Bonding detection data abnormity identification method and system

Abstract

The invention relates to the technical field of performance testing, and discloses a bonding detection data abnormity identification method and system.The method comprises the steps that force is applied to a wall arranged in a concrete column through a pressurizing rod, and bonding detection data of the wall is collected in real time through a strain sensor located on the concrete column; determining the stress fluctuation strength of the wall according to a stress fluctuation value in the bonding detection data; on the basis of timestamp information in the stress fluctuation value, performing fluctuation characteristic spectroscopy on the stress fluctuation intensity to construct a stress fluctuation intensity spectrum of the wall body; self-adaptive spectral domain feature merging is carried out on fluctuation intensity values in the fluctuation intensity spectrum, and aggregation fluctuation intensity values of the wall are obtained; the aggregation fluctuation strength value exceeding a preset strength threshold value is judged as bonding data abnormity, and a bonding abnormity judgment result of the wall is obtained; according to the bonding abnormity determination result, generating a bonding abnormity identification report of the wall body; according to the invention, the abnormity identification efficiency of the wall bonding detection data can be improved.

Description

Technical Field

[0001] This invention relates to the field of performance testing technology, and in particular to a method and system for identifying anomalies in adhesive testing data. Background Technology

[0002] In the field of structural safety testing for building engineering, the bonding performance between walls and concrete columns is one of the core indicators determining the overall structural stability, and the identification of anomalies in bonding test data is a crucial step in ensuring the validity of test results. With the expansion of building engineering scale and the increase in structural complexity, the requirements for the frequency of bonding test data collection and the accuracy of analysis are continuously increasing. The timeliness and accuracy of data anomaly identification are directly related to the overall effectiveness of project quality control, making the industry's demand for efficient bonding test data anomaly identification solutions increasingly urgent.

[0003] Existing technologies for identifying anomalies in bond testing data have significant performance shortcomings. Most rely on manual judgment or a single fixed threshold comparison method, which cannot accurately perceive the dynamic changes in stress fluctuations, nor can it perform multi-dimensional spectral feature mining of fluctuation data. This leads to frequent omissions and misjudgments of anomalies, seriously affecting the reliability of test results. At the same time, traditional technologies lack adaptive feature merging mechanisms, and the aggregation and analysis process for multi-frequency band fluctuation data is cumbersome and inefficient. It is difficult to quickly locate and trace the source of anomalies, and it is also difficult to efficiently generate standardized anomaly identification reports. The overall processing mode can no longer meet the timeliness and accuracy requirements of test data processing in engineering sites. Summary of the Invention

[0004] This invention provides a method and system for identifying anomalies in adhesive testing data to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for identifying anomalies in adhesion detection data, comprising: S1. Apply force to the wall placed in the concrete column through a pressure rod, and collect the bonding detection data of the wall in real time through a strain sensor located on the concrete column; S2. Determine the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data; S3. Based on the timestamp information in the stress fluctuation value, the stress fluctuation intensity is subjected to fluctuation feature spectroscopy to construct the stress fluctuation intensity spectrum of the wall. S4. Perform adaptive spectral domain feature merging on the wave intensity values ​​in the wave intensity spectrum to obtain the aggregate wave intensity value of the wall. S5. The polymerization fluctuation intensity value exceeding the preset strength threshold is determined as an abnormal bonding data, and the bonding abnormality determination result of the wall is obtained; S6. Based on the bonding anomaly determination result, generate a bonding anomaly identification report for the wall.

[0006] In a preferred embodiment, the step of applying force to the wall placed in the concrete column via a pressure rod and collecting real-time adhesion detection data of the wall via a strain sensor located on the concrete column includes: A load is applied to the wall placed in the concrete column by a pressure rod, and a strain sensor located on the concrete column is activated simultaneously to obtain the simulated strain electrical signal of the wall. The simulated strain electrical signal is interpreted to obtain the adhesion detection data of the wall.

[0007] In a preferred embodiment, determining the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data includes: The stress level center value of the stress fluctuation value sequence in the bonding test data within a preset time window is used as the representative benchmark value of the corresponding time window; Based on the representative benchmark value, identify the peak and trough values ​​of the stress fluctuation value sequence within the preset time window; The stress fluctuation intensity of the wall is obtained by evaluating the quotient between the absolute amplitude difference between the peak value and the valley value and the representative benchmark value.

[0008] In a preferred embodiment, the step of performing wave characteristic spectralization on the stress wave intensity based on the timestamp information in the stress wave value to construct the stress wave intensity spectrum of the wall includes: The stress fluctuation intensity and the timestamp information in the stress fluctuation value sequence are organized into the time-domain fluctuation signal of the wall. The time-domain fluctuation signal is mapped from the time dimension to the frequency dimension to obtain the strength information of the wall. Using the frequency in the frequency dimension as the horizontal axis and the intensity information as the vertical axis, a stress fluctuation intensity spectrum of the wall is constructed.

[0009] In a preferred embodiment, mapping the time-domain fluctuation signal from the time dimension to the frequency dimension to obtain the strength information of the wall includes: The time-domain fluctuation signal is smoothed to reduce spectral leakage, resulting in a smoothed time-domain fluctuation signal for the wall. Perform a Fourier transform on the smoothed time-domain fluctuation signal to obtain the smoothed frequency-domain fluctuation signal of the wall. Multi-dimensional parameter coupling is performed on the frequency and amplitude components of the smooth frequency domain fluctuation signal to obtain the fluctuation amplitude of the smooth frequency domain fluctuation signal, and the fluctuation amplitude is used as the strength information of the wall.

[0010] In a preferred embodiment, the step of adaptively merging spectral domain features of the wave intensity values ​​in the wave intensity spectrum to obtain the aggregated wave intensity value of the wall includes: Identify consecutive adjacent frequency points in the stress fluctuation intensity spectrum whose amplitude exceeds a preset noise benchmark; The effective fluctuation frequency band of the wall is obtained by defining the frequency band interval of the consecutive adjacent frequency points. Based on the magnitude of the average amplitude in the effective fluctuation frequency band, the effective fluctuation frequency bands are prioritized, and the effective fluctuation frequency band with the largest average amplitude is defined as the dominant fluctuation frequency band of the wall. The center point of the frequency range covered by the dominant wave frequency band is taken as the characteristic frequency, and the average value of the amplitude of the frequency points within the dominant wave frequency band is taken as the characteristic amplitude. The aggregate wave intensity value of the wall is generated based on the characteristic frequency and the characteristic amplitude.

[0011] In a preferred embodiment, the formula for calculating the polymerization fluctuation intensity value is as follows: ; In the formula, This represents the intensity value of the polymerization fluctuation. Indicates the first The characteristic amplitude of the dominant wave frequency band. Indicates the first The characteristic frequencies of the dominant wave bands. This indicates the total number of the dominant fluctuation frequency bands.

[0012] In a preferred embodiment, determining the polymerization fluctuation intensity value exceeding a preset strength threshold as an abnormal bonding data, and obtaining the bonding abnormality determination result of the wall, includes: The aggregation fluctuation intensity value is compared with a preset intensity threshold one by one. When the aggregation fluctuation intensity value exceeds the preset intensity threshold, the detection time period and location of the aggregation fluctuation intensity value are marked as abnormal to obtain the abnormal detection time period and abnormal location information of the wall. By combining the information from the normal detection period and the abnormal location, the adhesion abnormality determination result of the wall is obtained.

