judgment module is configured to count the effective number of all effective collaborative groups, calculate the structural damage consistency index based on the ratio of the effective number to the total number of damage collaborative analysis groups, and determine whether the structural damage consistency index is less than a preset damage judgment threshold. The determination module is configured to determine that the bridge deck has structural damage if the structural damage consistency index is not less than the damage determination threshold, and to determine the center range of the damage area based on the abrupt change location of the mode curvature distribution of all effective coordinating groups.

9. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1 to 7.

Images