[0013] In a preferred embodiment, generating an adhesion anomaly identification report for the wall based on the adhesion anomaly determination result includes: The abnormal detection period and the polymerization fluctuation intensity value in the bonding anomaly determination result are used as the abnormal fluctuation data of the wall. Based on the stress fluctuation intensity spectrum, a key impact analysis is performed on the abnormal fluctuation data to obtain the main frequency components of the abnormal fluctuation data; The abnormal detection period, the abnormal fluctuation data, and the main frequency components are integrated into the structured description data of the wall. The structured description data is reorganized using a document template to obtain an adhesion anomaly identification report for the wall.

[0014] To address the above problems, the present invention also provides a system for identifying anomalies in adhesive testing data, the system comprising: The data acquisition module is used to apply force to the wall placed in the concrete column through the pressure rod, and to collect the bonding detection data of the wall in real time through the strain sensor located on the concrete column; The stress fluctuation intensity assessment module is used to determine the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data. The stress fluctuation intensity spectrum construction module is used to perform fluctuation feature spectroscopy on the stress fluctuation intensity based on the timestamp information in the stress fluctuation value, so as to construct the stress fluctuation intensity spectrum of the wall. The fluctuation intensity aggregation module is used to perform adaptive spectral domain feature merging on the fluctuation intensity values ​​in the fluctuation intensity spectrum to obtain the aggregated fluctuation intensity value of the wall. An anomaly determination module is used to determine the polymerization fluctuation intensity value exceeding a preset strength threshold as an anomaly in the bonding data, and to obtain the bonding anomaly determination result of the wall. The report generation module is used to generate an adhesion anomaly identification report for the wall based on the adhesion anomaly determination result.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention determines the stress fluctuation intensity through time window benchmark value analysis, then maps the time-domain signal to the frequency dimension using Fourier transform to construct a stress fluctuation intensity spectrum. It can also define effective fluctuation frequency bands and filter dominant frequency bands based on noise benchmarks, and calculate the aggregated fluctuation intensity value by combining characteristic frequencies and amplitudes. This process achieves multi-dimensional spectral domain feature mining, effectively filters invalid noise interference, accurately marks the anomaly detection period and location, significantly reduces the probability of missed or false positives for anomaly data, and ensures the reliability of the identification results.

[0016] 2. This invention constructs a fully automated processing system from data acquisition to report output. From acquiring analog electrical signals from strain sensors and interpreting them into detection data, to constructing wave intensity spectra, adaptive feature merging, and anomaly detection, and then automatically integrating abnormal data with frequency components to generate a structured identification report, the entire process requires minimal manual intervention. Its adaptive spectral domain merging mechanism simplifies the analysis process of multi-frequency band data, and the standardized report generation module eliminates the tedious steps of manual document preparation, significantly improving the timeliness of detection data processing and meeting the high-efficiency detection needs of engineering sites. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for identifying anomalies in adhesive detection data according to an embodiment of the present invention. Figure 2 This is a functional block diagram of an anomaly identification system for bonding detection data provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for identifying anomalies in adhesive detection data. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for identifying anomalies in adhesive detection data can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a method for identifying anomalies in adhesive testing data according to an embodiment of the present invention. In this embodiment, the method for identifying anomalies in adhesive testing data includes: S1. Apply force to the wall placed in the concrete column through a pressure rod, and collect the bonding detection data of the wall in real time through a strain sensor located on the concrete column; In this embodiment of the invention, the step of applying force to the wall placed in the concrete column through a pressure rod and collecting the bonding detection data of the wall in real time through a strain sensor located on the concrete column includes: A load is applied to the wall placed in the concrete column by a pressure rod, and a strain sensor located on the concrete column is activated simultaneously to obtain the simulated strain electrical signal of the wall. The simulated strain electrical signal is interpreted to obtain the adhesion detection data of the wall.

[0021] First, rigidly connect one end of the pressure rod to the external loading drive device. The other end is precisely aligned with the test area inside the concrete column via a suitable connecting component. Simultaneously, ensure that the force direction of the pressure rod is perpendicular to the bonding interface between the wall and the concrete column. Next, attach a resistance strain gauge type strain sensor to the pre-defined monitoring area of ​​the concrete column using a special adhesive. This area should be selected at a pre-defined distance from the wall bonding interface and on a smooth, crack-free surface. After attaching the sensor, construct a Wheatstone bridge circuit for the sensor. Connect a stable DC power supply to the circuit's power supply end, and connect the signal output end to the input interface of the signal acquisition device. Then, start the external loading drive device to control the pressure application. The rod gradually applies axial load to the wall at a uniform speed. At the same time as the loading drive device is started, the power supply circuit of the strain sensor and the working circuit of the signal acquisition device are simultaneously connected. When the wall undergoes a small deformation under the load, it will cause the monitoring part of the connected concrete column to undergo a coordinated deformation. The strain sensor attached to the concrete column will change its own resistance value due to the deformation. This change in resistance value will cause the output voltage of the Wheatstone bridge circuit to fluctuate accordingly. The signal acquisition device will capture this continuously changing voltage signal in real time. This voltage signal is the simulated strain electrical signal of the wall. This signal can directly reflect the deformation state of the wall under the load and the stress at the bonding part with the concrete column.

[0022] After acquiring the simulated strain electrical signal of the wall, the signal is first fed into a passive RC low-pass filter circuit for preliminary signal conditioning. This circuit consists of a resistor with a fixed resistance and a fixed capacitor. Utilizing the capacitive reactance of the capacitor to high-frequency signals, it filters out high-frequency interference signals generated by surrounding building electrical equipment and environmental electromagnetic radiation, retaining only the low-frequency effective signal related to the wall strain. After filtering, the conditioned simulated strain electrical signal is fed into an analog-to-digital converter module. The sample-and-hold circuit inside this module locks the instantaneous value of the filtered voltage signal at fixed time intervals. Then, the quantization circuit converts the locked analog voltage value into the corresponding discrete digital quantity, thereby realizing the conversion from analog strain electrical signal to digital signal. Afterward, the converted digital signal is fed into a feature recognition unit composed of an operational amplifier circuit and a comparator circuit. The operational amplifier circuit first calculates the number of different loading stages. The voltage amplitude change corresponding to the digital signal is compared with the voltage peak difference between the initial loading stage and the peak load stage. Then, a dedicated analog bandpass filter circuit is used to filter out the specific low-frequency signal corresponding to the deformation of the bonding interface in the digital signal. The energy metering circuit calculates the proportion of the energy of the low-frequency signal to the total energy of the overall signal. At the same time, the slope calculation circuit obtains the voltage change rate of adjacent sampling points in the time domain waveform of the digital signal. This completes the multi-dimensional feature extraction of the analog strain electrical signal. Finally, all the extracted signal features are connected to the feature matching unit. This unit has a pre-stored correlation mapping relationship between signal features and wall bonding performance. The comparison circuit compares the currently extracted features with the pre-stored mapping relationship one by one. Based on the comparison results, specific indicators such as the bonding strength grade and bonding interface integrity status of the wall are determined. These indicators together constitute the bonding test data of the wall.

[0023] To ensure the accuracy of the simulated strain electrical signal acquisition, after installing the pressure rod and strain sensor, the Wheatstone bridge circuit must first be zeroed. By adjusting the variable resistor in the circuit, the output voltage of the signal acquisition device in the no-load state is brought to zero. Then, the entire loading and signal acquisition system is run under no-load conditions to confirm the stability of the pressure rod's force application and the continuity of the strain sensor's signal output. Only then can the load application and signal acquisition process be formally started. Throughout the process, the operating speed of the loading drive device must be kept stable to avoid invalid fluctuations in the simulated strain electrical signal due to sudden load changes, thus ensuring the accuracy of subsequent signal feature interpretation.

[0024] During the signal feature interpretation stage, for the digital signal output by the analog-to-digital converter module, a constant operating voltage needs to be provided to the feature recognition unit through a voltage regulator circuit to prevent voltage fluctuations from affecting the accuracy of feature extraction. For the amplitude change data output by the operational amplifier circuit, it needs to be stored in real time through a holding circuit so that it can be quickly retrieved and compared with the pre-stored data during the feature matching stage. For the calculation of the energy ratio of low-frequency band signals, the energy metering circuit will calculate the voltage square integral value of the specific frequency band signal per unit time through integration operation, and compare it with the voltage square integral value of the overall signal per unit time to obtain the corresponding ratio data. For the calculation of the slope of the time domain waveform, the slope calculation circuit will obtain the voltage difference between adjacent sampling points through differential operation, and calculate the voltage change rate per unit time by combining the sampling time interval. These accurately extracted signal features will be synchronously transmitted to the feature matching unit. This unit will complete the corresponding conversion between features and bonding detection data through a logic judgment circuit, and finally output bonding detection data that can accurately reflect the bonding state between the wall and the concrete column.

[0025] The beneficial effects include enabling simultaneous load application and strain signal acquisition on the inner wall of concrete columns. By building a dedicated circuit for the strain sensor and performing preliminary zeroing and no-load test runs, the stability and accuracy of the simulated strain electrical signal acquisition of the wall are ensured, effectively capturing the electrical signals corresponding to wall deformation and stress at the bonded areas. A passive RC low-pass filter circuit filters high-frequency interference, and an analog-to-digital converter completes signal format conversion. Combined with various dedicated circuits, multi-dimensional feature extraction is performed, including signal amplitude changes, energy proportions in specific frequency bands, and time-domain waveform slope. This fully extracts effective information from the simulated strain electrical signals. Furthermore, based on pre-stored feature mapping relationships with bonding performance, feature matching is completed, accurately generating bonding test data such as wall bonding strength grade and bonding interface integrity status. Throughout the process, voltage stabilization circuits and holding circuits ensure operational stability at each stage.

[0026] S2. Determine the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data; In this embodiment of the invention, determining the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data includes: The stress level center value of the stress fluctuation value sequence in the bonding test data within a preset time window is used as the representative benchmark value of the corresponding time window; Based on the representative benchmark value, identify the peak and trough values ​​of the stress fluctuation value sequence within the preset time window; The stress fluctuation intensity of the wall is obtained by evaluating the quotient between the absolute amplitude difference between the peak value and the valley value and the representative benchmark value.

[0027] First, the stress fluctuation value sequence extracted from the bonding test data is input into the time-series partitioning unit. The timing module built into this unit divides the entire stress fluctuation value sequence into several preset time windows of equal duration according to a pre-set time span, without any overlap. Then, all stress fluctuation values ​​within each preset time window are synchronously transmitted to the averaging integration circuit. This circuit first stores all stress fluctuation values ​​within the window one by one through the signal temporary storage subunit, and then starts the accumulation subunit to continuously sum all the stored stress fluctuation values. After the summation operation is completed, the averaging processing subunit divides the summed total value by the number of values ​​within the time window. The total number of stress fluctuation values ​​is used to obtain the average level of stress fluctuation values ​​within the time window. This average level is the representative benchmark value for the corresponding time window. The entire process relies on hardware circuits to complete data calculations. Each time window will generate a unique corresponding representative benchmark value. At the same time, the averaging circuit will complete the accumulation and averaging operations of the data within each window according to a fixed calculation sequence. The signal storage subunit will trigger the working instruction of the accumulation subunit after receiving all stress fluctuation values ​​within the window. After the accumulation subunit completes the summation, it will send a calculation completion signal to the averaging processing subunit. The averaging processing subunit will then start the averaging operation after receiving the signal.

[0028] The representative benchmark value corresponding to each preset time window is transmitted to the benchmark cache submodule of the extreme value identification unit. Simultaneously, all stress fluctuation values ​​within that time window are imported into the numerical traversal submodule of that unit. The numerical traversal submodule sends each stress fluctuation value to the comparison subcircuit one by one according to the chronological order of the stress fluctuation values. The comparison subcircuit compares the input individual stress fluctuation value with the representative benchmark value stored in the benchmark cache submodule in real time. When a stress fluctuation value is detected to be higher than the representative benchmark value, it is temporarily stored in the extreme value candidate submodule. This comparison is then processed by the extreme value candidate submodule, which collects all stress fluctuation values ​​within the window. After the fluctuation values ​​are compared, the extreme value candidate submodule will select the largest value in the temporary data. The extreme value locking submodule will mark it and store it as the peak value in the corresponding time window. When a stress fluctuation value is detected to be lower than the representative benchmark value, it will also be temporarily stored in another set of extreme value candidate submodules. After all values ​​are compared, this submodule will select the smallest value in the temporary data. The extreme value locking submodule will then mark it and store it as the valley value in the corresponding time window. The determination results of the peak value and valley value will be bound and stored with the representative benchmark value of the corresponding time window through the association storage submodule.

[0029] First, the identified peak and trough values ​​within the same preset time window are transmitted to the difference calculation unit. The absolute value processing submodule of this unit performs a subtraction operation on the peak and trough values, and then the polarity judgment submodule identifies the sign of the subtraction result. If the result is negative, a reverse conversion operation is initiated, ultimately obtaining the absolute difference between the peak and trough values. Then, before starting the quotient evaluation circuit, the absolute difference is transmitted to the threshold judgment submodule and compared with a preset reasonable fluctuation threshold. If the absolute difference is within the threshold range, the subsequent quotient evaluation process begins. If it exceeds the threshold, the anomaly marking submodule is activated to annotate the data, and simultaneously, the window re-division operation of the time sequence division unit is triggered to re-divide and calculate the time window for the abnormal period. After completing the validity verification, the qualified values ​​are... The absolute amplitude difference and the representative reference value of the corresponding time window are synchronously connected to the quotient evaluation circuit. The numerical adaptation submodule of this circuit first performs signal amplitude calibration on the absolute amplitude difference and the representative reference value to ensure that they are at the same data level. Then, the division operation submodule is started to perform division operation with the absolute amplitude difference as the dividend and the representative reference value as the divisor. During the operation, the accuracy compensation submodule corrects the error of the operation result to eliminate the small deviations generated during the circuit transmission. The final quotient value is the stress fluctuation intensity of the wall in the corresponding time window. The stress fluctuation intensity of all preset time windows is transmitted to the data integration module. This module arranges the stress fluctuation intensity of each window in an orderly manner according to the time sequence to form a complete time series map of wall stress fluctuation intensity.

[0030] The beneficial effects include enabling the time-series division of stress fluctuation value sequences in bond testing data and the accurate calculation of representative benchmark values ​​through hardware circuitry. Each circuit module works collaboratively according to a fixed time sequence, ensuring the orderliness and uniqueness of benchmark value calculations. It can accurately identify the peak and trough values ​​of stress fluctuations within each preset time window based on numerical traversal comparison and extreme value candidate screening mechanisms, and can achieve associated storage of extreme values ​​and corresponding benchmark values, facilitating subsequent data traceability. The absolute amplitude difference between peak and trough values ​​can be validated first to eliminate invalid extreme value interference, and then quotient evaluation can be completed through amplitude calibration and accuracy compensation to obtain accurate stress fluctuation intensity. Simultaneously, the stress fluctuation intensity of each time window can be integrated into a time-series graph.

[0031] S3. Based on the timestamp information in the stress fluctuation value, the stress fluctuation intensity is subjected to fluctuation feature spectroscopy to construct the stress fluctuation intensity spectrum of the wall. In this embodiment of the invention, the step of performing wave characteristic spectralization on the stress fluctuation intensity based on the timestamp information in the stress fluctuation value to construct the stress fluctuation intensity spectrum of the wall includes: The stress fluctuation intensity and the timestamp information in the stress fluctuation value sequence are organized into the time-domain fluctuation signal of the wall. The time-domain fluctuation signal is mapped from the time dimension to the frequency dimension to obtain the strength information of the wall. Using the frequency in the frequency dimension as the horizontal axis and the intensity information as the vertical axis, a stress fluctuation intensity spectrum of the wall is constructed.

[0032] The step of mapping the time-domain fluctuation signal from the time dimension to the frequency dimension to obtain the strength information of the wall includes: The time-domain fluctuation signal is smoothed to reduce spectral leakage, resulting in a smoothed time-domain fluctuation signal for the wall. Perform a Fourier transform on the smoothed time-domain fluctuation signal to obtain the smoothed frequency-domain fluctuation signal of the wall. Multi-dimensional parameter coupling is performed on the frequency and amplitude components of the smooth frequency domain fluctuation signal to obtain the fluctuation amplitude of the smooth frequency domain fluctuation signal, and the fluctuation amplitude is used as the strength information of the wall.

[0033] First, all stress fluctuation intensity data and corresponding timestamp information collected during the wall inspection process are matched one by one. Specifically, each stress fluctuation intensity data is bound to the timestamp information of the data collection time. During the binding process, it must be ensured that each set of bound data corresponds to the stress state of the same monitoring location and the same collection stage of the wall. Then, according to the chronological order of the timestamp information, all the bound stress fluctuation intensity data are linearly arranged, and the continuity of the timestamps must be maintained during the arrangement. If there are abnormal timestamp intervals, the stress fluctuation intensity data within the corresponding time period must be supplemented by linear interpolation to ensure the integrity of the time dimension. Finally, a time-domain fluctuation signal that can reflect the change law of wall stress over time is formed. This time-domain fluctuation signal must completely contain all stress fluctuation intensity data and their corresponding timestamp information, and the time span of the signal must be consistent with the overall detection duration.

[0034] First, a smoothing transition process is performed on the acquired wall time-domain fluctuation signal. The core purpose of this process is to reduce spectral leakage that may occur during the subsequent frequency domain conversion. Specifically, the neighborhood coverage range of the signal data points is first defined. Taking each stress fluctuation intensity data point in the time-domain fluctuation signal as the core, a fixed number of adjacent data points are selected to form a neighborhood data group for that core data point. The selection of the neighborhood data group must ensure symmetrical coverage and not exceed the overall time boundary of the signal. Next, weights are assigned to all data points within the neighborhood data group. The weight allocation follows the rule that the core data point has the highest weight, gradually decreasing towards the adjacent neighboring data points. Then, the neighborhood... The stress fluctuation intensity value of each data point in the data group is weighted and fused with its corresponding weight. The fused data value is used as the new stress fluctuation intensity value of the core data point. The weighted fusion operation of all data points in the time-domain fluctuation signal is completed in this way. After completion, edge transition processing is performed on the first and last data points of the overall signal. The stress fluctuation intensity values ​​of the first and last data points are made to approach the virtual stable value outside the signal by linear weighting, so as to avoid data abrupt changes at both ends of the signal. Finally, a smooth time-domain fluctuation signal of the wall is obtained. The stress fluctuation intensity data points of this signal are all smoothed and the signal as a whole shows the characteristics of a smooth transition, which can effectively avoid the problem of spectrum leakage.

[0035] A Fourier transform operation is performed on the generated smooth time-domain fluctuation signal of the wall to obtain a smooth frequency-domain fluctuation signal. Specifically, the smooth time-domain fluctuation signal is first organized into a continuous stress fluctuation data sequence in chronological order. Then, this data sequence is periodically extended. During the extension process, the entire time span of the original signal is considered as one period, and the signal is extended forward and backward by the same number of periods. The number of extension periods must be sufficient to fully represent the signal fluctuation characteristics. Next, vibration component decomposition is performed on the extended complete data sequence. During the decomposition process, different vibration frequencies contained within the sequence are identified one by one. For each vibration frequency, the amplitude and phase information of the stress fluctuation at that frequency are extracted. The amplitude information reflects the strength of the stress fluctuation at the corresponding frequency, and the phase information reflects the initial state of the stress fluctuation at the corresponding frequency. Then, each vibration frequency is bound to its corresponding amplitude and phase information. After binding, redundant information introduced in the extension period is removed, and only the vibration component data corresponding to the original signal time span is retained. Finally, the relevant data of all vibration frequencies are integrated to form a smooth frequency domain fluctuation signal of the wall. This signal takes the vibration frequency as the core dimension, and each frequency corresponds to a unique amplitude component and phase component.

[0036] First, all frequency components and their corresponding amplitude components are extracted from the smooth frequency domain fluctuation signal of the wall. Then, a multi-dimensional parameter coupling operation is performed. Specifically, for each frequency component, its corresponding amplitude component and vibration duration characteristics are retrieved. Vibration duration characteristics are the proportion of time that the frequency component is stably present in the smooth frequency domain fluctuation signal. Next, the magnitude of the amplitude component is correlated and integrated with the vibration duration characteristics. During integration, the amplitude component is used as the basis, and the amplitude component is corrected according to the vibration duration characteristics. The higher the proportion of vibration duration characteristics, the closer the corrected value is to the original amplitude component. The lower the proportion of vibration duration characteristics, the more moderately the original amplitude component is attenuated. After the correction, a second adjustment is made based on the distribution proportion of the frequency component in the overall signal. The distribution proportion is the proportion of the energy of the frequency component to the total energy of the smooth frequency domain fluctuation signal. The higher the distribution proportion, the higher the adjusted value is, and the lower the distribution proportion, the unchanged value is. The value obtained after the above two integration and adjustment is the fluctuation amplitude corresponding to the frequency component. The parameter coupling of all frequency components is completed in this way to obtain the corresponding fluctuation amplitude. Then, the fluctuation amplitudes corresponding to all frequency components are summarized. The summed fluctuation amplitude set is the strength information of the wall.

[0037] First, determine the coordinate range of the stress fluctuation intensity spectrum. The horizontal axis is based on all frequency values ​​obtained after mapping the frequency dimension as described above. These values ​​should be arranged in ascending order of frequency, and the interval between adjacent frequency values ​​should be uniform to ensure the standardization of the horizontal axis scale. The vertical axis is based on the wall strength information corresponding to each frequency. The scale range of the vertical axis should be determined based on the maximum and minimum values ​​of the strength information to ensure that all strength information data is completely presented within the coordinate range. Then, match each frequency value with its corresponding strength information coordinate points. During matching, ensure that the horizontal axis of each coordinate point is within the specified range. For each corresponding frequency, the vertical axis represents the intensity information at that frequency. Then, following the frequency order on the horizontal axis, all matched coordinate points are connected sequentially, maintaining the smoothness of the lines during the connection process. For blank areas within frequency intervals, linear filling is required based on the intensity information of adjacent coordinate points to ensure the continuity of the spectrum. Finally, a wall stress fluctuation intensity spectrum is constructed with frequency as the horizontal axis and intensity information as the vertical axis. This intensity spectrum must be able to intuitively present the distribution of stress fluctuation intensity of the wall at different frequencies, and the scale markings and line directions of the spectrum must accurately correspond to the relationship between all frequencies and intensity information.

[0038] The beneficial effects include the ability to form a complete and continuous time-domain fluctuation signal of the wall by accurately matching and completing the data with stress fluctuation intensity and timestamp information, ensuring a comprehensive presentation of the wall stress variation over time; by segmenting the time-domain fluctuation signal, extending the period, and decomposing the vibration characteristics, an effective mapping from the time dimension to the frequency dimension is achieved, which can accurately obtain the strength information of the wall at different frequencies and fully explore the stress fluctuation characteristics of the wall in the frequency dimension; by standardizing the coordinate value range, accurately matching coordinate points, and smoothly connecting and filling, an intuitive and clear wall stress fluctuation intensity spectrum is constructed, which can clearly present the stress fluctuation intensity distribution of the wall at various frequencies, providing an accurate and visualized detection basis for the comprehensive assessment of the wall stress state. The overall process ensures the integrity of the detection data, the accuracy of feature extraction, and the intuitiveness of the result presentation, which can effectively support the analysis and judgment related to the safety of the wall structure.

[0039] S4. Perform adaptive spectral domain feature merging on the wave intensity values ​​in the wave intensity spectrum to obtain the aggregate wave intensity value of the wall. In this embodiment of the invention, the step of adaptively merging the wave intensity values ​​in the wave intensity spectrum to obtain the aggregated wave intensity value of the wall includes: Identify consecutive adjacent frequency points in the stress fluctuation intensity spectrum whose amplitude exceeds a preset noise benchmark; The effective fluctuation frequency band of the wall is obtained by defining the frequency band interval of the consecutive adjacent frequency points. Based on the magnitude of the average amplitude in the effective fluctuation frequency band, the effective fluctuation frequency bands are prioritized, and the effective fluctuation frequency band with the largest average amplitude is defined as the dominant fluctuation frequency band of the wall. The center point of the frequency range covered by the dominant wave frequency band is taken as the characteristic frequency, and the average value of the amplitude of the frequency points within the dominant wave frequency band is taken as the characteristic amplitude. The aggregate wave intensity value of the wall is generated based on the characteristic frequency and the characteristic amplitude.

[0040] The formula for calculating the intensity value of the aggregation fluctuation is as follows: ; In the formula, This represents the intensity value of the polymerization fluctuation. Indicates the first The characteristic amplitude of the dominant wave frequency band. Indicates the first The characteristic frequencies of the dominant wave bands. This indicates the total number of the dominant fluctuation frequency bands.

[0041] First, retrieve the average background noise amplitude obtained from previous blank tests on similar walls and set this average as the preset noise benchmark. Then, compare the amplitude of each frequency point in the wall stress fluctuation intensity spectrum with the preset noise benchmark. During the comparison, ensure that the amplitude data of each frequency point is verified against the preset noise benchmark one by one. Mark all frequency points whose amplitude values ​​exceed the preset noise benchmark. Next, determine the adjacency of the marked frequency points. Starting from the lowest frequency end of the intensity spectrum, check whether each marked frequency point and its adjacent frequency points are all marked compliant frequency points. Group compliant frequency points that are connected without gaps into the same group. Group and classify all compliant frequency points in the intensity spectrum to ensure that the frequency points in each group are continuously adjacent and that no non-compliant frequency points are mixed in. Finally, identify all continuously adjacent frequency points in the stress fluctuation intensity spectrum whose amplitude exceeds the preset noise benchmark. This process must completely cover the entire frequency range of the intensity spectrum without missing any frequency points.

[0042] For each group of identified consecutive adjacent frequency points, a frequency band interval is defined. Specifically, the lowest frequency value in each group of consecutive adjacent frequency points is extracted and set as the lower limit of the corresponding frequency band interval. Then, the highest frequency value in the same group of consecutive adjacent frequency points is extracted and set as the upper limit of the corresponding frequency band interval. Next, the integrity of the defined frequency band interval is checked. During the check, all frequency points within the interval are examined one by one to confirm that there are no frequency points whose amplitude does not reach the preset noise benchmark. At the same time, it is confirmed that the upper and lower limits of the interval can completely cover all consecutive adjacent frequency points in the group, and no compliant frequency points exceed the interval range. After the frequency band intervals of all groups are defined and checked, each checked frequency band interval is uniformly named the effective fluctuation frequency band of the wall. Each effective fluctuation frequency band corresponds to a set of independent consecutive adjacent compliant frequency points, and there is no frequency range overlap between the effective fluctuation frequency bands.

[0043] First, the average amplitude is calculated for the effective fluctuation frequency band of each wall. During the calculation, the amplitude values ​​corresponding to all frequency points within the effective fluctuation frequency band are first summarized. Then, the total amplitude value is divided by the total number of frequency points within the effective fluctuation frequency band to obtain the average amplitude of the effective fluctuation frequency band. Next, based on the average amplitude of each effective fluctuation frequency band, all effective fluctuation frequency bands are prioritized. During the sorting, effective fluctuation frequency bands with higher average amplitude values ​​are placed with higher priority, and effective fluctuation frequency bands with lower average amplitude values ​​are placed with lower priority. After sorting, the effective fluctuation frequency band with the largest average amplitude value is selected and formally defined as the dominant fluctuation frequency band of the wall. The dominant fluctuation frequency band must simultaneously meet the requirements that the average amplitude is the maximum value among all effective fluctuation frequency bands and that the frequency points within the frequency band are all consecutive adjacent compliant frequency points.

[0044] First, the lower and upper limits of the frequency range of the dominant fluctuation band of the wall are extracted. Then, the frequency values ​​corresponding to the lower and upper limits are integrated, and the midpoint of the two values ​​is taken as the center point of the frequency range covered by the dominant fluctuation band. This center point is directly determined as the characteristic frequency. Next, the amplitude values ​​of all frequency points in the dominant fluctuation band are summarized. Then, the total summation of the amplitude values ​​is divided by the total number of frequency points in the dominant fluctuation band to obtain the average amplitude of the frequency points in the dominant fluctuation band. This average value is directly determined as the characteristic amplitude. Both the characteristic frequency and the characteristic amplitude are core parameters of wall stress fluctuation generated based on the dominant fluctuation band, and both are directly related to the frequency range and amplitude distribution of the dominant fluctuation band.

[0045] First, the characteristic amplitude and characteristic frequency are correlated and integrated. During integration, the characteristic amplitude is used as the core basis, and the fluctuation influence range corresponding to the characteristic frequency is numerically corrected. The fluctuation influence range corresponding to the characteristic frequency is determined according to its frequency position in the stress fluctuation intensity spectrum. The fluctuation influence range corresponding to the characteristic frequency in the high-frequency region of the intensity spectrum is larger, and the characteristic amplitude needs to be appropriately increased. The fluctuation influence range corresponding to the characteristic frequency in the low-frequency region of the intensity spectrum is smaller, and the original value of the characteristic amplitude remains unchanged. Then, the fluctuation stability of the corrected characteristic amplitude and characteristic frequency is integrated a second time. The fluctuation stability is determined according to the dispersion of the amplitude of the frequency point within the dominant fluctuation frequency band. The lower the dispersion, the higher the fluctuation stability, and the integrated value is increased accordingly. The higher the dispersion, the lower the fluctuation stability, and the corrected characteristic amplitude value remains unchanged. The final value obtained after completing the two integration operations is the aggregate fluctuation intensity value of the wall. This aggregate fluctuation intensity value can comprehensively reflect the core characteristics of the stress fluctuation of the wall.

[0046] The corresponding characteristic amplitude is the average amplitude of the frequency points within the dominant fluctuation frequency band. The dominant fluctuation frequency band is the frequency band with the largest average amplitude selected from the effective fluctuation frequency bands of the wall. The effective fluctuation frequency band is obtained by defining the frequency band intervals of consecutive adjacent frequency points in the stress fluctuation intensity spectrum of the wall that exceed the preset noise benchmark. It is necessary to first identify these consecutive adjacent frequency points and define the corresponding frequency band intervals. After determining the effective fluctuation frequency band, the average amplitude of each frequency band is calculated. Then, the frequency band with the largest average amplitude is selected as the dominant fluctuation frequency band. Finally, the average amplitude of all frequency points within the dominant fluctuation frequency band is calculated, which is the characteristic amplitude.

[0047] The corresponding characteristic frequency is the center point of the frequency range covered by the dominant fluctuation frequency band. The frequency range of the dominant fluctuation frequency band is determined when defining the frequency band interval of consecutive adjacent frequency points in the wall stress fluctuation intensity spectrum whose amplitude exceeds the preset noise benchmark. It is necessary to first define the lower limit and upper limit of the dominant fluctuation frequency band, and then take the intermediate value of these two frequency values ​​to obtain the characteristic frequency corresponding to the dominant fluctuation frequency band.

[0048] The total number of corresponding dominant wave frequency bands is the number of dominant wave frequency bands defined in the wall. The dominant wave frequency band is the frequency band with the largest average amplitude selected from all effective wave frequency bands. It is necessary to first calculate the average amplitude and prioritize the effective wave frequency bands of the wall, and then count the number of those identified as dominant wave frequency bands, which is the total number.

[0049] This formula is used to generate the aggregate fluctuation intensity value of the wall. This value integrates the correlation results of the characteristic amplitude and characteristic frequency corresponding to each dominant fluctuation frequency band. Through this integration method, the comprehensive state of wall stress fluctuation can be quantitatively characterized, providing an intuitive core indicator for the assessment of wall structure stress state and safety judgment.

[0050] When the characteristic amplitude increases, its corresponding square value increases accordingly. Combined with the characteristic frequency value, the product of the two increases, thus increasing the final aggregate wave intensity value. When the characteristic frequency increases, even if the characteristic amplitude remains stable, the product of the two will also increase, similarly increasing the aggregate wave intensity value. If the characteristic amplitude or characteristic frequency of multiple dominant wave bands increases simultaneously, the increase in the aggregate wave intensity value will be more significant.

[0051] The beneficial effects include: accurately identifying meaningful consecutive adjacent frequency points in the stress fluctuation intensity spectrum through a preset noise benchmark, effectively eliminating invalid frequency data corresponding to background noise, and ensuring the reliability and relevance of subsequent analysis data; defining the frequency band intervals of consecutive adjacent frequency points, clearly delineating the effective fluctuation frequency band of the wall, defining a clear range for the subsequent extraction of core stress fluctuation features, and avoiding omissions or confusion of effective frequency intervals; prioritizing and determining the dominant fluctuation frequency band based on the average amplitude, accurately locking the core frequency interval of wall stress fluctuations, and highlighting... The key fluctuation characteristics are identified; by extracting the characteristic frequencies and amplitudes corresponding to the dominant fluctuation frequency bands, the core parameters of wall stress fluctuations can be condensed, providing a core basis for the quantitative assessment of the overall fluctuation state; by combining the characteristic frequencies and amplitudes to generate aggregated fluctuation intensity values, the comprehensive integration and quantitative characterization of wall stress fluctuation characteristics can be achieved, providing intuitive and accurate core indicators for the comprehensive assessment and safety judgment of wall structure stress state. The entire process is progressive, ensuring the accuracy, pertinence, and comprehensiveness of wall stress fluctuation analysis, and effectively improving the reliability of wall structure condition detection.

[0052] S5. The polymerization fluctuation intensity value exceeding the preset strength threshold is determined as an abnormal bonding data, and the bonding abnormality determination result of the wall is obtained; In this embodiment of the invention, determining the polymerization fluctuation intensity value exceeding a preset strength threshold as an abnormal bonding data, and obtaining the bonding abnormality determination result of the wall, includes: The aggregation fluctuation intensity value is compared with a preset intensity threshold one by one. When the aggregation fluctuation intensity value exceeds the preset intensity threshold, the detection time period and location of the aggregation fluctuation intensity value are marked as abnormal to obtain the abnormal detection time period and abnormal location information of the wall. By combining the information from the normal detection period and the abnormal location, the adhesion abnormality determination result of the wall is obtained.

[0053] First, retrieve the preset strength threshold. To determine this threshold, select multiple groups of normal walls with the same material, structure, and bonding process as the current wall. Conduct multiple stress fluctuation tests on these walls, obtain the polymerization fluctuation strength value corresponding to each test, and then count the maximum stable value among these values, that is, the highest value that appears repeatedly in multiple tests. This value is then used as the preset strength threshold.

[0054] Then, retrieve all current aggregate fluctuation intensity values ​​of the wall. Each aggregate fluctuation intensity value must be associated with a corresponding detection period, which is the continuous time interval corresponding to the value acquisition process, accurate to the start and end times of acquisition. Simultaneously, each aggregate fluctuation intensity value must also be associated with a corresponding detection location, which is a pre-defined monitoring point on the wall, clearly indicating its area affiliation. Next, verify each aggregate fluctuation intensity value against a preset intensity threshold one by one. During verification, the value of a single aggregate fluctuation intensity value must be directly compared with the threshold, and the comparison result must be recorded immediately after each comparison is completed.

[0055] When a value of aggregate fluctuation intensity exceeds a preset intensity threshold, the corresponding detection time period record and detection location record are retrieved directly. An "abnormal" label is added to the corresponding entry in the time period record, and the same "abnormal" label is added to the corresponding entry in the location record. When adding the label, it must be ensured that the label only corresponds to the value entry that exceeds the threshold and does not affect other entries. After completing the comparison and labeling of all aggregate fluctuation intensity values, all time period record entries and location record entries with the "abnormal" label are filtered out. The time period information and location information corresponding to these entries are extracted and organized into a structured information set, which is the abnormal detection time period and abnormal location information of the wall. Each entry in this information set contains both the abnormal detection time period and the corresponding abnormal location, and the information is accurate and without redundancy.

[0056] First, retrieve the compiled information on abnormal detection time periods and locations of the walls. Categorize all entries in this information set according to their detection locations, grouping all abnormal detection time period entries corresponding to the same detection location into the same group. During categorization, verify the location information of each entry one by one to ensure no misclassification. For each abnormal detection time period entry within each group, analyze the time intervals of each entry, checking for any continuous overlap between the time intervals corresponding to different entries. If the end and start times of two time intervals are connected or overlap, merge these two time intervals into a single continuous abnormal time interval. After merging, clearly label the overall start and end times of this interval. If there are time intervals between time intervals, maintain the independence of each time interval and do not merge them.

[0057] Next, a correlation description is provided for the detection locations and abnormal time periods corresponding to each group. The description must include the regional affiliation of the detection location, the specific time range of the corresponding abnormal time period, and the situation where the aggregation fluctuation intensity value at that location exceeds the preset intensity threshold within the corresponding time period. The description must be clear and unambiguous. After completing the correlation descriptions for all groups, the descriptions for all groups are arranged sequentially according to the regional distribution of the detection locations on the wall. During the arrangement, each group's description must be kept as a separate paragraph to avoid information mixing. Finally, the title "Wall Adhesion Anomaly Judgment Result" is added at the beginning of the arranged content, confirming the overall content as the final judgment result. This result fully presents the locations of adhesion anomalies on the wall and the corresponding time periods of occurrence, and can be directly used for assessing the wall adhesion status.

[0058] The beneficial effects include: determining the preset strength threshold by selecting normal wall data of the same material, structure, and process, ensuring the rationality and adaptability of the threshold, and providing a precise benchmark for subsequent comparisons; accurately locating each aggregate fluctuation strength value by binding it to the corresponding detection time period and location and comparing it with the threshold one by one, avoiding omissions or misjudgments of abnormal information, and directly completing the abnormal identification, ensuring the accuracy and correspondence of abnormal detection time period and abnormal location information; classifying abnormal information by detection location and merging consecutive time periods, making abnormal time periods at the same location more systematic and clear, and clearly presenting the correspondence between location and time period in the correlation description, avoiding information confusion; and finally, the wall bonding anomaly judgment results are arranged in an orderly manner by region, completely presenting the location of bonding anomalies and corresponding time periods, providing an intuitive and accurate basis for wall bonding status assessment, effectively improving the pertinence and reliability of wall bonding anomaly detection, and contributing to more accurate assessment of wall structure safety.

[0059] S6. Based on the bonding anomaly determination result, generate a bonding anomaly identification report for the wall.

[0060] In this embodiment of the invention, generating an adhesion anomaly identification report for the wall based on the adhesion anomaly determination result includes: The abnormal detection period and the polymerization fluctuation intensity value in the bonding anomaly determination result are used as the abnormal fluctuation data of the wall. Based on the stress fluctuation intensity spectrum, a key impact analysis is performed on the abnormal fluctuation data to obtain the main frequency components of the abnormal fluctuation data; The abnormal detection period, the abnormal fluctuation data, and the main frequency components are integrated into the structured description data of the wall. The structured description data is reorganized using a document template to obtain an adhesion anomaly identification report for the wall.

[0061] First, retrieve the wall adhesion anomaly judgment results. These results are obtained from the previously collected information on anomaly detection periods and locations. Extract the anomaly detection period for each entry from these results, ensuring each period includes a specific start and end time. Simultaneously, retrieve the corresponding aggregate fluctuation intensity value for each anomaly detection period. This value is generated earlier by combining characteristic frequency and characteristic amplitude. Bind each anomaly detection period to its corresponding aggregate fluctuation intensity value. During the binding process, verify the correlation between the records in the judgment results to ensure that each period corresponds only to the aggregate fluctuation intensity value generated during its detection process, preventing mismatches between period and intensity value. After completing all binding operations, organize the bound information, removing any duplicate entries. Name the organized information set "Wall Anomaly Fluctuation Data." Each entry in this data set contains a single anomaly detection period and its corresponding aggregate fluctuation intensity value, and the information in each entry accurately corresponds to the record in the judgment results.

[0062] First, retrieve the stress fluctuation intensity spectrum of the wall. This spectrum is constructed with frequency as the abscissa and intensity information as the ordinate. Then, retrieve the abnormal fluctuation data of the wall. Based on each abnormal detection period in the abnormal fluctuation data, locate the frequency and amplitude distribution area mapped to the corresponding time interval in the stress fluctuation intensity spectrum. Within this area, examine the amplitude of each frequency one by one, and identify the frequency components with relatively prominent amplitudes. These frequency components must be components whose amplitudes are significantly higher than those of other frequencies in the surrounding area. During the identification process, each frequency in this area must be compared one by one, without missing any frequency with prominent amplitude. Summarize these identified frequency components, excluding frequency components with lower amplitudes in the area, and retaining only the frequency components with prominent amplitudes. At the same time, associate and mark the summarized frequency components with the corresponding abnormal fluctuation data entries to ensure that each frequency component corresponds to its source abnormal fluctuation data. The associated and marked frequency components are determined as the main frequency components of the abnormal fluctuation data. These main frequency components must fully reflect the core characteristics of the corresponding abnormal fluctuation data in the frequency dimension.

[0063] First, retrieve the abnormal detection periods, abnormal fluctuation data, and main frequency components of the wall. The abnormal detection periods are derived from the bonding anomaly judgment results, the abnormal fluctuation data is a set obtained from the previous binding periods and strength values, and the main frequency components are based on stress fluctuation intensity spectrum analysis. Match each abnormal detection period with its corresponding abnormal fluctuation data entry, ensuring that each period corresponds to the correct strength value entry based on previously recorded correlations. Then, associate the matched combinations with the corresponding main frequency components, verifying the source markers of the main frequency components to ensure that each combination corresponds to its analyzed frequency components. Finally, each abnormal detection period, its corresponding abnormal fluctuation data, and its corresponding main frequency components form an independent information unit. Arrange all information units in chronological order of the abnormal detection periods, maintaining the integrity of each unit without splitting its content, while ensuring the overall logical coherence of the arranged information. The arranged information is then defined as the structured description data of the wall. Each unit within this data contains complete abnormal detection periods, abnormal fluctuation data, and main frequency components, which can be directly used for subsequent report reorganization.

[0064] First, prepare a document template for the wall adhesion anomaly identification report. This template should contain fixed content sections, including basic anomaly information, details of abnormal fluctuation data, analysis of main frequency components, and a comprehensive description of the anomaly. Each section should have clearly defined content fields, and each field should be labeled with its corresponding content type. Next, retrieve the structured description data of the wall and assign each information unit in the structured description data to a different section of the template. Specifically, the anomaly detection period corresponds to the time period field in the basic anomaly information section, the aggregated fluctuation intensity value in the abnormal fluctuation data corresponds to the numerical value field in the abnormal fluctuation data details section, and the main frequency components... The information units are assigned to the component entry areas of the main frequency component analysis section. When splitting and matching, the content of each information unit must be checked against the requirements of the template section to ensure that the content of each information unit is accurately filled in the corresponding section without any errors or omissions. After completing all the content, the content in the template is formatted to maintain consistency in the layout of each section. At the same time, the core content of each information unit is briefly summarized in the anomaly summary section to make the report content clearer and easier to understand. The formatted and supplemented template content is then used as the wall adhesion anomaly identification report, which fully presents the relevant data and analysis results of the wall adhesion anomaly.

[0065] The beneficial effects include: extracting abnormal detection periods and corresponding polymerization fluctuation intensity values ​​from the adhesion anomaly judgment results to form abnormal fluctuation data, accurately integrating core data related to wall adhesion anomalies, avoiding the dispersion and confusion of abnormal information, and ensuring the accuracy of subsequent analysis. By conducting key impact analysis on the abnormal fluctuation data based on stress fluctuation intensity spectrum to obtain the main frequency components, it is possible to deeply explore the core characteristics of abnormal fluctuations in the frequency dimension, providing targeted basis for investigating the root causes of adhesion anomalies. By integrating abnormal detection periods, abnormal fluctuation data, and main frequency components into structured descriptive data, abnormal information from various dimensions forms an orderly and interconnected whole, improving the organization and usability of the information. By reorganizing the structured descriptive data into document templates to obtain a wall adhesion anomaly identification report, the abnormal data and analysis results are presented in a standardized and clear form, facilitating quick access to core content and providing intuitive and complete reference for assessing the wall adhesion status and subsequent handling. The overall process makes the information processing of wall adhesion anomalies, from data integration to result output, more accurate, orderly, and practical.

[0066] like Figure 2 The diagram shown is a functional block diagram of an adhesion detection data anomaly identification system provided in an embodiment of the present invention.

[0067] The bonding test data anomaly identification system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the bonding test data anomaly identification system 100 may include a data acquisition module 101, a stress fluctuation intensity assessment module 102, a stress fluctuation intensity spectrum construction module 103, a fluctuation intensity aggregation module 104, an anomaly determination module 105, and a report generation module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0068] In this embodiment, the functions of each module / unit are as follows: The data acquisition module 101 is used to apply force to the wall placed in the concrete column through the pressure rod, and to collect the bonding detection data of the wall in real time through the strain sensor located on the concrete column; The stress fluctuation intensity assessment module 102 is used to determine the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data. The stress fluctuation intensity spectrum construction module 103 is used to perform fluctuation feature spectroscopy on the stress fluctuation intensity based on the timestamp information in the stress fluctuation value, so as to construct the stress fluctuation intensity spectrum of the wall. The fluctuation intensity aggregation module 104 is used to perform adaptive spectral domain feature merging on the fluctuation intensity values ​​in the fluctuation intensity spectrum to obtain the aggregated fluctuation intensity value of the wall. The anomaly determination module 105 is used to determine the polymerization fluctuation intensity value exceeding the preset strength threshold as an abnormal bonding data, and obtain the bonding anomaly determination result of the wall. The report generation module 106 is used to generate an adhesion anomaly identification report of the wall based on the adhesion anomaly determination result.

[0069] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0070] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0071] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0072] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0073] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for identifying anomalies in adhesive testing data, characterized in that, The method includes: S1. Apply force to the wall placed in the concrete column through a pressure rod, and collect the bonding detection data of the wall in real time through a strain sensor located on the concrete column; S2. Determine the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data; S3. Based on the timestamp information in the stress fluctuation value, the stress fluctuation intensity is subjected to fluctuation feature spectroscopy to construct the stress fluctuation intensity spectrum of the wall. S4. Perform adaptive spectral domain feature merging on the wave intensity values ​​in the wave intensity spectrum to obtain the aggregate wave intensity value of the wall. S5. The polymerization fluctuation intensity value exceeding the preset strength threshold is determined as an abnormal bonding data, and the bonding abnormality determination result of the wall is obtained; S6. Based on the bonding anomaly determination result, generate a bonding anomaly identification report for the wall.

2. The method for identifying anomalies in adhesive testing data as described in claim 1, characterized in that, The process involves applying force to the wall placed within the concrete column using a pressure rod, and collecting real-time bond detection data of the wall using a strain sensor located on the concrete column, including: A load is applied to the wall placed in the concrete column by a pressure rod, and a strain sensor located on the concrete column is activated simultaneously to obtain the simulated strain electrical signal of the wall. The simulated strain electrical signal is interpreted to obtain the adhesion detection data of the wall.

3. The method for identifying anomalies in adhesive testing data as described in claim 1, characterized in that, Determining the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data includes: The stress level center value of the stress fluctuation value sequence in the bonding test data within a preset time window is used as the representative benchmark value of the corresponding time window; Based on the representative benchmark value, identify the peak and trough values ​​of the stress fluctuation value sequence within the preset time window; The stress fluctuation intensity of the wall is obtained by evaluating the quotient between the absolute amplitude difference between the peak value and the valley value and the representative benchmark value.

4. The method for identifying anomalies in adhesive testing data as described in claim 1, characterized in that, The step of performing wave characteristic spectroscopy on the stress wave intensity based on the timestamp information in the stress wave value to construct the stress wave intensity spectrum of the wall includes: The stress fluctuation intensity and the timestamp information in the stress fluctuation value sequence are organized into the time-domain fluctuation signal of the wall. The time-domain fluctuation signal is mapped from the time dimension to the frequency dimension to obtain the strength information of the wall. Using the frequency in the frequency dimension as the horizontal axis and the intensity information as the vertical axis, a stress fluctuation intensity spectrum of the wall is constructed.

5. The method for identifying anomalies in adhesive testing data as described in claim 4, characterized in that, The step of mapping the time-domain fluctuation signal from the time dimension to the frequency dimension to obtain the strength information of the wall includes: The time-domain fluctuation signal is smoothed to reduce spectral leakage, resulting in a smoothed time-domain fluctuation signal for the wall. Perform a Fourier transform on the smoothed time-domain fluctuation signal to obtain the smoothed frequency-domain fluctuation signal of the wall. Multi-dimensional parameter coupling is performed on the frequency and amplitude components of the smooth frequency domain fluctuation signal to obtain the fluctuation amplitude of the smooth frequency domain fluctuation signal, and the fluctuation amplitude is used as the strength information of the wall.

6. The method for identifying anomalies in adhesive testing data as described in claim 1, characterized in that, The step of adaptively merging spectral domain features of the wave intensity values ​​in the wave intensity spectrum to obtain the aggregated wave intensity value of the wall includes: Identify consecutive adjacent frequency points in the stress fluctuation intensity spectrum whose amplitude exceeds a preset noise benchmark; The effective fluctuation frequency band of the wall is obtained by defining the frequency band interval of the consecutive adjacent frequency points. Based on the magnitude of the average amplitude in the effective fluctuation frequency band, the effective fluctuation frequency bands are prioritized, and the effective fluctuation frequency band with the largest average amplitude is defined as the dominant fluctuation frequency band of the wall. The center point of the frequency range covered by the dominant wave frequency band is taken as the characteristic frequency, and the average value of the amplitude of the frequency points within the dominant wave frequency band is taken as the characteristic amplitude. The aggregate wave intensity value of the wall is generated based on the characteristic frequency and the characteristic amplitude.

7. The method for identifying anomalies in adhesive testing data as described in claim 6, characterized in that, The formula for calculating the intensity value of the aggregation fluctuation is as follows: ； In the formula, This represents the intensity value of the polymerization fluctuation. Indicates the first The characteristic amplitude of the dominant wave frequency band. Indicates the first The characteristic frequencies of the dominant wave bands. This indicates the total number of the dominant fluctuation frequency bands.

8. The method for identifying anomalies in adhesive testing data as described in claim 1, characterized in that, The step of determining the polymerization fluctuation intensity value exceeding a preset strength threshold as an abnormal bonding data, and obtaining the bonding abnormality determination result of the wall, includes: The aggregation fluctuation intensity value is compared with a preset intensity threshold one by one. When the aggregation fluctuation intensity value exceeds the preset intensity threshold, the detection time period and location of the aggregation fluctuation intensity value are marked as abnormal to obtain the abnormal detection time period and abnormal location information of the wall. By combining the information from the normal detection period and the abnormal location, the adhesion abnormality determination result of the wall is obtained.

9. The method for identifying anomalies in adhesive testing data as described in claim 1, characterized in that, The step of generating an adhesion anomaly identification report for the wall based on the adhesion anomaly determination result includes: The abnormal detection period and the polymerization fluctuation intensity value in the bonding anomaly determination result are used as the abnormal fluctuation data of the wall. Based on the stress fluctuation intensity spectrum, a key impact analysis is performed on the abnormal fluctuation data to obtain the main frequency components of the abnormal fluctuation data; The abnormal detection period, the abnormal fluctuation data, and the main frequency components are integrated into the structured description data of the wall. The structured description data is reorganized using a document template to obtain an adhesion anomaly identification report for the wall.

10. A system for identifying anomalies in adhesive testing data, characterized in that, The system for implementing the method for identifying anomalies in adhesive detection data as described in claim 1 includes: The data acquisition module is used to apply force to the wall placed in the concrete column through the pressure rod, and to collect the bonding detection data of the wall in real time through the strain sensor located on the concrete column; The stress fluctuation intensity assessment module is used to determine the stress fluctuation intensity of the wall based on the stress fluctuation value in the bonding test data. The stress fluctuation intensity spectrum construction module is used to perform fluctuation feature spectroscopy on the stress fluctuation intensity based on the timestamp information in the stress fluctuation value, so as to construct the stress fluctuation intensity spectrum of the wall. The fluctuation intensity aggregation module is used to perform adaptive spectral domain feature merging on the fluctuation intensity values ​​in the fluctuation intensity spectrum to obtain the aggregated fluctuation intensity value of the wall. An anomaly determination module is used to determine the polymerization fluctuation intensity value exceeding a preset strength threshold as an anomaly in the bonding data, and to obtain the bonding anomaly determination result of the wall. The report generation module is used to generate an adhesion anomaly identification report for the wall based on the adhesion anomaly determination result.

Images